The Archaeology of Death in Roman Syria: Burial, Commemoration, and Empire 9781107131415, 9781316443231

Table of contents :
Cover
Frontmatter
Contents
List of Figures
List of Tables and Charts
Acknowledgments
List of Abbreviations
Introduction
1 - Locating The Dead: Space, Landscape, And Cemetery Organization
2 - The Tomb: Architecture And Decoration
3 - Gifts For The Dead: Function And Distribution Of Grave Goods
4 - The Dead: Bones, Portraits, And Epitaphs
5 - Funerary Beliefs: Differentiation, Continuity, And Change In Ritual
6 - The Global And The Local: Romanization, Globalization, And The Syrian Cemetery
Postscript
Appendix 1 - Sites
Appendix 2 - Tomb Types
List of Online Appendices
Bibliography
Index

Citation preview

embellished underground and above ground tombs filled the cemeteries of the Roman province of Syria. Its inhabitants used rituals of commemoration to express messages about their local identity, family, and social position, while book, Lidewijde de Jong investigates these customs and the belief systems that governed the choices made in the commemoration of men, women, and children. Providing the first all-inclusive overview of the archaeology of death in Roman Syria, the book combines spatial analysis of cemeteries with the study of funerary architecture, and decoration, and grave goods, as well as information about the deceased provided by sculptural, epigraphic, and osteological sources. It also sheds a new light on life and death in Syria and offers a novel way of understanding provincial culture in the Roman Empire. Lidewijde de Jong is Assistant Professor in Archaeology at the University of Groningen. She has extensive fieldwork experience in the Middle East and codirected projects in Syria. She was awarded a Visiting Scholar Fellowship at the Institute for the Study of the Ancient World (NYU). She has published widely on mortuary archaeology and Roman Mesopotamia. She serves on the Advisory Board of the American Journal of Archaeology and is the Chair of the Center for the Study of Culture, Religion and Society – Interdisciplinary Studies in the Ancient World (CRASIS) at the University of Groningen.

Printed in the United Kingdom

Cover image: Jørgen Christan Meyer

The Archaeology of Death in Roman Syria

simultaneously ensuring that the deceased was given proper burial rites. In this

de Jong

In the first centuries of the Common Era, an eclectic collection of plain and

The Archaeology of Death in Roman Syria Burial, Commemor ation, and Empire Lidewijde de Jong

THE ARCHAEOLOGY OF DEATH IN ROMAN SYRIA

In the first centuries of the Common Era, an eclectic collection of plain and embellished underground and aboveground tombs filled the cemeteries of the Roman province of Syria. Its inhabitants used rituals of commemoration to express messages about their local identity, family, and social position, while simultaneously ensuring that the deceased were given proper burial rites. In this book, Lidewijde de Jong investigates these customs and the belief systems that governed the choices made in the commemoration of Syrian men, women, and children. Presenting the first all-inclusive overview of the archaeology of death in Roman Syria, this book combines spatial analysis of cemeteries with the study of funerary architecture, decoration, and grave goods, as well as information about the deceased provided by sculptural, epigraphic, and osteological sources. It sheds a new light on life and death in Syria and offers a novel way of understanding provincial culture in the Roman Empire. Lidewijde de Jong is Associate Professor in Archaeology at the University of Groningen. She has extensive fieldwork experience in the Middle East and has co-directed projects in Syria. She was awarded a Visiting Scholar Fellowship at the Institute for the Study of the Ancient World (NYU). She has published widely on mortuary archaeology and Roman Mesopotamia. She serves on the Advisory Board of the American Journal of Archaeology and is the chair of the Center for the Study of Culture, Religion, and Society – Interdisciplinary Studies in the Ancient World (CRASIS) at the University of Groningen.

Tomb 86 (“Funerary temple”) in Palmyra. Photo by Jørgen Christian Meyer, www.hist.uib.no/ antikk/dias/Palmyra/Funerarytempel/Data/page.htm?2,0 – with permission

THE ARCHAEOLOGY OF DEATH IN ROMAN SYRIA BURIAL, COMMEMORATION, AND EMPIRE LIDEWIJDE DE JONG University of Groningen

University Printing House, Cambridge cb2 8bs, United Kingdom One Liberty Plaza, 20th Floor, New York, ny 10006, USA 477 Williamstown Road, Port Melbourne, vic 3207, Australia 4843/24, 2nd Floor, Ansari Road, Daryaganj, Delhi - 110002, India 79 Anson Road, #06-04/06, Singapore 079906 Cambridge University Press is part of the University of Cambridge. It furthers the University’s mission by disseminating knowledge in the pursuit of education, learning and research at the highest international levels of excellence. www.cambridge.org Information on this title: www.cambridge.org/9781107131415 doi: 10.1017/9781316443231  C Lidewijde de Jong 2017

This publication is in copyright. Subject to statutory exception and to the provisions of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press. First published 2017 Printed in the United Kingdom by TJ International Ltd. Padstow Cornwall A catalogue record for this publication is available from the British Library. isbn 978-1-107-13141-5 Hardback Cambridge University Press has no responsibility for the persistence or accuracy of URLs for external or third-party internet websites referred to in this publication, and does not guarantee that any content on such websites is, or will remain, accurate or appropriate.

CONTENTS

List of Figures List of Tables and Charts Acknowledgments List of Abbreviations INTRODUCTION

page vii xi xiii xv 1

1 LOCATING THE DEAD: SPACE, LANDSCAPE, AND CEMETERY ORGANIZATION

20

2 THE TOMB: ARCHITECTURE AND DECORATION

37

3 GIFTS FOR THE DEAD: FUNCTION AND DISTRIBUTION OF GRAVE GOODS

77

4 THE DEAD: BONES, PORTRAITS, AND EPITAPHS

102

5 FUNERARY BELIEFS: DIFFERENTIATION, CONTINUITY, AND CHANGE IN RITUAL

146

6 THE GLOBAL AND THE LOCAL: ROMANIZATION, GLOBALIZATION, AND THE SYRIAN CEMETERY

175

POSTSCRIPT

217

appendix 1. sites

225

appendix 2. tomb types

314

List of Online Appendices

335

Bibliography

339

Index

363

v

FIGURES

1

2 3 4 5

6 7

8 9

10

11

12

Front cover: Tomb 86 (“Funerary temple”) in Palmyra. Photo by Jørgen Christian Meyer, www.hist.uib.no/antikk/dias/Palmyra/ Funerarytempel/Data/page.htm?2,0 – with permission Tomb of Sampsigeramos, Homs. Engraving by Cassas (1756–1827). Permission granted by the Universitäts- und StadtBibliothek Köln Map of sites mentioned in the text. Drawn by author Tomb of Hamrath, Suweida. Drawing by M. de Vogüé and E. Duthoit (1865–1877). De Vogüé 1865–1877, plate I Al-Bass Cemetery in Tyre. Images by author Cemeteries of Palmyra, with detail of West Cemetery. Drawn by author, after Schnädelbach 2010, image p. 31; permission granted by the German Archaeological Institute (copyright remains with the German Archaeological Institute) Villages on the Limestone Plateau. Tchalenko 1953, plates XCIV, CXXXIII Shahba¯ . A: Butler 1903, photo p. 381; permission granted by the Howard Crosby Butler Archive, Department of Art and Archaeology, Princeton University. B: Drawn by author Pit- and cist-graves. A,B: Bounni & Saliby 1956, plate 5. C: Griesheimer 1997a, fig. 8. D,E: Zaqzûq 2001, plates 13.5, 13.6 Hypogea. A,B: Christensen et al. 1986, fig. 23c,f. C: Toll 1946, plate VIII. D: Ingholt 1938, Plate XLV, reprinted with permission from the Editor, Faculty of Arts and Sciences, American University of Beirut. E: Badawi 2007, fig. 17 Mausolea. A–C: Schmidt-Colinet 1992, abb 16a, o, u. D,F: Sartre-Fauriat 2001, figs. 41, 187. E: Butler 1914, ill. 244; permission granted by Howard Crosby Butler Archive, Department of Art and Archaeology, Princeton University. G: Balty 1981, fig. 187; permission granted by Mission archéologique belge à Apamée de Syrie (MABAS) Funerary enclosures (al-Bass Cemetery, Tyre). A: Complex 4 and 5. B: Complex 4 (left) and Complex 5 (right). Drawn/taken by author Tower-tombs. A: www.hist.uib.no/antikk/dias/Palmyra/ NekroW/Data/page.htm?15,0 (photographer: Jørgen Christian

2 3 4 22

23 26

27 42

43

44

45

vii

viii

LIST OF FIGURES

13

14

15 16

17

18 19

20

21 22 23

24

Meyer). B: www.hist.uib.no/antikk/dias/Palmyra/NekroW/ Data/page.htm?14,0 (photographer: Jørgen Christian Meyer). C: Toll 1946, plate LXV Jar-burials. A,B: Vandenabeele 1972, plates XVI, XX; permission granted by Mission archéologique belge à Apamée de Syrie (MABAS). C: Badawi 2007, fig. 8. D: Zaqzûq 2001, plate 13.3b Sarcophagi in the open air. A: Balty 1981, fig. 195; permission granted by Mission archéologique belge à Apamée de Syrie (MABAS). B: Butler 1916, ill. 351; permission granted by Howard Crosby Butler Archive, Department of Art and Archaeology, Princeton University Tumuli. Photo and drawings by W. Oenbrink Funerary stelae. A: Palmyra Museum, inv. A91. B: Photo by W. van Rengen; permission granted by Mission archéologique belge à Apamée de Syrie (MABAS). C: Balty 1981, fig. 213; permission granted by Mission archéologique belge à Apamée de Syrie (MABAS). D: Sartre-Fauriat 2001, fig. 360 Sarcophagi. A: Chéhab 1968, plate XLI; permission granted by the Ministry of Culture/Directorate General of Antiquities of Lebanon. B: Photo by L. Petersen. C: van Rengen 1972, plate XXVI, 1; permission granted by the Mission archéologique belge à Apamée de Syrie (MABAS) Rock-reliefs (Qatura). Tchalenko 1953, plates LXII, 1, 3; CLXXV, 3 Pre-Roman graves. A: Curvers & Stuart 1997, fig. 7; permission granted by the Ministry of Culture/Directorate General of Antiquities of Lebanon; drawn by Curvers & Stuart. B: Thorpe 1998–1999, fig. 6; permission granted by the Ministry of Culture/Directorate General of Antiquities of Lebanon; drawn by Thorpe; site director: R. Thorpe. C–E: Novák 2000, abb. 590, 32, 586 Plaster and painted decoration. A: Ingholt 1938, plate L, 3; reprinted with permission from the Editor, Faculty of Arts and Sciences, American University of Beirut. B: Le Lasseur 1922a, plate III, A. C: Dunand 1965, plate II; permission granted by the Ministry of Culture/Directorate General of Antiquities of Lebanon Tomb of Hiram (Tyre region). Renan 1864–1874, plates XLVII, XLVIII Tomb at Amrit. Renan 1864–1874, plate XVII Finds from loculus 1a in the hypogeum at Deb’aal (glass vessels, golden earring, golden necklace, golden ring). Hajjar 1965, plate XVII, XVIII, XIX; permission granted by the Ministry of Culture/Directorate General of Antiquities of Lebanon Items of personal adornment. A,D: van Ess & Petersen 2003, fig. 21–22; permission granted by the Ministry of Culture/Directorate General of Antiquities of Lebanon. B,C: Chéhab 1986, plates I,

47

48

49 50

51

54 55

62

68 72 74

78

LIST OF FIGURES

25

26 27 28 29

30

31 32

33 34

35 36 37

38 39

40

VIII; permission granted by the Ministry of Culture/Directorate General of Antiquities of Lebanon. E,F: Seyrig 1952a, plate XXVI Vessels. A,B: Chéhab 1986, plates II, XLVI; permission granted by the Ministry of Culture/Directorate General of Antiquities of Lebanon. C: Christensen et al. 1986, fig. 34h. D: Sartre-Fauriat 2001, fig. 203. E: Vandenabeele 1972, plate XX; permission granted by Mission archéologique belge à Apamée de Syrie (MABAS) Terracotta lamps from Tomb C (Palmyra). Higuchi & Izumi 1994, fig. 71 Miscellaneous finds. A: Seyrig 1952a, plate XXII. B: Abdul-Hak 1954–1955, plate V Tomb C (T. 32), Palmyra. Higuchi & Izumi 1994, figs. 25, 34 Tomb of Abedrapsas (T. 60), Frikya. A,C: Butler 1903, fig. 104, image p. 280; permission granted by Howard Crosby Butler Archive, Department of Art and Archaeology, Princeton University. B: Griesheimer 1997a, fig. 33 Co-buried individuals. A: Amy & Seyrig 1936, plate XLVI. B: Chéhab 1984, plate XVIII; permission granted by the Ministry of Culture/Directorate General of Antiquities of Lebanon. C: Photo by L. Petersen. D: Higuchi & Izumi 1994, plate 45 Habbasi Tomb (T. 66), Hama. Christensen et al. 1986, figs. 24, 25, 28, 29, 31 Tomb of Nasrallat (T. 76), Palmyra. Ingholt 1935, Plate XLIII, XLIV, 2; reprinted with permission from the Editor, Faculty of Arts and Sciences, American University of Beirut Stele from Qamishly. Cumont 1933, plate XL Military stelae. A: Yale University Art Gallery, inv. 1938.5356. B,C: Photo by W. van Rengen; permission granted by Mission archéologique belge à Apamée de Syrie (MABAS) Tombstone of Vibius Apollinarius, Palmyra. IGLS XVIII/1, fig. p. 371 Loculus M 1–1 in Tomb C (T.32), Palmyra. Higuchi & Izumi 1994, fig. 37 Hypogeum at al-Awatin (T. 45). Dunand 1965, plate XXII; permission given by the Ministry of Culture/Directorate General of Antiquities of Lebanon Grave goods in Tomb C (T. 32), Palmyra. Higuchi & Izumi 1994, fig. 29 Al-Bass Cemetery (Tyre). A: Photo by author. B: Chéhab 1985, plate LXXXVI; permission granted by the Ministry of Culture/Directorate General of Antiquities of Lebanon Mythological scenes. A: Chéhab 1968, plate VIII; permission granted by the Ministry of Culture/Directorate General of Antiquities of Lebanon. B: Dunand 1965, plate V; permission given by the Ministry of Culture/Directorate General of Antiquities of Lebanon

ix

81

85 88 91 109

113

115 118

123 127

133 136 149

151 153

155

161

x

LIST OF FIGURES

41 42 43 44

45 46

47

48 49 50 51 52

53 54 55 56 57 58 59 60

61 62

Grave goods surrounding funerary portrait in Tomb C (T. 32), Palmyra. Higuchi & Izumi 1994, fig. 36 Tomb of Tiberius (T. 7), Beshindlaye (Limestone Plateau). De Vogüé 1865–1877, plate 92 Limestone bust of Menophilia daugther of Diodoros (Hama, T. 66) Christensen et al. 1986, fig. 28b Mausolea. A: Butler 1915, ill. 280; permission granted by the Howard Crosby Butler Archive, Department of Art and Archaeology, Princeton University. B: al-As’ad & Schmidt-Colinet 1995, fig. 57 Aboveground tombs on the Limestone Plateau. Tchalenko 1953, plates CLXXV, CXCI Elevated sarcophagus (Complex XVI), al-Bass Cemetery (Tyre). Chéhab 1984, plate LXIV; permission granted by the Ministry of Culture/Directorate General of Antiquities of Lebanon Tower-tombs in the West Cemetery of Palmyra. Photo by Jørgen Christian Meyer (www.hist.uib.no/antikk/Dias/Palmyra/ NekroW/Data/page.htm?1,0) Plan of Apamea. Drawn by author Plan of Baalbek. Drawn by author Plan of Beirut. Drawn by author Plan of Bosra. Drawn by author Deb’aal 1: plan of tomb. Hajjar 1965, fig. 1; permission granted by the Ministry of Culture/Directorate General of Antiquities of Lebanon Plan of Dura Europos. A: Drawn by author. B: Toll 1946, plates I, II Plan of Hama. Drawn by author Map of the Hauran. Drawn by author Plan of Homs. Drawn by author Plan of Jebleh. Drawn by author Map of the Limestone Plateau. Drawn by author Plan of Qatura. Tchalenko 1953, plate LIX Plan of Palmyra. Drawn by author, after Schnädelbach 2010, image p. 31; permission granted by the German Archaeological Institute (copyright remains with the German Archaeological Institute) Map of Tyre and surroundings. Drawn by author Plan of al-Bass Cemetery. Drawn by author

163 164 178

187 193

201

212 226 230 234 238

242 244 248 252 264 268 272 282

287 306 310

TABLES AND CHARTS

tables 1 2 3 4 5 6 7 8 9 10

Tombs per site Items of personal adornment Miscellaneous finds Finds in Hellenistic-Parthian tombs Distribution of inscriptions (site and language) from cat. 1 tombs Distribution of inscriptions (site and language) from cat. 2 tombs Distribution of grave goods, Palmyra Distribution of grave goods, al-Bass Cemetery, Tyre Tomb types per site (known types) Tomb types per site (unknown types)

16 82 90 96 105 106 292 312 314 325

charts 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

Distribution of tomb types, cat. 1 – known types Distribution of tomb types, cat. 1 – unknown types Distribution of tomb types, cat. 2 – known types Distribution of tomb types, cat. 2 – unknown types Distribution of tomb types – Hellenistic and Parthian periods Distribution of grave goods over main categories Vessels Distribution of tomb types, Apamea Distribution of grave goods, Apamea Distribution of tomb types, Baalbek Distribution of grave goods, Baalbek Distribution of tomb types, Beirut Distribution of grave goods, Beirut Distribution of tomb types, Bosra Distribution of grave goods, Deb’aal Distribution of tomb types, Dura Europos Distribution of grave goods, Dura Europos Distribution of tomb types, Hama Distribution of grave goods, Hama Distribution of tomb types, Hauran Distribution of grave goods, Hauran Distribution of tomb types, Homs Distribution of grave goods, Homs Distribution of tomb types, Jebleh

38 39 39 40 63 80 84 227 228 230 231 235 237 239 243 246 247 250 251 254 256 266 267 270 xi

xii

LIST OF TABLES AND CHARTS

25 26 27 28 29 30 31 32 33

Distribution of grave goods, Jebleh Distribution of tomb types, Limestone Plateau Distribution of tomb types, Palmyra Distribution of tomb types, Selenkahiye Distribution of grave goods, Selenkahiye Distribution of grave goods, Tell Kazel Distribution of tomb types, Tyre Distribution of grave goods Djel el-’Amed/al-Awatin Tombs, Tyre Coffin materials

270 274 290 303 303 305 307 308 326

ACKNOWLEDGMENTS

While writing this book, I benefited greatly from contacts with many people and institutions. The concept of this book was developed during a sabbatical leave provided by the University of North-Carolina in Chapel Hill and a Visiting Scholar Fellowship of the Institute for the Study of the Ancient World (ISAW) of New York University. I thank both for their generous support. Grants to visit Lebanon and Syria were provided by the Vice Chancellor for Research and Economic Development (University of North Carolina – Chapel Hill), the Department of Classics (University of North Carolina – Chapel Hill), and the Office of Research Services (Wilfrid Laurier University). As this book is in part based on my doctoral thesis, gratitude to those who supported my dissertation research is in place. In particular, I wish to thank the faculty and graduate students in Archaeology, Classics, and Anthropology at Stanford University, and the funding sources at Stanford University and the Andrew W. Mellon Foundation. I am grateful for the opportunity to discuss my work with colleagues at Wilfrid Laurier University, the University of North Carolina – Chapel Hill, and the University of Groningen, and with the fellows and faculty at ISAW. I am indebted to Jacco Dieleman, Maura Heyn, Vana Kalenderian, Onno van Nijf, Jennifer Trimble, and Cecil Wooten for their helpful reflections on particular themes of the book, and to the publishers and anonymous reviewers at Cambridge University Press for their valuable suggestions. A special thanks goes out to all the participants in the graduate seminar I taught at UNC in 2012 on the Archaeology of Death, and to John Esposito and Angelina Phebus, who assisted with data collection. I wish to thank the authors, antiquities services, museums, and libraries that offered permission to use imagery from their collections. Jean-Charles Balty, Ross Burns, Hans Curvers, Agnes Henning, Jørgen Christian Meyer, Werner Oenbrink, Lars Petersen, Kiyohide Saito, Andreas Schmidt-Colinet, Barbara Stuart, and Ruben Thorpe were so kind as to make their materials available. William Torres provided valuable assistance with the illustrations. Throughout the years, the Directorate General of Antiquities of Lebanon and the Directorate General of Antiquities and Museums of Syria have assisted me in various ways, for which I offer my gratitude. xiii

xiv

ACKNOWLEDGMENTS

My family deserves special attention. My parents, Klaas and Tilly, never questioned my, perhaps odd, career moves, and came to visit me in the trenches of post-civil war Beirut and on a tell in the Syrian steppe. I owe much, very much, to my husband, Willie, for his loving support throughout this project, and his unfailing enthusiasm for visiting yet another Roman tomb while on holiday. Our daughter, Roxie, was born as this book took shape, and she gave me a most excellent and joyful counter-balance to work life. One day, I hope to be able to take her to see the sites in Syria. When I started with this book in 2010, the Arab Spring erupted in Tunisia and Egypt, and stirred in Syria. By the time I was writing the pieces on the Hauran in southern Syria, the same region had become the focus of uprisings. As I dove into the analysis of Syria’s rural sites and urban centers, many of them featured in international news headlines. With intense sadness, I witnessed the downfall of this beautiful country, with its warm and hospitable people. This book, therefore, is dedicated to the people of Syria.

ABBREVIATIONS

AE CIL CIS ii

IGLS IV

IGLS V IGLS VI IGLS XIII/1 IGLS XIII/2

IGLS XV

IGLS XVIII/1 PAT RES II RES IV RES VII

L’Année épigraphique. 1888–. Königlich Preussische Akademie der Wissenschaften zu Berlin, Corpus inscriptionum latinarum (Berlin 1893–). Corpus Inscriptionum Semiticarum. Pars secunda: Inscriptiones Aramaicas continens. Tomus I, 1889, Académie des Inscriptions et Belles Lettres, Paris. Jalabert, L. & Mouterde, R. 1955, Inscriptions Grecques et Latines de la Syrie. Tome IV: Laodicée. Apamène. Paul Geuthner, Paris. Jalabert, L. & Mouterde, R. 1959, Inscriptions grecques et latines de la Syrie. Tome V: Émésène, Paul Geuthner, Paris. Rey-Coquais, J. 1967, Inscriptions grecques et latines de la Syrie. Tome VI: Baalbek et Beqa’, Paul Geuthner, Paris. Sartre, M. 1982, Inscriptions Grecques et Latines de la Syrie. Tome XIII/1: Bostra, Librairie Orientaliste Paul Geuthner, Paris. Sartre, M. 2011, Inscriptions Grecques et Latines de la Syrie. Tome XIII/2: Bostra (supplément) et la plaine de la Nuqra, Presses de l’IFPO, Beirut. Sartre-Fauriat, A. & Sartre, M. 2014, Inscriptions Grecques et Latines de la Syrie. Tome XV: Le Plateau du Trachȏn et ses bordures, Presses de l’IFPO, Beirut. Yon, J. 2012, Inscriptions Grecques et Latines de la Syrie. Tome XVII/1: Palmyre, Presses de l’IFPO, Beirut. Hillers, D.R. & Cussini, E. 1996, Palmyrene Aramaic Texts, Johns Hopkins University Press, Baltimore & London. Chabot, J.B. 1907–1914, Répertoire d’épigraphie sémitique. Tome II, Imprimerie nationale, Paris. Chabot, J.B. 1919, Répertoire d’épigraphie sémitique. Tome IV, Imprimerie nationale, Paris. Chabot, J.B. 1936, Répertoire d’épigraphie sémitique. Tome VII, Imprimerie nationale, Paris.

xv

INTRODUCTION

DEATH IN ROMAN SYRIA

A traveler on the way to the Roman city of Berytus, modern Beirut, would pass through the cemetery just before reaching the town. Tall, built structures, elevated on bedrock outcrops, the tombs of this cemetery of the 1st and 2nd c. CE could not escape the attention of any passer-by. Inside, marble sarcophagi imported from Turkey and locally produced limestone coffins held members of the prominent families of the city. Others were buried in pits or in stacked niches inside the walled enclosures of the tombs. Gifts of jewelry, clothing, glass vessels, and coins accompanied the deceased. In this way, the people of Roman Beirut followed the funerary rituals of their forefathers, who placed great importance on the adornment of the body and the need to protect and appease the dead. The shapes of the tombs surrounding Roman Beirut, however, were hardly reminiscent of earlier types. The simple and unmarked graves of the pre-Roman burial grounds had given way to a great variety of shapes, sizes, and modes of decoration. Tombs now rose prominently along the main roads. Brightly painted decoration adorned the walls and well-executed reliefs covered the expensive coffins. Beirutis were not the only ones to revamp the architecture of funerary commemoration. Several hundred kilometers to the northeast, in the rural Syrian Limestone Plateau, inhabitants of the town of Qatura carved stylized portraits in the cliff wall, accompanied by Greek epitaphs, dedicated in the 2nd c. CE to deceased fathers, wives, and husbands. 1

2

INTRODUCTION

1. Tomb of Sampsigeramos, Homs. Engraving by Cassas (1756–1827)

The practice of using prominent cliffs to cut out spaces for the dead had an earlier history in this region, although it was never employed on such an elaborate scale. New in the Roman period were the additions of dates, names, and faces to the grave-sites and an emphasis on individual identity and family groups. In Emesa, modern Homs, descendants perhaps of local royalty built a tomb that towered over the settlement and made use of a novel material: Roman brick work or opus reticulatum. The tomb itself is long gone, but we get a glimpse of what it may have looked like in an engraving made in the late 18th c. by Louis-François Cassas (Figure 1). As an archaeology student working on the Roman cemetery of Beirut and traveling through Lebanon and Syria, I was struck by the rich material culture of mortuary practices from the Roman period. Extensive Roman necropoleis extended outside the ancient cities of Palmyra and Tyre. Local and national museums seemed overflowing with sarcophagi and funerary stelae. I wondered why the material remains of mortuary customs were so visible in the Roman period. Was it a pure chance of archaeological recovery or was there something different, new about the way people were burying the dead? This book is a response to this fascination and has two objectives. The first is to reconstruct the mortuary customs of people in the Roman province of Syria, providing an analysis of the practices and beliefs around what happened after death and what constituted a proper burial. The second objective is to situate these mortuary customs in the wider political-cultural context of the time, tracing their chronological development and the patterns of continuity and change. Did incorporation into the Roman Empire somehow inspire the construction of the monumental tombs outside Beirut, or was this a trend that pre-dated

DEATH IN ROMAN SYRIA

2. Map of sites mentioned in the text

Roman control or which had nothing to do with the new imperial overlords? Such questions are rarely asked of Syrian funerary materials, in large part due to the deplorable state of the remains and of the publication record. This book, thus, addresses a methodological challenge and develops a framework to deal with complex and decontextualized funerary assemblages. It is also about mortuary archaeology, and the translation from material and epigraphic evidence to the reconstruction of funerary beliefs. A book about culture change in a Roman province is a book about Romanization, or changes in cultural practices stemming directly from incorporation into the Roman Empire. In more general – and perhaps more fashionable – terms, it is about local versus global identity. The two interconnected processes of globalization and localization provide the theoretical framework developed in this book in an attempt to understand change. As we will see, mortuary practices in general, and tombs from Roman Syria in particular, have a great deal to tell us about the impact of empire on local communities, and vice versa.

Rediscovering the Tombs of Syria The material remains of tombs and cemeteries in Roman Syria, roughly corresponding with modern Syria and Lebanon (Figure 2), are perhaps not obvious

3

4

INTRODUCTION

3. Tomb of Hamrath, Suweida. Drawing by M. de Vogüé and E. Duthoit (1865–1877)

candidates for the study of the interaction between Rome and its provincial subjects. This is not for lack of material, as I explain later, but rather is related to the state of the remains, the manner of publication, and disciplinary divides. Today, we know thousands of tombs from over 200 sites across Syria. We also have a plethora of sarcophagi, portrait busts, and epitaphs in Semitic languages, Greek, and Latin. Unfortunately, the often dismal state of publication, inaccessibility, and decontextualized discussion have turned many interested scholars away from the region. Travelers and adventurers of the 18th and early 19th c. were the first to introduce the monumental tombs of Roman Syria to the Western world. Descriptions and drawings of the enigmatic tower-tombs of Palmyra and the tall mausolea in the Hauran spurred on new generations of explorers (Figure 3). Initial research on funerary practices in the region was directed at texts. Epigraphists had deciphered Palmyrene inscriptions by the mid-18th c., for example – almost 150 years before the first excavation took place. Ernest Renan’s exploration of the Levantine coast in the 1860s heralded a phase of intensive architectural surveys. Between 1899 and 1909, Howard Butler and his team covered the Limestone Plateau in Northwest Syria and the Hauran in the south. These surveys yielded an abundance of information about tombs, many of which have long since vanished. The first excavation aimed at Roman remains started in Baalbek in 1899. The monumental sacral architecture of this site was the object of study, but smaller explorations by Theodor Wiegand in the 1910s also yielded information about cemeteries

DEATH IN ROMAN SYRIA

surrounding the Roman city. The Interbellum witnessed the greatest period of exploration and excavation of Roman remains in the Syrian region. Research continued at Palmyra and the Hauran, and new cemeteries were discovered in Dura Europos and nearby Baghuz, as well as in Apamea, Beirut, Hama, Homs, Halabiye, and the region of Tyre. The survey work of Georges Tchalenko, undertaken between 1935 and 1942, refined the understanding of the architecture and epigraphy of the Limestone Plateau, the hinterland of Antioch. The tireless efforts of Henri Seyrig, head of the Antiquities Service during the mandate era, disclosed numerous tombs, inscriptions, sarcophagi, and other funerary materials. The period before the Second World War also witnessed the birth of epigraphic corpora such as the Répertoire d’ épigraphie sémitique (RES), Corpus Inscriptionum Semiticarum (CIS), and Inscriptions Grecques et Latines de la Syrie (IGLS). After the war, the playing field and research questions had changed. Roman remains were no longer at the forefront of international excavations. The Antiquities Services of Syria and Lebanon now discovered the majority of the Roman tombs, often as part of rescue operations. In Syria, local and foreign teams continued excavating Apamea, Dura Europos, and Palmyra. The French Institute (IFPO) has conducted new projects in the Hauran and Bosra since 1974, but few have yielded Roman tombs. Georges Tate revisited Tchalenko’s work on the Limestone Plateau. The Syrian Antiquities Service recorded tombs at Jebleh, Palmyra, and the Hauran. Most archaeological attention in Syria, nevertheless, was directed at the earlier Bronze and Prehistoric ages. Archaeologists interested in these periods avoided Roman remains when possible, although small collections of Roman tombs were published from excavations at Tell Kazel (in the 1960s and 80s) and Selenkahiye (1960s–70s). Archaeology in Lebanon followed a different trajectory. Members of the Lebanese Antiquities Service uncovered a painted tomb at Deb’aal in 1961 and an extensive burial ground at Tyre in 1959–75. Incidental discoveries of tombs were reported at Beirut and Sidon. By the mid-1970s, however, civil war halted large-scale excavations in Lebanon.1 The history of research explains to some extent the problems with the study of funerary practices in Roman Syria and the minimal attention given to these materials by scholars of the Roman world and the Near East. Early projects primarily focused on architecture, sculpture, and epigraphy. Information about the chronological development of cemeteries and tombs, as can be gleaned from stratigraphic analysis, is limited at best. Archaeologists during the Interbellum usually discarded human remains without analysis, and never collected faunal or botanical remains. Limited land-measuring and recording technologies 1

For a description of the excavation history of the sites mentioned in the text, see Appendix 1. Introductions to the history of research are provided by Butcher 2003, de Jong 2007, and Sartre-Fauriat 2001.

5

6

INTRODUCTION

resulted in gaps in our topographical understanding of the sites, as well as a serious dearth in plans of tombs and cemeteries. The Lebanese Civil War hindered the full publication of important excavations such as the Roman cemetery at Tyre. Other discoveries were only published in interim reports, many of which remained buried deep in a few libraries across the world. Looting, destruction, and the removal of portable elements of funerary architecture for reuse elsewhere have decontextualized large portions of the archaeological record. Most material pertaining to Roman tombs was disclosed through excavation reports and epigraphic collections. Both present a restricted picture, either focused on a single site and isolated from contemporary cemeteries in Syria, or focused solely on the textual content,divorced from the tomb it once adorned.One can hardly blame the scholars, whose purpose it was to disseminate newly discovered archaeological materials or to provide collections of inscriptions, leaving little room for cross-provincial comparison. Yet, the site- or inscription-based focus tells us little about the development of cemeteries in relation to larger themes of the Roman Empire, such as economic and social change, or their connection to long-term Near Eastern traditions concerning death and burial. This last point is particularly problematic. Disciplinary divides between scholars of the Near East and those interested in Greece and Rome have stood in the way of comprehensive studies of the mortuary material. Near Eastern scholars usually turn their backs on post-Alexandrian periods, whereas Greek and Roman archaeologists and historians primarily discuss Syria in terms of cultural influence from these western territories. Semitic inscriptions are largely published separate from their Greek and Latin counterparts. Comparative and diachronic research is often lacking. The Achaemenid centuries (6th–4th c. BCE), for instance, are usually not part of the training of students of the Roman or Greek world, but, as this book demonstrates, they had a profound impact on the cultural developments of later Syria. In the last 30 years or so, the material remains of the Roman period have gained new,although modest,interest.Millar’s history of the region,followed by Butcher’s archaeological account, provided new insights. Konrad and Butcher published short overview articles or book sections on funerary practices. Investigations of Roman tombs by German and French scholars were conducted on the Limestone Plateau and in the Hauran in the 1990s. Based on these projects, Sartre-Fauriat published the most extensive report on funerary practices of the region thus far, concentrating on the Roman Hauran. Griesheimer produced a smaller study of the Limestone Plateau, and Schmidt-Colinet, although focusing only Palmyra, adopted a comparative approach and incorporated themes concerning the larger Roman world in his study of funerary architecture. Sadurska and Bounni’s report of 1994 was the first full contextual report of Palmyrene funerary sculpture, and Henning wrote an extensive study of the Palmyrene tower-tombs. Japanese projects in the Southeast Cemetery of

DEATH IN ROMAN SYRIA

Palmyra yielded the first full osteological report of this site.2 Of all the sites in the region,Palmyra has received the most scholarly attention,and this site leaves a heavy imprint on the analysis presented in this book. The end of the civil war in Lebanon heralded the start of large projects, most notably at Baalbek, Beirut, and Sidon. These excavations have yielded a mass of data, of which analysis and publication have only just started. This book builds on the work of these scholars and extends their analyses to cover the entire Roman province. Despite the preceding complaints about the publication record on tombs in Roman Syria, my book is first and foremost a celebration of the work of early travelers and pioneer surveyors, and of the tireless efforts of archaeologists belonging to foreign teams and, in particular, the Antiquities Services of Syria and Lebanon.

Setting the Stage: A Short History of Syria Later, I discuss the methodology used to handle the fragmented assemblage of funerary materials from Roman Syria. Here, I provide a general overview of the period and region under study in this book. When Roman soldiers entered Syria in the 1st c. BCE, they encountered a variety of landscapes. Hilly regions bisected by rivers and two mountain ranges, the Lebanon and antiLebanon, characterize the western portion of Syria. Large urban centers were situated here, most notable in the Orontes Valley (Apamea, Antioch) and along the coast (Beirut, Laodicea, and Tyre). Antioch, modern Antakya, was the primary city, and its hinterlands included the rocky landscape of the Limestone Plateau. South of Damascus stretched the pasturelands and fertile plains of the basaltic area. Bosra was the main town in this region, known as the Hauran. The Euphrates and Tigris rivers cut through the semi-arid steppe landscape of eastern Syria. This area was barely urbanized before the Roman period, and its local population included farmers and nomadic pastoralists. Earlier, in the 4th– 3rd c. BCE, colonies had been founded along these rivers, with Dura Europos the most well-known example. Desert landscape largely covers Central Syria. This was the home of nomadic pastoralists and long-distance caravans, which met at the oasis settlement of Palmyra. Roman armies also entered a fractured political landscape, a result of waning centralized power on the part of the Seleucid kings and the rise of local leaders. After Alexander the Great ended Persian-Achaemenid rule in the region in 333–332 BCE, most of Syria became part of Seleucid territory. This kingdom was named after its first king, one of Alexander the Great’s generals, Seleukos the Great. In 200 BCE, southern Syria and the coastal cities in Lebanon and 2

Butcher 2003; Griesheimer 1997a; Henning 2003, 2013; Konrad 2004, 2007; Millar 1993; Sadurska & Bounni 1994; Sartre-Fauriat 2001; Schmidt-Colinet 1997, 2004. Japanese research in Palmyra: Higuchi & Izumi 1994; Higuchi & Saito 2001; Saito 1995. See also de Jong 2010.

7

8

INTRODUCTION

Syria, up to the river Eleutheros near Tell Kazel, were added to the Seleucid kingdom. Afterward, the territory of the Seleucid Empire gradually diminished. Parthians from northeast Iran expanded their control westward, and took eastern Syria in the 120s BCE. The Nabataean kingdom in Jordan controlled sections of southern Syria by 160–150 BCE. Movements for local independence in the same century left a patchwork of small kingdoms and tetrarchies, or chiefdoms. By the 1st c. BCE, chiefs and kings possibly ruled over Chalcis in North Syria, Chalcis in Lebanon, the Homs region, Abila in Lebanon, Arca (Beqa’ Valley), and Osrhoene, with their capital at Edessa. Textual sources mention Ituraeans in the Beqa’ Valley and Mount Lebanon, and nomad chiefs on the right bank of the Euphrates in Central–East Syria. Although this is not a study of the Hellenistic and Parthian periods, the preRoman centuries are relevant in assessing the long-term continuity of Roman period funerary customs. The archaeology of both Parthian and Hellenistic Syria is, however, woefully understudied, and the narrative of these periods primarily rests on historical accounts. These inform us that Seleucid kings rekindled a phase of urbanization in Syria by founding numerous colonies and bringing about major shifts in population. Northwest Syria, in particular, was transformed by the foundation of Antioch, Apamea, Laodicea, and Seleuciaad-Pieria. Military colonies have been identified through excavation at Jebel Khalid and Dura Europos on the Euphrates. The impact of the foundation of new cities on the surrounding countryside has not yet received extensive study. Further east,in the hinterland of the Mesopotamian colonial capital at Seleuciaon-the-Tigris,archaeological surveys have revealed increases in settlement density and cultivation, as well as changes in local forms of water management.3 The Seleucid Empire brought with it a new lingua franca, Greek, and sociopolitical institutions originating from the Greek world. It should be noted that evidence for these institutions, such as the urban polis model, is limited and mostly comes from later (Roman) periods. The cultural influence of the Hellenistic world, or Hellenization, is a topic that we will return to in this book.4 The Parthian presence in East Syria (North Mesopotamia) remains somewhat of an enigma. The region does not appear to have been urbanized, but results from archaeological surveys demonstrate a densely filled countryside.5 Material from the few excavated sites and cemeteries points to a thriving period in the

3

4

5

Adams 1965. For the countryside of North Syria, see Casana 2007; Ur 2010; Wilkinson 2003; Wilkinson & Rayne 2010. The issue of Hellenization is further explored in de Jong 2007. For recent overviews of Hellenistic Syria, see Butcher 2003; Grainger 1990, 1991; Kosmin 2014; Kuhrt & Sherwin White 1987; Millar 1987; Sartre 2005; Sherwin-White & Kuhrt 1993. De Jong 2011; de Jong & Palermo in press; Wilkinson 1997–1998. Parthians in Northern Mesopotamia have, thus far, received little attention, but see Downey 1988; Hauser 2012; Millar 1998; Wiesehöfer 1998.

DEATH IN ROMAN SYRIA

1st c. BCE and CE. Dura Europos, for instance, witnessed a boom in public building programs. Excavators of Tell Sheikh Hamad, a small rural settlement in the Lower Khabur region, noted a considerable degree of affluence, evidenced by high-quality architecture and rich graves. Houses and tombs included foreign imports from the eastern Mediterranean and Iran.6 Even if the details of cultural and political influence are often obscure, it is clear that Romans encountered diverse socio-political and cultural landscapes. Initially, Rome appears to have exerted direct control only over limited portions of the provincial territory, and much of the fractured political situation was left intact.7 Rome in any case was in a state of civil war for much of the 1st c. BCE, and some of the fighting took place in and over Syrian territory. Parthians also set their sights on Syria, and the century was one of frequent battles and incursions. The rise of the first Roman emperor, Augustus, may have brought more peace in the area, although Romans were still pitted against Parthians in the 1st c. CE. In general, the 1st c. BCE and CE left few archeological traces in the Syrian province. This holds true for the cemeteries, and I will come back to this issue in the following chapters. One of the few areas where activity was visible was in temples. Several were rebuilt, and these may have functioned as regional rather than urban sanctuaries, for instance, at Si’ in the Hauran, Palmyra in the desert, and Baetocaece by the Syrian coast. The poor visibility of these centuries was likely related to the developments of the subsequent ones. Starting late in the 1st c. CE, the archaeological record suddenly explodes with material, and the following centuries produced one of the most archaeologically visible periods in the history of the region. The visible and decorated tombs that feature at the start of this book should be placed in this period. Most evidence comes from the cities of the Syrian province, which expanded and were revamped in the 1st and 2nd c. CE.8 This urban redevelopment centered on three aspects. First, the central thoroughfare in the town was formalized by the construction or repaving of the central street, and often by the addition of a colonnade. Second, new public buildings arose, most commonly baths, temples, theatres, and, less frequently, circuses and arches, constructed in stone on monumental scale. Aqueducts and other canals bringing water arose outside the towns. The third aspect of urban transformation was expansion. Most cities grew beyond their former size in the Roman period. The peak of building activity occurred between the late 1st and early 3rd c. CE. Parallel trends can be detected in the countryside. Two rural areas that had been fairly uninhabited 6

7

8

Dura Europos: Downey 2000, 155; Leriche 2002, 103. Sheikh Hamad: Kühne 2005; Novák 2000. This pattern of bloom and trade connections with the Roman world also emerges from the assemblages at sites in North Iraq (de Jong & Palermo in press). Overviews of the history and archaeology of Roman Syria are provided by Butcher 2003; Millar 1993; Sartre 2005. See also de Jong 2007. See also MacDonald 1986 and Segal 1997 and the description of sites in Appendix 1.

9

10

INTRODUCTION

in the centuries before the Roman era, the Hauran and the Limestone Plateau, now slowly filled with villages and agricultural installations.9 Most people in these regions appear to have lived in nucleated settlements, or in one of the larger centers, such as Brad, Cyrrhus, and Qatura in the Limestone Plateau and Bosra, Qanawat, Shahb¯a, Suweida in the Hauran. Syria had a long history of urbanization, and the rise in the number and size of cities accelerated in the Roman period. In this context, the term “city” refers to function and status, instead of (only) size. Cities had the right to mint coins and hold markets, and included a territory over which the urban institutions exerted power. This status was often bestowed on a settlement by Rome in the form of a title (“municipia,” “metropolis,” “colonia”). The process of awarding civic status continued throughout the Roman centuries, and towns such as Damascus, Dura Europos, and Palmyra received it only in the 3rd c. CE. A distinction needs to be made between the bestowal of civic rights and the founding of colonies, whereby the new administrative designation was accompanied by large groups of immigrants, the colonists. Only one such colony was founded in Roman Syria, at Beirut in the late 1st c. BCE. Baalbek in the Lebanese Beqa’ Valley perhaps lay in the territory of the Roman colony. High numbers of Latin inscriptions provide evidence for the presence of colonists in the vicinity of Baalbek. Inscriptions also indicate the scattering of veterans from the Roman army over the province. They mostly settled in rural areas. The majority of the “foreigners” in the province, nevertheless, were those on active duty. The location of Syria as a battleground of late Republican generals and local usurpers, and on the border with two powerful enemies (Parthians and Sassanians), ensured the continuous presence of large armed forces. Sartre estimates that four legions were stationed in Syria, in addition to numerous auxiliary units. These numbers fluctuated in the 2nd and 3rd c. CE.10 The soldiers were garrisoned in cities across the province and, by the late 1st c. CE, in military camps along the eastern border. This border was pushed further east after wars waged by Lucius Verus (161–166). Mesopotamia (eastern Syria and Iraq) transferred from Parthian to Roman hands, and Parthian sites such as Dura Europos and possibly Tell Sheikh Hamad received a garrison of Roman soldiers. At least from the 1st c. CE onwards, and possibly earlier, local Syrians were also drafted into the Roman army. Syrians reached the Roman senate in the second half of the 1st c. CE, and their number grew in the early 3rd c. CE under the influence of Emperor Septimius Severus’ Syrian wife and family.11 9

10 11

Hauran: Dentzer 1985, 1986; Dentzer et al. 2010; Graf 1992a, 450. Limestone Plateau: Strube 1996; Tate 1992; Tchalenko 1953, 1958. See also Callot & Marcillet-Jaubert 1984; Steinsapir 2005. Sartre 2005, 60–61. See de Jong & Palermo in press for a discussion of the militarization of eastern Syria. Syrians in the army and senate: Bowersock 1994, 141; Graf 1992b, 29; Smith 2013, 163–172.

SAMPLE BOUNDARIES

In the 3rd c. CE, three Syrians made it to the imperial throne: Elagabalus (218–222), Severus Alexander (222–235), and Philip the Arab (244–249). The latter directed many resources to the embellishment of his hometown, Shahb¯a in the Hauran, including a memorial and possible tomb for his father. The second half of the same century, however, was a troubled one for Syria. Sassanians, who earlier had wrestled control from the Parthians, started to apply pressure on the Roman border in the 230s. Twenty years later, they had pushed Roman control westward to Nisibis and Circesium on the Euphrates. Roman military sites at Dura Europos and the Lower Khabur region were abandoned. Sasanians also made inroads deeper into Syria, and briefly captured, for instance, the main cities, Apamea and Antioch, in 256 CE. Rome itself was in trouble in the 3rd c. CE, when emperors rose and fell in rapid succession. It is in the context of a loosened grip on its territories and a reliance on local leaders to counter some of its border attacks that we should read Queen Zenobia’s rebellion. This queen expanded the territory of Palmyra in the 260s, a deed read as rebellious by Rome. Their face-off lasted until 272 CE, when Emperor Aurelian crushed the Palmyrenes and took the city. Palmyra did not regain its former glory, and its cemeteries illustrate the punishment and downfall of local elite families. Construction activities and possible rises in population numbers indicate that Syria had regained its vitality, or some of it, by the 4th c. CE. Resources were now directed to other areas of social life, such as large residences with lavish opus sectile or mosaic floors or ecclesiastic architecture, as well as to non-urban areas. Smaller settlements such as Hama expanded in size, and the rural areas of the Limestone Plateau and the Hauran grew in population density and agricultural cultivation. The steppe of eastern Syria, perhaps for the first time, became involved in similar processes. Large military and economic centers developed at Nisibis and Raqqa. The border zone between Sura and Palmyra was refortified in the late 3rd and early 4th c. CE, and some of the camps, for instance, Resafa, evolved into larger centers. Habitation also moved into other arid zones, such as around Androna in Central Syria. By the mid 4th c. CE, the Late Roman or Byzantine period had started in Syria, the effective end point of my analysis. SAMPLE BOUNDARIES

This book investigates the funerary customs of communities in the Roman province of Syria (Figure 1). Later, I discuss what funerary customs entail, but here it is necessary to discuss the chronological and regional boundaries drawn for this analysis. The Roman province of Syria was formally created by general Pompey the Great in 64/63 BCE and covered an area roughly corresponding to modern Syria and Lebanon, as well as the small portion of Turkey on the eastern Levantine coast. Neither the boundaries of this province nor its political organization were stable and easy to trace. The Parthian Empire marked

11

12

INTRODUCTION

the eastern border of Roman Syria, but the extent of the Parthian territory remains obscure for the first 100 years or so. It would be a mistake in any case to see this border as a linear feature. It was a permeable gray zone, in which imperial control centered on the economic and military hubs. Rome had a considerable degree of influence in the affairs of Palmyra, but not in Dura Europos, which fell in the Parthian zone. Political control of the steppe and desert lands surrounding both towns was ambiguous, at least from our current perspective. By the late 1st c. CE, the eastern border, or limes, probably followed the Euphrates from Zeugma down to Sura and then south into the Central Syrian Desert. Whereas Dura Europos and the Balikh and Khabur valleys were in the Parthian realm of influence, this is less certain for Raqqa (ancient Nikephorion/Callinicum).With the conquest of Mesopotamia (161–166 CE), these territories fell into Roman hands. As we have already discussed, eastern Syria, with the rest of Mesopotamia, was then lost again to the Sasanians in the mid 3rd c. CE. In the same period, a string of Roman legionary and auxiliary fortresses arose in the Syrian Desert and along the Euphrates and Khabur rivers. Now, for the first time, a clearly demarcated border zone existed on the eastern edge of the Roman Empire.12 The southern limits of the province also moved back and forth. The region of the Hauran was partly incorporated into the Roman province of Syria in 64 BCE and partly stayed in the Nabataean kingdom. This kingdom, centered on Petra in southern Jordan, had expanded northward in the mid or late 2nd c. BCE. The northern Hauran, including the sites of Si’, Qanawat, and perhaps Suweida, was probably transferred to King Herod in 27 or 23 BCE, whereas the southern part, including Bosra, remained Nabataean. Throughout the 1st c. CE, control of the region alternated between the client-kings of the Herodian dynasty and Roman governors. By the late 1st c. CE, the Roman administrators incorporated the northern Hauran into the province of Syria. When the Romans in 106 CE annexed the Nabataean kingdom, the southern Hauran became part of the new province of Arabia, with Bosra as its primary center. Discussion exists about the northern boundaries of the province of Syria. The region of Cilicia and the town of Zeugma in modern Turkey are sometimes considered part of it.13 Sommer argues that the lower Balikh Valley became part of the new province of Mesopotamia in the 2nd c. CE, whereas the upper Balikh remained in the client-kingdom of Osrhoene. Under Septimius Severus (193–211 CE), the Roman province of Mesopotamia encompassed the entire Balikh Valley.14 This emperor also divided the province into Syria Coele in the north and Syria Phoenice in the south. Diocletian changed the 12

13

14

See the work of Konrad (2001, 2003, 2008) for the archaeological evidence of this eastern border. According to Butcher 2003, 108, eastern Cilicia was part of Syria until Vespasian. Edwell 2008, 15–17 suggests that Zeugma was the northern boundary of Syria until this same period. Sommer 2005, 69–71.

MORTUARY RITUALS IN PRACTICE

borders again at the end of the 3rd c. CE. Provincial boundaries thus fluctuated not only as Rome established control but also throughout the period of its domination. They were flexible and changeable. To make matters worse, even within the core area of the Roman province, the modes of political organization varied. One cannot equate the establishment of the province with direct rule by a Roman governor, at least not before the late 1st c. CE. As mentioned earlier, general Pompey and his successors followed the Roman policy of incorporating existing chiefdoms (tetrarchies) and small city-states, which, after showing allegiance to the Roman rulers, were left in charge of local governance. Textual sources demonstrate the continuation of several of these small political units into the 1st c. CE. Over the course of this century, the client-territories were annexed and placed under direct rule of a Roman administrator. This also happened with the two larger kingdoms south of the border: the Herodian client-kingdom in the Palestinian region (92 CE) and the Nabataean kingdom in the Jordanian area (106 CE). Parts of these territories were added to the province of Syria. If we consider the Roman province of Syria as a demarcated territory under direct control of a Roman governor, it is hard to pinpoint when this period began and where the boundaries lay. What we can say with certainty is that by 64 BCE, political leaders in the region were tied to Rome to a greater or lesser degree. This includes the Herodian and Nabataean territories in southern Syria and Palmyra. The Roman period in this book, therefore, starts in 64 BCE, or around the middle of the 1st c. BCE. The territorial boundaries are those of modern Syria and Lebanon. This allows me to include the steppe lands on the changing eastern frontier and parts of the rural south. It has the drawback of cutting across several cultural boundaries. Funerary sculpture at Zeugma, for instance – outside the territory discussed in this book – was similar to the assemblage of northern Syria. As we shall see, tombs in eastern Syria show greater affinity with Parthian northern Mesopotamia than with elsewhere in the province. The end point of the study coincides with the move of the capital of the Roman Empire to Constantinople in 330 CE, or the start of the Byzantine period. No tombs dating after 330 CE are incorporated in the sample. This is largely a mechanical division, although by that time, several cross-provincial trends had disappeared and the building of new tombs had slowed down. Furthermore, the official installation of Christianity in the early 4th c. CE, with its own ideas about death and the afterlife, eventually led to a new role for tombs in local communities. MORTUARY RITUALS IN PRACTICE

Funerary practices are the ways in which a society takes care of a dead body and deals with the loss of one of its members. They concern several interconnected

13

14

INTRODUCTION

aspects: the physical body, the memory of the deceased, and continuity within the surviving community. Taking care of the corpse is a prime concern, and the deceased body is often the focus of funerary ritual, during which it is physically, symbolically, and socially transformed from living to deceased.15 No longer a functioning member of society, the persona of the deceased acquires new roles, ranging from actively mediating spirits to inactive collectives of distant ancestors. Equally important is the symbolic separation of the deceased from the living. The surviving community, now a member short, has to restructure and reallocate the positions of its members, as well as find avenues through which to mitigate powerful emotions tied to bereavement. Funerary practices thus concern not only the deceased but also the relationship between individual humans, dead and alive, and between all humans and cosmological beliefs, or how people view their place in the world.16 Funerary beliefs involve a series of transformations: from living to dead, from house to grave, from the world of the living to the afterlife, and from life with to without the deceased. They constitute a particular class of ritual that concerns the transition from one social state to another, famously coined “rites of passage” by Van Gennep.17 This anthropologist demonstrated that these types of ritual – birth, initiation/coming of age, marriage, and death – share a common tripartite structure: separation, liminality, and reincorporation into a new status. The dead and the living proceed through these stages, which often involve distinct ritual actions, at various speeds and intensities. The reconstruction of mortuary rites, therefore, centers on identifying these transitions. The primary evidence for the funerary practices of Roman Syria comes from material remains: tombs, headstones, coffins, and artifacts that were placed with the deceased. Material culture is a means of communication, and plays a role in processes of social reproduction. Patterning in the material record is created by human (social) acts, which are informed by and situated in the norms and rules of a society. Actions and people stand in a discursive relation to each other, i.e., actions are defined by social norms, while at the same time, people define these norms through actions.18 In the material record pertaining to the funerary realm, people and societies reproduce the existing norms concerning the relation between the living, the dead, and the cosmological, and define these norms through mortuary behavior. Funerary practices involve ritual actions, which are distinct symbolic actions reproducing and reworking these relationships.19 By creating and acting out funerary practices, the living community reproduces socially agreed conventions, which are not only presented but also challenged and renegotiated. In a ritual practice, participants 15 17 18

16 Cf., Gramsch 2013; Williams 2013. Nilsson Stutz & Tarlow 2013, 4. Van Gennep 1960 [1908]; see also Hertz 1907. 19 Hodder 1986, 1995; Morris 1992, 1. Verhoeven 2011, 118. See also Tarlow 2013, 618.

MORTUARY RITUALS IN PRACTICE

reproduce – and not always consciously – socially agreed conventions.20 Key in funerary ritual is the shaping of memory, revolving around an individual deceased and/or of the collective persona of the dead. The dead are given place in the past, and memory of them can serve goals in the present, both in legitimization and in challenges of social order. Connerton writes about the link between practice and memory, or the bodily performance of commemorative practices as a way of shaping and reinforcing the identity of a group. Such practices differed from long-term cultural memory, which was inscribed in tombs and text. Through both means – inscribed and embodied memory – people gave shape to the present by drawing on the past.21 Not all dead held such roles, as some could be actively “forgotten.” They were denied a memory. This book analyzes the identity (or identities) of the deceased. Identity, in this context, concerns the broader groups with which individuals and communities identify.22 In the Syrian funerary context, status, gender, age, ethnicity, and professional identity were, to a greater or lesser extent, marked. A common pitfall in mortuary studies is the equation of identity in life with status in death. Despite Ucko’s early warnings based on ethnographic examples, archaeologists, for instance, still often link resources spent on a tomb directly to the wealth of the deceased or of his or her family.23 Low numbers of child burials are interpreted as resulting from an inferior status of children in society, and gendered grave goods such as weapons or weaving implements are used to establish the sex of the deceased. Funerary practices, however, do not portray the status of the living, but the agreed-upon norms around how to bury a person. These norms vary depending on aspects of a person’s social persona (i.e., gender or position), but are also related to other factors, such as how they died or the existence of egalitarian ideologies. Minimal burial differentiation between genders, for instance, does not mean that the living members of different gender groups did not hold distinct positions. Rather, the burying community chose to emphasize other aspects of the identity of the deceased, and not to elaborate or carry over gender divisions into death in a similar manner. The dead, in such a case, were less gendered than the living.24 In a similar way, greater elaboration of funerary practices does not solely or necessarily indicate the wealth of those building the tombs. Rather, there was something about the occupant that required enhancement. Morris has demonstrated how increased variation in the amount of resources spent on tombs was the result of changing norms about public displays of wealth.25 Furthermore, those with a particularly untimely or violent death, or those dying in the battlefield, could

20 21 22 25

Parker Pearson 2008 [1999], 194. See also Barrett 1990, 182; Nilsson Stutz & Tarlow 2013, 4. Connerton 1989, 2–4, 41–47; see also Joyce 2001 and Williams 2013. 23 24 Díaz-Andreu et al. 2005, 1. Ucko 1969. Cf., Sofaer 2006. See, for example, Morris 1992, 128–155.

15

16

INTRODUCTION

table 1. Tombs per site Site

Cat. Cat. 1 tombs 2 tombs Site

Apamea Baalbek Beirut Bosra Deb’aal Dura Europos Hama The Hauran (12 sites)

63 15 15 15 1 11 16 63

128 99 91 295 – 19 2 486

Cat. Cat. 1 tombs 2 tombs

Homs 23 Jebleh 8 Limestone Plateau (30 sites) 77 Palmyra 115 Selenkahiye 45 Tell Kazel 3 Tyre 47 Total 517

129 17 97 361 8 2 64 1797

stand out by a different and more elaborate form of burial.26 This is not to say that untimely death always resulted in burial differentiation, or that wealthy tombs never held the rich members of a community. Rather, we should understand mortuary practices to be operating on the nexus of cosmological beliefs, emotional and practical responses of the living, and the social persona of the deceased. METHODOLOGY

This book considers all material remains from mortuary practices (architecture, furniture, grave goods, and human remains), as well as the epigraphic evidence (funerary inscriptions). The historiographic section preceding this illustrated the difficulties with both material and epigraphic data from Roman Syria. Their publication record has severe gaps and inconsistencies, obstructing comprehensive synthesis. Since that is what this book aims to achieve, a discussion of the methodology is at place. Tombs are recorded from over 200 sites in Syria and Lebanon. In the analysis, I concentrate on assemblages from thirteen sites and two rural regions. These are described individually in Appendix 1. This selection is primarily based on the availability of contextual information, concerning the town or city associated with the necropolis. In the few cases where this data lacks, such as the Deb’aal tomb in Lebanon, it is the wealth of mortuary data that prompted its selection.The sample represents different types of settlement and a variety of regions (Table 1, Figure 1). It includes the only Roman colony in the province, Beirut, as well as one of its possible satellites, the sanctuary site of Baalbek. Larger cities and their territories include Apamea, Bosra, and Tyre. Smaller settlements in western Syria include Hama, Homs, Jebleh, and Tell Kazel. Three sites are chosen from Central–East Syria: Dura Europos and Selenkahiye in the Jezirah and the city of Palmyra in the desert. 26

Robb 2007, 292.

METHODOLOGY

I have also added to the sample two rural areas that have yielded a considerable number of tombs and inscriptions: the Limestone Plateau in Northwest Syria and the Hauran south of Damascus. Burial evidence from the Hauran comes from eight small villages and farmsteads and four larger towns. On the Limestone Plateau, twenty-eight small sites and two larger centers yielded useable funerary materials. Such a selection has a number of biases. First of all, the most important city of the region, Antioch, is omitted. This site has not yielded an extensive set of funerary materials, at least not in published form. Eastern Syria is underrepresented, with only two sites, again due to a dearth of usable data. The inclusion of Palmyra in the sample – a city that yielded a voluminous and relatively well-studied data set – may cause the analysis to tilt toward Palmyrene rather than Syrian practices. This is particularly the case in the discussion of funerary epigraphy and portraiture. When possible, the funerary material is compared to the tombs directly predating the period of Roman expansion. This includes Hellenistic material from western Syria (4th–2nd c. BCE) and Hellenistic and Parthian material from eastern Syria (4th c. BCE–1st c. CE). This material record is not well published either, and many details of these pre-Roman cemeteries remain unclear. In the analysis, pre-Roman tombs are included from Beirut, Dura Europos, Jebleh, Palmyra, and Tell Kazel. These are described in Appendix 1. Tombs from two more sites in Syria are included that yielded useable pre-Roman, but not later material. The Hellenistic colony or military foundation at Jebel Khalid on the Euphrates had a large cemetery, from which forty-eight graves were dated to the Hellenistic period (2nd c. BCE). At Tell Sheikh Hamad, a small rural site in the eastern steppe, a large cemetery extended in the lower town. A total of 314 tombs are published, of which sixty-five date to the Hellenistic-Parthian period (200 BCE–160 CE), seventy-five to the Parthian period (122 BCE–160 CE), and twenty to the Parthian-Roman period (122 BCE–220 CE).

Categories of Tombs At the time of press, thousands of tombs from over 200 sites are known from Syria, dating to between the 1st c. BCE and the mid 4th c. CE. As explained in the previous section, this material for the most part is inaccessible, decontextualized, and poorly published, leading to large gaps in chronological, architectural, and spatial information. Since my research depends on well-dated and contextualized tombs, I have set up parameters by which to select appropriate materials. The first criterion is a secure Roman date that can be verified based on published information. Several tombs carry a construction date in their foundation inscription, but we are not always so lucky. In other cases, nomenclature

17

18

INTRODUCTION

or references to military unit enables dating of the epitaphs. When inscriptional information is not available or useable, tombs are dated based on stylistic elements of the architecture and decoration, including funerary portraits and sarcophagi, or by association with tombs carrying a dated epitaph. In a few cases, grave goods, mostly glass vessels, are used to establish chronology. The dates of construction and the phases of usage for each tomb are included in the online appendix (www.cambridge.org/dejonglidewijde). The second criterion is the availability of information about the construction of the grave: whether we know anything about the shape of the tomb, the stele, or the coffin. This is based on published descriptions and, to a lesser extent, illustrations. Many publications merely report the presence of tombs or necropoleis, without any further elaboration. As such, they can hardly be used in the reconstruction of funerary ritual. The final selection criterion, applied less rigidly, concerns spatial location. Where was the tomb located and what was the spatial association with the settlement or other elements in the landscape, such as farming installations, roads, hills, and older cemeteries? After applying these parameters, 517 tombs from fifty-five sites remained. These are listed as category 1 tombs, and the analysis of this book rests upon this assemblage. Many of the tombs held multiple graves, and the minimum number of burial spots is around 4300. The thirteen sites and two rural regions yielded many more tombs and funerary materials that are not incorporated into the core sample. Vital information is lacking about these tombs. However, I did not want to discard this data entirely, as it holds clues about mortuary practices, even if only in numerical terms. This collection, therefore, is incorporated as category 2 tombs. These do not form part of the core sample, but I refer to them when they support, contradict, or add to a pattern established by the cat. 1 tombs. The cat. 2 sample consists of a minimum of 1797 tombs and around 15 000 burial spots. Table 1 illustrates the numerical distribution across sites. Tombs from both categories are described in detail in the online appendix. All available epigraphic, material, and topographical data concerning the tombs was collected. In the following chapters, I explore their location, architecture, tomb furniture, grave goods, skeletal remains, figural sculpture, and epigraphy. The limitations of each category are discussed in the individual chapters. Detailed information was usually available concerning architecture, inscriptions, coffins, and decoration, and cursory data existed on the layout of the burial grounds and the grave-good assemblages. Human remains were rarely published, and nothing is known about faunal or botanical data. The emphasis in this book thus gravitates to the spatial components of the funerary practices (i.e., the architecture and the placement of the tomb in the landscape).

METHODOLOGY

Set-Up The aim of this book is to provide a diachronic and cross-provincial comparison of the cemeteries. Such an approach requires a degree of abstraction. I do not provide a detailed iconographic study of the sculpture or funerary painting in the tombs, nor do I spend much time on architectural development or architectural decoration within each tomb type. The uneven publication of the tombs would in any case complicate such an analysis on a cross-provincial level. The epigraphic evidence is mined for dates and biographical information concerning the burying community, but is not discussed in linguistic or onomastic terms. There is much more to say and investigate with regards to the iconography, epigraphy, and contents of the graves. I leave this to others. This book should be read as a starting point for the discussion of the rich funerary material of Roman Syria and its relevance to themes current in the study of the Roman world. Chapters 1–4 compare tombs and cemeteries chronologically and across Syria, in order to establish provincial, regional, and local patterns, as well as changes over time. They examine spatial setting, funerary architecture, grave goods, and the deceased, respectively. Chapter 5 combines the results of this analysis into a reconstruction of the beliefs and ritual practices regarding burial and commemoration. Chapter 6 then takes these funerary customs to the political and cultural setting of Syria under Rome. The Postscript summarizes the results and reflects on current and future challenges. Two appendices offer a full description of the tombs and cemeteries and a brief discussion of the associated settlements (Appendix 1) and a longer description of the architectural features characterizing the distinct tomb types (Appendix 2). Online, the reader will find the database that forms the core of the analysis of this book, with descriptions of each tomb (www.cambridge.org/dejonglidewijde).

19

CHAPTER ONE

LOCATING THE DEAD: SPACE, LANDSCAPE, AND CEMETERY ORGANIZATION

W

here to place the tomb? this first chapter focuses on the choices made in the selection of a place for burial. Space is an important component of cemeteries, related both to the internal organization of a burial ground and to the surrounding landscape. Cemeteries and individual tombs were part of various, intersecting landscapes: the rolling hills and expansive steppe of the natural landscape, the houses and roads of the built landscape, the olive presses and irrigation canals of the productive landscape. This chapter also investigates the placement of burial grounds and individual tombs in relation to one another. How were cemeteries organized, and where were they placed vis-à-vis older burial grounds? What do these choices reveal about the relationship between communities and the landscape they inhabited?1 The emphasis in this chapter is thus placed on landscapes of the dead: cemeteries as landscapes and cemeteries in (built and natural) landscapes. Spatial links are key in the discussion of landscape and cemetery organization, yet the available evidence imposes limitations. The absence of spatial data, sometimes even in the most basic form of a plan of the cemetery, prohibits detailed geographical information system (GIS) examinations and statistical analyses of, for example, spatial clustering or visibility analysis. We concentrate on the available evidence, which varies in quality for each of the cemeteries discussed. The first section of the chapter addresses the location 1

20

Cf., von Hackwitz & Lindholm 2015, 146.

THE SETTLEMENT AND THE TOMB

of the cemeteries in relation to the built, natural, and past landscapes. The discussion centers on the study of the spatial connection between the cemetery and the town, as well as the zone around the settlement where agricultural installations, animal pens, quarries, roads, and features of the water supply were located. It then moves to questions concerning the visibility of tombs and the separation of settled and cemetery landscapes. The second section investigates the layout of the cemeteries, by focusing on spacing, orientation, clustering, spatial hierarchy, and intercutting of tombs. The conclusion highlights the aspects that characterize cemeteries across the province and can be considered part of cross-regional mortuary practices. THE SETTLEMENT AND THE TOMB

Cemeteries in Roman Syria extended along the edges of settled areas and were generally spatially associated with the town. This holds true for both urban cemeteries and those of villages.Only a few tombs were associated with isolated farmsteads. Larger sites often had more than one cemetery, ranging from two at Apamea to five at Palmyra and Bosra. It remains uncertain, however, whether one is able to distinguish between separate cemeteries in each case. As will become clear in the discussion that follows, physical boundaries were absent from the burial grounds, and the lines between cemeteries and non-funerary spaces were often blurred. The large sample of tombs surrounding the city of Tyre, ancient Tyrus, on the Lebanese coast, illustrates the main choices guiding the placement of burial grounds. Tombs covered most of the plain and foothills around the city, an area of ca. 14 km2 , and generally followed the main roads meandering from the city to the countryside. One cemetery started directly on the other side of a monumental arch marking the entrance to the city (Figure 4). This al-Bass Cemetery (1st–7th c. CE) flanked both sides of Tyre’s main paved route leading to the city. Tombs followed this road over a distance of 1.4 km inland. The alBass Cemetery presumably housed the urban population of Tyre, yet it remains unclear where the urban burial ground ended and those of villages in Tyrian territory started. None of the cemeteries appear demarcated in a physical form, by walls or natural features. In the case of Tyre, this may be related to the fact that archaeologists have not had the opportunity to conduct a full-scale spatial analysis and to map the smaller sites around the city. Yet, unclear demarcations between the cemeteries of cities and those associated with nearby villages may have been a common pattern across the Syrian province. Tombs, for instance, extended north of the city of Bosra in southern Syria as far as the hamlet of Jmarr¯ın, approximately 2.5 km away. It is uncertain whether clear boundaries existed between the two. If there were demarcations between the cemeteries of the city and nearby villages, they remain archaeologically invisible.

21

22

LOCATING THE DEAD: SPACE, LANDSCAPE, AND CEMETERY ORGANIZATION

A

B

C

D

4. Al-Bass Cemetery in Tyre. A: Plan of cemetery and surroundings. B: Entrance to Complex 33 with aqueduct pillar in the back. C: Roman arch. D: Roman–Byzantine road aligned with sarcophagi

When the al-Bass Cemetery of Tyre was constructed in the 1st c. CE, it had a spacious layout, with scattered tombs. By the 3rd c. CE, the open areas had made way for a densely filled space, where tombs sometimes literally piled up against and on top of one another.Rather than tombs being constructed further from the city as time progressed, burial grounds grew increasingly dense. The West Cemetery of Palmyra illustrates this point more clearly (Figure 5). Here,

THE SETTLEMENT AND THE TOMB

5. Cemeteries of Palmyra, with detail of West Cemetery

the oldest tombs of the late 1st c. BCE lay well over a kilometer from the city in the surrounding desert. Starting in the 1st c. CE, new tombs arose in the open spaces between the older tombs and the city, and gradually filled in the burial ground. Only after the 2nd c. CE were new cemeteries created further away from town, for instance, at Beirut, Bosra, Baalbek-Douris, and Hama.

23

24

LOCATING THE DEAD: SPACE, LANDSCAPE, AND CEMETERY ORGANIZATION

The crowded feel of the al-Bass Cemetery was not just related to infilling but also to the large-scale building projects of non-funerary nature that took place there. Between the 1st and 2nd c. CE, a circus, monumental arch, wide paved road, aqueduct, and perhaps an Apollo shrine arose in and around the cemetery (Figure 4). It is possible that gardens extended behind the tombs, as indicated by basins, canals, and an irrigation-system, although these may be of later date (5th–6th c. CE). In any case, rather than conceptualizing the spaces of the dead as quiet and removed from day-to-day life,the example of Tyre teaches us that Roman cemeteries were busy and full, with tombs, infrastructure, and public architecture competing for space. In the urban cemeteries elsewhere, we encounter theaters (Baalbek, Bosra, Qanawat), circuses (Beirut, Bosra), sanctuaries (Baalbek, Qanawat), and bath complexes (Brad and perhaps Dana). It would nevertheless be a mistake to consider the burial ground as part of the city proper. Tombs extended outside the settled area, and this fact was one of the main guiding principles of the location of cemeteries in Syria. In those settlements surrounded by fortifications, tombs arose outside the walled space, usually immediately outside the gates of the city. At Apamea in the Syrian Orontes Valley, for instance, if we assume that the currently visible (3rd c. CE) city wall followed the outline of an earlier rampart, the first tomb arose a few meters outside the North Gate (see Figure 48). In places where city walls have not been identified, there is nothing to suggest that tombs were built in the settlement. Furthermore, there are several instances of the decommissioning of older cemeteries when the growing city threatened to incorporate them. The Iron Age–Hellenistic cemetery of Beirut (Bey 018/063/040) was possibly still in use early in the Roman period, but was abandoned sometime between the late 1st c. BCE and the 2nd c. CE. Burial moved to new locations on the hills outside the city (see Appendix 1). Similarly, when the new town of Palmyra expanded, late in the 1st c. BCE or slightly after, the Hellenistic Baalshamin Tomb (2nd–1st c. BCE) was no longer used (Figure 5).2 An inscription dating to 11 CE describes the closing and possible purification ritual that followed the abandonment of the tomb (see p. 285). The separation of settled space and burial grounds is a common phenomenon in the ancient world, and we will come back to this topic later. At the same time, there are indications that this separation may not have been as strict and inflexible as was thought. At least four tower-tombs at the aforementioned West Cemetery of Palmyra lie within the walls surrounding the site (T. 1–4, see Concordance List). Perhaps this 2

See also the example of Shahb¯a (p. 260). The 2nd millennium BCE cemetery at Byblos on the Lebanese coast was reopened in the Hellenistic and early Roman periods, but abandoned when the city grew to incorporate this area after the 1st c. CE (Salles 1980, 9–19). A similar expansion in the mid 2nd c. CE or later is suggested for the Sheikh Abdallah Cemetery in Baalbek (see p. 231 and de Jong 2011 [2015]).

THE SETTLEMENT AND THE TOMB

was a chronological development, as the construction date of the tombs may post-date the use of the wall. In the town of Brad on the Limestone Plateau in northern Syria, a mausoleum stood between a domestic quarter with a bath complex and the rest of the town (T. 5; Figure 6). Shahb¯a, in the Syrian Hauran, gained new status in the 3rd c. CE as the birthplace of Emperor Philipp. A monument honoring his father, the Philippeion, occupied the center of the newly expanded town (T. 6). This building is often identified as a tomb, but we are, in fact, uncertain whether it housed the grave of Julius Marinus, Philipp’s father, or served as a commemorative building without holding his physical remains (Figure 7).3 These examples illustrate that exceptions were made and that tombs or buildings associated with funerary commemoration were constructed inside urban boundaries. Furthermore, due to expanding size, many cities did include in their walled space, the cemeteries of the recent past. In most cases, nevertheless, one would enter the cemetery upon leaving the city gates. As we have seen in Tyre, the burial grounds extended on both sides of the main paved road connecting the city to the countryside (Figure 4). This is another main guiding principle of Syrian cemeteries. Whereas the peninsula of Tyre left room for only a single main route, multiple roads radiated from the other cities.We find cemeteries along most of them.Not every burial ground or tomb flanked a road, but roadside location is a dominant characteristic among burial grounds in Roman Syria. Roadside cemeteries provided a spatial link between the deceased community and the – living – urban population. Upon approaching a settlement, one would first see its burial grounds. Again, it is clear that cemeteries were not spatially isolated or located settings far removed from daily life. Instead, they were very much part of the visible landscape, and were frequented by travelers, farmers, worshippers, and revelers of the circus and public bath. Other features in the built and natural landscape seem to have had little impact on the choice of location for Syrian burial fields. Instead, local topography and land use determined the selection of the place of burial. Some cemeteries,for instance,lay at a distance from agricultural fields and permanent water sources. The tombs of Palmyra and Dura Europos occupied marginal lands on the desert and steppe edge, and the cemeteries of Baalbek extended on all sides except its western edge, where the land was likely under cultivation. In the villages in rural Syria, by contrast, agricultural and hydraulic installations such as cisterns and olive presses surrounded the tombs. Basins, canals, and terracotta pipes in the cemetery at the al-Bass Cemetery of Tyre may indicate the presence of gardens close by, although, as already stated, these could be Byzantine 3

Other possible examples come from Suweida and inscriptions found at rural sites in the Hauran. They are discussed on p. 148.

25

26

LOCATING THE DEAD: SPACE, LANDSCAPE, AND CEMETERY ORGANIZATION

A

B

6. Villages on the Limestone Plateau. A: Bamuqqa in the 1st–2nd c. CE (left) and in the 6th c. CE (right). Tomb 1 is incidated as Tombeau. B: Brad, with Tomb 1 indicated as mausolée

THE SETTLEMENT AND THE TOMB

27

A

Citywall Cemetery

Temple?

Plaza T. 1a/1b Philippeion Theater

Cemetery Baths

Stadium

N

B

7. Shahb¯a. A: Philippeion. B: Plan of site

0

100

200

28

LOCATING THE DEAD: SPACE, LANDSCAPE, AND CEMETERY ORGANIZATION

rather than Roman in date.4 Quarries could serve as burial ground. Tombs were adapted to local natural features, and cut into the hillside, partly sunk into portions of the protruding bedrock, or dug out of the compact clayish soil. At times, natural features appear to have been used to enhance the height and visibility of tombs. The tall tower-tombs of Palmyra often were erected on hills or slopes, and prominent natural features on the Limestone Plateau, such as hilltops, rocky outcrops, and cliff façades often held tombs. Local conditions of terrain,therefore,directed much of the selection of burial grounds, yet cross-provincial patterns also emerge from the preceding discussion. The importance of extramural burial points to the existence of strong sentiments about the segregation of inhabited and funerary space. The movement of cemeteries as towns expanded further underscores the concern with maintaining this separation.At the same time,the cemeteries were in close proximity to the town, and surrounded by buildings of a civic and religious nature, or by agricultural and hydraulic features. Many burial grounds flanked the main roads leading to the settlement. The cemeteries formed part of the lived landscape. Despite the extramural location, a spatial link remained between the settlement and the grave field. Outside the urban boundaries proper, or on the outskirts of the inhabited space, appeared, alongside theaters, circuses, gates, and aqueducts, the necropoleis. In the following chapters, I argue that the proximity of burial and inhabited space was intentional, and served both to increase the visibility of the deceased and to stress their connection to the urban community. In the rural areas of Syria, the patterns around choice of location are less clear-cut (Figure 6). Tombs could be close to the settlement or several hundred meters or more away from it. Their placement alongside roads is less obvious, although often we lack of knowledge about the presence of roads in these regions. Several grave fields are not clearly associated with a settlement.5 A small number of tombs stood isolated in the rural landscape (T. 7–10). More research may identify the associated farms or hamlets, yet one cannot escape the general impression that the rules by which urban cemeteries were planned held less sway in the countryside. Visibility, on the other hand, was key in both rural and urban contexts, and often, the natural landscape was used to heighten the visibility of tombs. 4

5

De Jong 2010, 608–609, 627. Similar evidence for gardens comes from the Limestone Plateau in the 4th–6th c. CE (Griesheimer 1997a, 195–196). The Roman cemetery of Abila in northern Jordan was used for horticulture, pasture, and pottery production (Smith 1992, 222–224). See also Sartre-Fauriat 2001, vol. II, 20. In some cases, this is likely due to archaeological recovery, such as at Tell Kazel, Deb’aal, and Selenkahiye, where little is known about Roman-period habitation. In other cases, archaeological research has not identified nearby sites, such as in the grave fields of Nawa-tell Umm al-Hauran in the Hauran and Marina,Millis,Sardin,and Ir-Rubbeh on the Limestone Plateau. Both regions also yielded a small number of single tombs that were not associated with a settlement.

THE SETTLEMENT AND THE TOMB

A New Location? To what extent were these spatial preferences a continuation of older, preRoman practices? Overall, pre-Roman cemeteries remained in use in the Roman period, provided that they were not covered by the expanding city. At Hama and Tyre, the burial grounds were in close proximity to older ones (8th–4th c. BCE), although not those immediately preceding the Romanperiod constructions. It is not uncommon to find evidence for reuse of older tombs in the Roman period, as illustrated by Iron Age and Hellenistic tombs in Beirut,Jebleh,and perhaps Dura Europos.6 When new cemeteries were created, the primary reasons seem to have been the need for new space to accommodate the growing population and the encroachment of the inhabited area on existing burial grounds. The precise chronology of abandonment and formation of new cemeteries in Syria often remains hazy, yet several general patterns can be detected. First, when possible, the pre-Roman burial grounds remained in use, and new tombs were constructed in or close to the older cemeteries. The choice of location of tombs in the Roman period, therefore, followed older patterns of extramural burial and proximity to the settlement. Intramural burial did occur on the eastern edge of the study region, in Hellenistic and Parthian Mesopotamia. Its westernmost attestation, thus far, was at Dura Europos, where a group of twelve pit-graves was discovered on the citadel.7 Second, the construction of non-funerary architecture in the cemetery, and its roadside location, were mostly new features of the Roman period. Thus far, no buildings or paved roads have been identified in pre-Roman cemeteries, although the location of the main roads of Hellenistic settlements is not always clear. The exception is the burial ground of Dura Europos, which likely grew close to the main route of entry, and held a temple dedicated to Palmyrene gods. Third, when issues of space and expanding cities required the planning of new cemeteries in the Roman period, roadside location and proximity to other civic buildings became important guiding principles. In the concluding section of this chapter, I argue that both are related to the wider phenomenon 6

7

This common practice is also known from others sites in Roman Syria: Achaemenid-period (6th–4th c. BCE) tombs at Amrit were reused throughout the Hellenistic and early Roman years (Dunand et al. 1954–1955, 200–203; Saliby 1989, 26–27); several chambers of a rockcut tomb in Necropolis K in Byblos dating to the early 2nd millennium and 6th c. BCE yielded Hellenistic and early-Roman finds (3rd c. BCE–1st c. CE, Salles 1980); a Bronze Age tumulus in Der’a in the Hauran was reused for burial in the Roman period. See also the Achaemenid burial ground at Kamid el-Loz (Heinz et al. 2004) and Sarepta (Saidah 1969). For more examples of the longevity of Achaemenid-period tombs, see Nunn 2001, 397. The citadel was abandoned at the time (1st c. CE); see Matheson 1992, 125–127; Toll 1946, 6. Examples from Mesopotamia come from Hellenistic Nimrud (Oates & Oates 1958) and Hellenistic and Parthian Seleucia-on-the-Tigris (Graziosi 1968–1969; Hopkins 1972; Invernizzi 1967; Negro Ponzi 1970–1971, 1972, 2002; Valtz 1986, 1988; Waterman 1931; Yeivin 1933). In the Levant, the latest attestation was perhaps in the Achaemenid period (Nunn 2001, 393).

29

30

LOCATING THE DEAD: SPACE, LANDSCAPE, AND CEMETERY ORGANIZATION

of urban and extra-urban renewal that occurred between the late 1st c. BCE and the 2nd c. CE. As the layout of cities in Syria transformed in the Roman period, the cemeteries followed suit. Tombs constructed in the Roman period stand in stark contrast to their Hellenistic and Parthian predecessors in one area: visibility. There is little to suggest that pre-Roman cemeteries used rocky outcrops or other features of the landscape to emphasize the height and visibility of the tombs. In fact, the next chapter demonstrates that visibility was also a key distinguishing factor between Roman and pre-Roman funerary architecture. The importance of the visibility of funerary space will emerge as a major theme in this book. Throughout the Roman period, the choices relating to the locations of burial grounds appear to have changed little. In general, cemeteries were used for centuries, and maintained in the same location; although, as already noted, in some instances, the post-2nd c. CE cemeteries extended further from town. The few indications of the disturbance of cemeteries and the demolition of tombs tell an interesting story. From Dura Europos originates a stele with a Latin inscription, probably belonging to a Roman soldier, that was reused in the floor of the Artemis temple (T. 11). The date of this act is uncertain, but it must be placed between the presence of Roman soldiers in Dura Europos (116–121 CE or, more likely, after 165 CE) and the destruction of the town in 256–257 CE. At Apamea, perhaps as many as 130 funerary stelae commemorating Roman soldiers ended up in the city wall, which was rebuilt after an attack by the Sasanian Emperor Shapur in 256 CE. These stelae were dedicated in the first half of the same century, and, in some cases, only a few years before Shapur’s attack. In times of need, cemeteries were mined for building materials, and it was perhaps not a coincidence that the graves belonged to soldiers of the Roman army. This group had few personal ties in the town, and may even have prompted negative sentiments. After the soldiers had left, no one was present or willing to take care of their tombs. Perhaps the opposite occurred at Palmyra after the rebellion of Queen Zenobia and its reconquest by Emperor Aurelian in 272 CE. A new city wall now enclosed numerous tombs in the urban space, and many others became part of its construction (Figure 5). It is likely that this rampart was built around 300 CE by the Roman army, whose camp the wall partly encircled. Although the tombs were probably not in use anymore, this act of incorporation reads as a strong imperial Roman statement directed at the inhabitants of Palmyra. The tombs that were once built by the important families of this city and had been in use for more than 100 years were now rendered useless and served as building material for the Roman army. Although this book stops around 330 CE, there is little to suggest that the primary guidelines behind the selection of funerary space changed in the early Byzantine age, as most cemeteries were still in use (or were reused) in this period. There is one important exception. When Christianity was officially

CEMETERY ORGANIZATION

established as the religion of the empire, burials appeared in and around churches, which sometimes stood inside town. This practice is found in 5th c. Apamea, Qanawat, and several towns on the Limestone Plateau, but it may have already started a century earlier. The long-standing ban on intramural burial was now permanently lifted. CEMETERY ORGANIZATION

When we move from the level of the cemetery to that of the tombs, or more precisely, to the placement of the tombs in relation to one another, a different picture emerges. In contrast to similarities across the province concerning the placement of the cemetery, very few commonalities existed in cemetery organization. We can illustrate this point by comparing the two sites with extensively studied and published burial grounds. Those at Palmyra yielded at least 476 tombs. The West Cemetery, or Valley of the Tombs, represents the earliest, and contained a minimum of 141 tombs, dating between the late 1st c. BCE and the 3rd c. CE. It covered an area between 138 and 220 ha, within which tombs could be spaced tens of meters apart or more than 300 m from one another. As we have already seen, over time, the tombs grew closer to one another and to the city, thereby filling in some of the open space. Yet with 141 tombs in an area of over 100 ha, the cemetery remained spacious. In the mid–late 1st c. CE, additional cemeteries were created north, east, and south of Palmyra. Their size and density varied from 85 tombs in an area of 7.8 ha in the Southwest Cemetery to 80 ha with a minimum of 150 tombs in the North Cemetery. Clusters of tombs were aligned with geographical features such as hills and wadis (Figure 5). These tombs were built around the same time, and may have formed groups of similar types. The West Cemetery, for instance, had a cluster of tower-tombs on the hill of Umm Belqis, mausolea and towertombs along the Wadi Qubur, and a third cluster of hypogea and tower-tombs on the slope of Jebel Husayniye. Many tombs were aligned with roads leading to the city, and their entrances could either face the road or be found on the opposite side, facing away from it. A different picture is provided by the alBass Cemetery of Tyre. Between forty-nine and sixty tombs stood here, in an area of 1.5 ha or more. As discussed, in the 1st c. CE, tombs lay scattered in the field, but by the 3rd c. CE, the cemetery was densely filled and all tombs abutted another (Figure 4). Rows of funerary enclosures with doors opened toward the road and created the appearance of a street with tombs. By contrast, Palmyrene tombs never abutted, but were spread out over several kilometers in the hills and valleys encircling the city. Even when, over time, tombs grew closer to one another and to the city, they never formed “streets of tombs” as in Tyre. This variation in tomb density, cemetery organization, and size exemplified by the necropoleis of Palmyra and Tyre is a feature of the entire assemblage of

31

32

LOCATING THE DEAD: SPACE, LANDSCAPE, AND CEMETERY ORGANIZATION

Roman Syria. The absence of clear spatial information and maps complicates the analysis of the cemetery layout, but few cemeteries looked alike in this respect. We find crowded burial grounds crossed by paths and stairs (Beirut), pit-graves dug in a row (Baalbek-Douris), and fairly regularly spaced tombs at Jebleh and perhaps Baalbek’s Sheikh Abdallah Cemetery. Other sites had more spacious burial grounds, where the placement of tombs did not reveal strong patterns of clustering or spacing, such as Hama (Karm el-Haurani). One of the few aspects that the cemeteries seem to share is a low level of intercutting. When possible, older tombs remained untouched as new ones were dug. Since the burial grounds were not demarcated, and few have been completely excavated or surveyed, it is difficult to establish their size. Even the surface area of the cemeteries of the most complete sample, at Palmyra, is not always recorded or known. Clustering of tombs according to architectural type and geographical features, as described in the Palmyrene case, is not found elsewhere. Little spatial hierarchy or differentiation can be detected in the cemeteries. In other words, tombs were not singled out by a more prominent or isolated location,or by clustering.The only important type of clustering appears to have been alongside or close to the main roads. There was a strong preference for cardinal positions: 80% of all tombs for which the orientation was recorded followed a cardinal direction. This number increased during the Roman period, from around 60% in the 1st c. BCE and CE to 90% in the 3rd and 4th c. CE. Since many tombs were perpendicular or parallel to the roads, these patterns in orientation are related to the roads, which more or less followed cardinal directions. The Roman cemeteries in Syria, although placed at a similar location visà-vis the town and the road, did not follow the same internal plan. Instead, local decisions regarding issues of space, terrain, and type of tomb guided the planning of the burial fields. If it is assumed that the excavated portions are representative of the entire burial ground, then strong patterning in spatial hierarchy within the cemeteries was lacking. People were buried in similar locations. A handful of isolated tombs in the rural areas and the clusters in Palmyrene burial grounds form exceptions to this rule. Overall, the organization of the cemetery, or lack thereof, did not differ significantly from pre-Roman times. The Hellenistic cemeteries of Beirut and Tell Kazel were densely packed, as were the burial grounds of Jebel Khalid (Hellenistic) and Tell Sheikh Hamad (Hellenistic-Parthian). The tombs of Palmyra and Dura Europos could lie several hundred meters apart.8 No graves or groups of graves stand out in isolation or prominence. As in the Roman 8

The pre-Roman cemeteries of Beirut, Dura Europos, Palmyra, and Tell Kazel are described in Appendix 1. Jebel Khalid: Jackson & Littleton 2002; Littleton & Frohlich 2002; Littleton et al. 1996–1997. Tell Sheikh Hamad: Novák 2000.

DISCUSSION: MEMORIAL LANDSCAPES AND URBAN DEVELOPMENT

period, intercutting was uncommon, and a preference for cardinal positions can be detected. DISCUSSION: MEMORIAL LANDSCAPES AND URBAN DEVELOPMENT

What does the spatial setting of the tombs and the cemeteries in Roman Syria tell us about funerary rituals? First of all, the diversity in layout reveals the existence of locally specific guidelines about where to place the tomb. Here, we operate on the level of the community in the village, town, or city. The placement of the cemetery in relation to other landscape features, on the other hand, illustrates guidelines that were more or less shared across the province. Cemeteries were places of deep time and continued interaction. It was important to maintain older cemeteries and to avoid disturbing existing tombs. Infilling and reuse perhaps illustrate that proximity to older tombs was important. Burial grounds engaged with the long-term memory of the community, and linked them with the past. The proximity of these memorial landscapes to the settlement allowed for frequent interaction between the living and the dead. Far from being marginal places, the urban burial grounds were integrated into the sub-urban fabric. They were well trafficked and, to some extent, multipurpose spaces. Funerary space was not physically separated from civic, religious, or agricultural space. This does not mean that the local communities did not have strong sentiments about what constituted burial space and what did not. Yet these delineations were not physically inscribed in the landscape in a way that is visible to archaeologists. The delineations may not have been spatial at all, but, for instance, pertaining to time. There was a time for farming, enjoying horse races, visiting a bath, and a time for burying. In Syria, roadside location and proximity ensured a spatial link between the living (the settlement) and the dead (the cemetery). The fact that the dead were visible and close suggests an active role of the memory of the dead in the living community. If one were to focus on location alone, few distinctions are apparent among the deceased, whose tombs, with exceptions at Palmyra, were not clustered, isolated, or in other ways placed to stand out. In spatial terms, those who received an archaeologically visible burial are represented as a collective. This image, however, evaporates rapidly when we address the shape of the tomb in the next chapter. The dead were not buried in similar types of tombs. The discussion of the placement of cemeteries has also revealed strong sentiments about keeping the dead away from the inhabited space. Such separations are common in the ancient world, and are usually tied to concepts of sacred boundaries and pollution. In ancient Jewish, Roman, Mesopotamian, and Greek contexts, the dead were considered polluting and the living became ritually impure when coming in contact with the dead. A state of impurity prohibited or limited interaction with the sacred. Purification was thus necessary

33

34

LOCATING THE DEAD: SPACE, LANDSCAPE, AND CEMETERY ORGANIZATION

for those individuals who partook in funerary practices or interacted in another way with the dead. Cemeteries could not be located within the sacred space of a settlement. It was important to keep people away from the dead, and to keep the dead (and those dealing with the dead professionally) out of sacred locations.9 Lacking textual evidence, it is difficult to say whether similar concepts of pollution dictated the placement of tombs in Syria, although it may be likely given their predominance in the ancient Mediterranean and Near East. The most direct evidence comes from the Baalshamin Tomb in Palmyra, where an inscription appears to refer to an act of purification, probably when it was abandoned and the area was incorporated into the temenos of a new temple (see p. 169). The purification did not entail moving the tomb or its contents, but rather covering it. The fact that the cemeteries were well-trafficked spaces indicates the same point. The space of the cemetery itself was not polluting, and did not cause trouble to travelers, circus-goers, or worshipers on their way to an extramural sanctuary. In Chapter 5, I discuss how this fact may be related to the pollution being restricted to the first stages of burial. The ritual landscape of the burial ground simultaneously was kept separate and intersected with other landscapes, such as those of production, leisure, or worship. Archaeologists tend to look for clear lines or demarcations in the data set. In Roman Syria, we find blurred lines: between different cemeteries, between burial and non-burial space, between urban cemeteries and those of nearby villages. The only clear lines were formed by roads, radiating from the settlement and taking with them the tombs. These tomb-flanked roads represent a dominant characteristic of Roman-period burial grounds and form the starting point for discussing long-term trends in Syrian mortuary practices. What, if anything, changed in the Roman period? If possible, Syrians continued to bury the dead in the older, pre-Roman cemeteries. The choice in location of these burial grounds was not new, therefore, but tied to long-standing principles. It was only when cities and populations began to expand that new locations for burial were sought. Here, we can identify changing concepts of space. The new cemeteries occupied a more fixed place in the landscape compared to earlier periods. This pattern was most pronounced in the context of the cities of Roman Syria, whose cemeteries stretched out along the roads and close to the city walls and non-funerary architecture. Rural burial grounds continued to display greater variation in choice of location. 9

For instance: Avni et al. 2008; Kearns 2010; Lepetz & Van Andringa 2011; McCane 2003, 30; Parker 1996 [1983]. Lindsay 2000 has argued that these beliefs are arcane by the Imperial period in Rome itself, and now concern hygienic practices. See also Kaizer 2002, 185–187 for possible parallels in Palmyra and Lightfoot 2003, 510–512 on purity regulations in Lucian’s On the Syrian Goddess.

DISCUSSION: MEMORIAL LANDSCAPES AND URBAN DEVELOPMENT

The development of the new roadside cemeteries was gradual. The earliest examples come from the Hauran and Bosra, and perhaps date to the 1st c. BCE. In Palmyra, the oldest tombs along the wadi entrance west of the site were far from the road, and tombs started to line up with the road in the mid 1st c. CE or later (Figure 5). More robust evidence comes from the late 1st and the 2nd c. CE, exemplified by burial grounds at Apamea, Baalbek, Beirut, Bosra, Hama, Palmyra, Qanawat, Qatura, and Tyre. The encroachment of non-funerary architecture on the tombs was also clearly visible in the (early) 2nd c. CE. By this century, therefore, burial grounds had acquired a more standardized location, along the roads and surrounded by buildings and infrastructure. This chronological development followed closely what we know about settlement expansion and urban renewal throughout Roman Syria (see also Introduction). Beirut offers a good example of this phenomenon. Its pre-Roman town covered the tell (“acropolis”) and harbor area, and by the 2nd c. BCE expanded farther south. A city wall demarcated the boundaries of the Hellenistic city, and beyond the wall, the cemeteries extended, west, southwest, and east of the city. In the 1st c. CE, the city grew again, and showed the first signs of redevelopment and monumentalization of the urban landscape. The main roads were (re)paved, and a bath, aqueduct, and colonnade were erected. A century later, more baths and colonnades adorned the city. The discovery of a 2nd c. CE well cutting the Hellenistic city wall in the eastern section of town indicates that wall was out of use by that time.10 Threatened by the expanding city, the old cemeteries were abandoned, and new ones spread over the two hills east and west of the city center, loosely following the main roads out of the settlement. Other extramural architecture, most notably a circus, arose in close proximity to the tombs. As made clear in the introduction, other urban centers in Syria witnessed similar phases of urban renewal between the 1st c. BCE–1st c. CE and the 2nd c. CE. The transformation of larger settlements in Syria led to a degree of similarity, with each town boasting a similar package of public and often civic buildings, and comparable infrastructure and passage features. As part of this transformation, the cemeteries moved to a new and more fixed position. The standardization of the location of urban cemeteries can thus be linked to larger changes in the urban landscapes of the province. Despite a physical separation between the inhabited areas and the burial grounds, the latter were part of the (sub)urban landscape, in spatial, chronological, and – as illustrated in the next chapters – symbolic terms. This trend was less obvious in the smaller settlements of the province. The burial grounds of villages and hamlets in the Hauran and Limestone Plateau could be close to or at some distance from the habitation. Some aligned with 10

Curvers & Stuart 2005.

35

36

LOCATING THE DEAD: SPACE, LANDSCAPE, AND CEMETERY ORGANIZATION

the roads, but this is not obvious in each case. Furthermore, some tombs were seemingly isolated from a settlement. The fact that spatial associations with the town, roads, and non-funerary buildings was more pronounced in the larger towns in the same regions, such as Brad, Qanawat, Si’, and Suweida, suggests that the larger and more “urban” the town, the closer the physical links with the cemetery. Both in the rural areas and in the urban cemeteries of Roman Syria, emphasis was put on visibility. Proximity to the settlement, elevated placement, and roadside location increased the opportunity for tombs to be seen. The burial grounds visually displayed the community,or at least those members with access to burial in the settlement cemeteries. The next chapter demonstrates that visibility also became a major characteristic of funerary architecture: the shape of the tomb.

CHAPTER TWO

THE TOMB: ARCHITECTURE AND DECORATION

T

his chapter investigates a different type of space, that of the built tomb. What were the dominant guidelines in designing a final resting place? How were tombs built, organized, and decorated? This chapter is not concerned with in-depth descriptions of the architectural features of the tombs, an effort that would be thwarted in any case by large gaps in the available data. Rather, it aims to identify the overarching patterns in the way people were buried, and how these inform us about funerary rituals. The previous chapter signaled the persistence of older customs in the Roman period, such as the continuous use of pre-Roman cemeteries. As cities expanded, additional burial grounds were created, and it is here that new concepts of cemetery space, related to visibility and spatial connections between the tomb and settlement, started to appear. This chapter takes the theme of continuity and change to the tomb itself. The first part of the chapter presents an overview of the different architectural types by summarizing the more extensive discussion of tomb types presented in Appendix 2. A discussion about the reliability of the distributions of tomb types precedes this section. The second part investigates the combinations of different types across space and time, and assesses the degrees of diversity and uniformity in Syrian cemeteries. These are then compared to pre-Roman practices and connected to changing concepts of funerary display. The discussion at the end of the chapter problematizes the sample a bit further by focusing on research biases inherent in the data set. It concludes with an investigation of the possible architecture models for the new elements in the 37

38

THE TOMB

180 160 140 120 100 80 60 40 20 0

chart 1. Distribution of tomb types, cat. 1 – known types

tomb architecture, and places Syrian mortuary practices in the cultural milieu of the East Mediterranean, Rome, and the Iranian world.

TOMB TYPES

We can distinguish at least nine distinct architectural types of tomb in the sample from the Roman period. These immediately tell us that Syrians did not choose similar grave forms, or bury the dead in tombs of the same size, outlook, and material. Far from it, in fact. The funerary assemblage is characterized by a great eclecticism in architectural shapes and forms of decoration. In order to give a sense of the range, this section describes each of the architectural types separately. First, however, a few methodological notes are necessary. In the archaeology of Syria, and archaeology in general, little consensus exists about how to describe funerary architecture and what constitutes distinct tomb types. For the purposes of this book, I have attempted to bring some order to the collection of tombs by dividing them into categories based on three characteristics: shape, construction above or below ground, and single or communal burial, i.e., whether the tomb was designed for one person or more. Here, I have assumed that these features were the result of conscious planning or decision making among the burying community, and thus reveal what it thought was important. Such divisions are to some degree artificial, as they present my emphases rather than those of the people of the past. As we shall see, some overlap and mix forms existed. However, a degree of “boxing in,” putting an architectural label on a tomb, is vital in order to get a grip on the thousands of tombs of the Syrian province. The “unboxing” follows in the subsequent sections and chapters. The types that I distinguish are hypogea, mausolea, funerary enclosures, tower-tombs, tumuli, pit-graves, cist-graves, jar-burials, and sarcophagus groups placed in the open air (Chart 1). Each is described in this section. Three more

TOMB TYPES

39

70 60 50 40 30 20 10 0 stele (63)

coffins (7)

rock-reliefs (5)

chart 2. Distribution of tomb types, cat. 1 – unknown types

categories are identified in which the exact tomb shape is unclear (Chart 2). The first two are sarcophagi and stelae, which may have been placed in or on top of a grave, but have long since been removed from their original location. The third type consists of funerary rock-reliefs, where the associated site of burial was not found. The discussion of all tomb types is hindered by the uneven, incomplete, and limited state of publication. Images or drawings often lack, and the descriptions can be rather cursory. In the Introduction, I have outlined the methodology used to incorporate as much of this problematic material as possible into the study, and to add degrees of reliability. The sample on which Chart 1 is based represents those of the highest quality, or cat. 1. Chart 3 shows the distribution of tomb types from the second tier group, the so-called cat. 2 material. This group includes the same types, and adds various fragmentary shapes of which the original placement remains unknown, such as cippi, lintels, or sculpture, as well as a large group of unknown shapes about which the publications did not care to say more than that they were “tombs” (Chart 4).

350 300 250 200 150 100 50 0

chart 3. Distribution of tomb types, cat. 2 – known types

40

THE TOMB

450 400 350 300 250 200 150 100 50 0

chart 4. Distribution of tomb types, cat. 2 – unknown types

A comparison of the types in the cat. 1 and 2 sample (Charts 1 and 3) allows for several observations. Hypogea and pit-graves represent more than half of the assemblage, and various other types make up the rest. The assemblage, therefore, is split between two highly common shapes and a wider collection of less popular types of tomb. The differences between the two charts can, in part, be explained by archaeological recovery. For instance, the greater presence of mausolea and funerary enclosures in the cat. 1 data set is linked to their size and elaborate decoration: they were hard to miss and easier to date by archaeologists. Pit-graves, if not marked by a stele or filled with rich grave goods, were less likely to be noted or published. In general, the sample is skewed toward the larger (communal) and visible tomb types, which are more likely to be discovered and reported than small, single, underground forms such as jar-burials and cist- and pit-graves. Stelae and other inscribed stones are often published without context, with most emphasis on the inscription rather than the shape or material. They are more likely to end up in the cat. 2 assemblage. Compare, for instance, Chart 2 to Chart 4. As discussed later, the inscribed stones originally belonged to pit-graves, hypogea, and perhaps other communal tombs. It thus follows that the total number of these types of tomb was higher. We can also assume that sarcophagus burials were more popular than the sample suggests, because of their portability and their widespread reuse in later times. The high number of tower-tombs in Chart 3 reflects their common occurrence at only one site, Palmyra. In sum, based on the comparison of Charts 1–4, the distribution of tombs types should probably be modified as follows: mausolea, tower-tombs, and funerary enclosures were less common than the charts imply, and regionally constricted. Hypogea, pit-graves, cist-graves, jar-burials, and possibly sarcophagi in the open air were more frequent. It is

TOMB TYPES

good to emphasize, however, that none of these charts can be taken as reflecting a real distribution of types in Roman Syria. Rather, they provide an impression of the range of types that Syrians chose for their tombs, and their relative popularity.

Tomb-Types: Known Shapes (Chart 1) Before looking at the occurrence of the tomb types across time and space, we first briefly discuss each type, in order of popularity. A fuller description of the chronological development, regional spread, and architectural details can be found in Appendix 2. The simple pit-grave, cut into the bedrock or dug into the earth and consisting of a single burial space, was the most common form of burial and was used throughout the period at hand (Figure 8). Variation mostly occurred in methods of closing, which could be done by stone slabs, terracotta tiles, or heaps of earth and rubble. At Selenkahiye and other sites in the Middle Euphrates region, storage jars covered the pits. Sarcophagus lids more commonly closed the pit-graves of Northwest Syria. Some types of cover, such as sarcophagus lids and earthen heaps, were visible above the ground. Possibly more pit-graves had markers, for instance, in the form of funerary stelae (see discussion of stelae, p. 48). Yet, few examples have remained in situ. Pit-graves generally contained a single individual, who was placed directly on the floor or in a coffin made of terracotta, wood, stone, or lead. A variation of the pit-grave type was the cist-grave, with a burial pit aligned with stones or bricks forming a cist, covered by tiles, slabs, or earth. This type was uncommon and largely disappeared after the 2nd c. CE. Hypogea, or underground chamber-tombs, form the other most common type of burial. They were dug wherever the landscape allowed: in rocky outcrops and hills, and, in the case of Palmyra, into the hard clay surface. Many variations existed in the form and execution of the hypogeum, but the basic type consisted of an entrance corridor (dromos) leading to a small vestibule that opened to a central chamber (Figure 9). Burial usually took place in rectangular burial niches (loculi) cut in the three walls of the chamber, but it also occurred in free-standing stone sarcophagi and in pit-graves dug in the chamber floor. On average, six places were available for burial in the hypogea, but this number varied considerably. The large Palmyrene examples could accommodate hundreds of loculi. As in the case of the pit-graves, hypogea were located below the surface and could be marked above the ground. At least a third of all hypogea in the sample incorporated aboveground portions in various shapes: double columns, stelae, stone platforms, earthen tumuli, and decorated façades. With the exception of the earthen mounds, the aboveground features were introduced in the 2nd c. CE. In the same period, the hypogea became increasingly ornate. Reliefs of cornices, engaged columns, and other architectural features

41

42

THE TOMB

8. Pit- and cist-graves. A: Nawa-tell Umm el-Hauran, Tomb 139. B: Nawa-tell Umm el-Hauran, Tomb 156. C: Herbet Kalil, Tomb 1. D: Selenkahiye, Tomb X 23.3. E: Selenkahiye, Tomb T. 06.9

were added to the exterior façade; while inside, (figural) reliefs and plastered or painted walls could be found. Mausolea, the next type, are built, aboveground tombs. They came in several forms: circular and rectangular, in the shape of temples and towers, relatively solid stone squares with flat roofs, and open structures with columns supporting a pyramid-shaped roof (Figures 10, 44). Mausolea contained multiple burials, usually placed in loculi or coffins. Hybrid forms also existed, whereby a

TOMB TYPES

9. Hypogea. A: Hama, Tomb G XXVII (central chamber: 2.40 × 1.63 m). B: Hama, Tomb G XXVII (3.57 × 3.45 × 2.15/1.60 m). C: Dura Europos, Tomb 3 (central chamber: 3.10 × 2.70 × 3.5 m). D: Palmyra, Tomb of ’Abd’astor (central chamber: ca. 17.25 × 2.25–3 m). E: Jebleh, Tomb D (central chamber: 1.22 × 0.63 × 0.15 m)

mausoleum stood on top of a hypogeum, for instance, at Palmyra and Qanawat. The great diversity in size, shape, and finish of the mausolea highlights strong regional trends of this tomb type. Circular mausolea were characteristic of the Hauran and date to the 1st c. BCE–1st c. CE. They were sparsely decorated. By the 2nd c. CE, rectangular and square mausolea had replaced the

43

44

THE TOMB

10. Mausolea. A: Tomb 86 (Palmyra). B: Tomb 85b (Palmyra). C: Tomb 150 (Palmyra). D: Tomb 8 (Bosra). E: Tomb 7 (Bosra). F: Tomb 1 (R¯ımet al-Lohf). G: Mausoleum 1 (Apamea)

circular types in the Hauran and spread to other parts of Syria. Tetrastyle tombs, consisting of four columns or pillars supporting a roof that was often pyramidshaped, appeared on the Limestone Plateau, and those in Palmyra included an elaborately sculpted front reminiscent of temple and theater architecture. Like hypogea, mausolea were increasingly decorated in the 2nd c. CE. Thus far, no mausolea appear to have been reported from the Lebanese coast or in the steppe interior of Syria.

TOMB TYPES

A

B

11. Funerary enclosures (al-Bass Cemetery, Tyre). A: Complex 4 and 5. B: Complex 4 (left) and Complex 5 (right)

A type related to the mausoleum is the funerary enclosure. We have already seen the type in the al-Bass Cemetery of Tyre described in Chapter 1. It was only found on the Lebanese coast. It consists of a large, enclosed, but open-air space divided into several rooms or sections (Figure 11). Inside, multiple forms of burial took place: in rock-cut pits, sarcophagi, and loculi stacked in built platforms. The enclosures were decorated with relief, plaster, and paint, and

45

46

THE TOMB

some Tyrian examples included mosaic floors. On average, they held 22 burial spots,but the overall dimensions and number of spots varied considerably.Some included benches constructed against the interior walls, and rooms that appear not to have been used for burial. Gardens were sometimes incorporated in the structures. The earliest funerary enclosure arose late in the 1st c. CE, and the type dies out after the 4th c. CE. Tower-tombs are relatively well studied compared to the rest of the assemblage. The term is used fairly indiscriminately across the architectural types of Roman Syria. It is applied to tall, aboveground buildings of various shapes, such as the Palmyrene towers, the Tomb of Samsigeramos at Homs, and the Hermel building. This book makes a distinction between mausolea, such as those at Homs and Hermel, and tower-tombs proper. The latter are square buildings that were higher than wide and had several stories, which could be reached by a winding interior staircase (Figure 12). They are mostly known from Palmyra, but were also erected along the Euphrates north and east of Palmyra (see Appendix 2, p. 322). The earliest Palmyrene examples date to the second half of the 1st c. BCE. This type had loculi that were accessible from the exterior façade. Over time, burial moved inside, to tiers with stacked loculi and, starting in the mid-2nd c. CE, also to sarcophagi. The towers grew in size and the number of burial spots could reach several hundreds. Their construction continued until the early 2nd c. CE. Plastered walls were common in the oldest tower-tombs, whereas those of the late 1st and 2nd c. CE were decorated with reliefs, pilasters, moldings, and coffered ceilings. Combined with figural sculpture and great size, these Palmyrene tower-tombs became some of the most lavish funerary monuments of the Syrian province. A different type is represented by jar-burials: large to medium-sized jars placed in pits (Figure 13). Only a small number is included in the cat. 1 sample from Roman Syria, and the type appears more frequently beyond the eastern edges of the Roman province. Jar-burials were used for both cremation and inhumation burials. In the case of inhumations, the jars contained or covered the remains of infants and small children. The practice of cremation was unusual, and perhaps tied to the Roman military, as discussed in Chapter 4 (p. 110). Roman Syria has yielded large quantities of sarcophagi, mostly without context. They feature more extensively in the next section. A small group, however, forms a separate type, labeled “sarcophagi in the open air.” The type consists of groups of two or three stone sarcophagi raised on a pedestal (Figure 14). Often, the coffins formed a triclinium shape. At least four examples come from the database, and the ten additional groups in the cat. 2 sample suggest a modest popularity on the Limestone Plateau in the 2nd and 3rd c. CE. Tumuli form the final category.As in the case of the tower-tombs,the definition of “tumulus” varies, and the term is applied in the scholarship to different

TOMB TYPES

12. Tower-tombs. A: Tomb of Elahbel (Palmyra). B: Tomb of Kithot (Palmyra). C: Tower-tomb at Baghuz

types, such as round mausolea and hypogea under an earthen mound. In this book, tumuli are communal tombs consisting of a circular mound of earth and rubble in which pit-graves were dug, or which covered a built grave chamber (Figure 15). They are distinct from hypogea covered by a mound and from circular, built tombs. The cat. 1 database contains only three tumulus-tombs, but the type was more widespread both during and before the Roman period.

47

48

THE TOMB

13. Jar-burials. A: Tomb 19 (Apamea). B: Section with Tomb 19 (Apamea). C: Section (top) and plan (middle, bottom) of Tomb B (Jebleh). D: Tomb R21.5 (Selenkahiye)

Several regional traditions in tumulus construction appear to have existed. Those in the Hauran consisted of a rubble mound covering a small burial chamber with a central support pillar. Their construction was perhaps limited to the 1st c. BCE and the 1st c. CE. In the Syrian Upper Euphrates area around the Taqba dam,tumulus-tombs were larger than the examples from the Hauran, and possibly of 3rd c. CE or later date.

Unknown Shapes: Stelae (Chart 2) A significant portion of the funerary material from Roman Syria comes in fragmentary form. Here, the appearance of the original tomb is not known or certain. Funerary stelae form the largest group, and were common throughout the Roman province, except in East Syria (Figure 16). The majority date

TOMB TYPES

14. Sarcophagi in the open air.A:Tomb 7 (Si’,three sarcophagi on triangular pedestal).B:Pedestal of Sarcophagus Tomb 1 (Apamea)

to the 2nd and 3rd c. CE, after which their number declined. A major question concerning the stelae is their original placement. According to a short report published in the 1930s, stelae still stood atop pit-graves at the Northeast Cemetery of Palmyra. Unfortunately, photos or drawings lack. On the Limestone Plateau, stelae sometimes marked hypogea and pit-graves (T. 12–14), and at Umm el-Jimal in North Jordan, twenty-three examples stood in rows against an exterior wall of a partly sunken mausoleum.According to Butler,these stelae

49

50

THE TOMB

A

B

15. Tumuli. A: Tumulus Q10 in Qanawat. B: Plans of tumuli in the Hauran

stood in their original location.1 Stelae, thus, could be placed outside different types of tombs. At least in one case, a stele was found inside a hypogeum (T. 15). The shape of the stele could also indicate the original placement. The pointed 1

Butler 1920, 209–210.

TOMB TYPES

16. Funerary stelae. A: Stele 51 (Palmyra). B: Stele 3 (Apamea). C: Stele 44 (Apamea). D: Stele 32 (Si’, cat. 2)

51

52

THE TOMB

base of Palmyrene stelae and the undecorated lower third portion of the stelae at Apamea suggest that they were inserted into the ground or into rock cuttings. It remains uncertain, nevertheless, whether all stelae physically belonged to a grave or whether they were set up in non-funerary settings, where they served as cenotaphs or other types of memorials. We return to this question in Chapter 5. The published funerary stelae usually had an inscription, sometimes in combination with an image in relief. Often, the deceased are depicted, in bust form or as standing, seated, or reclining full-length figures. Additional relief decoration comes in the form of floral and geometric motifs. The stelae varied considerably in shape, decoration, and inscription, often forming distinct regional groups. Examples from Palmyra, for instance, depicted hanging curtains and palm leaves coupled with Palmyrene-Aramaic texts (Figure 16). Stylized figural representations are common on the porous basalt stelae from the Hauran. Soldiers of the Roman army were commemorated in stelae, often depicting military attributes (Figure 34).

Unknown Shapes: Coffins (Chart 2) Almost 600 coffins were counted in the sample of cat. 1 tombs from Syria. Some formed their own category, such as the sarcophagi groups placed in the open air described earlier. Others stood inside tombs. The seven cases listed as a separate category in Chart 2 are published without any information of their original location. Because coffins were an important feature of Roman tombs, this section discusses the main patterns of the entire assemblage, divided by material of construction. Wooden coffins were likely the most widespread type, but are rarely preserved. Their presence is usually deduced from finds of metal coffin fittings and multiple iron nails, some with pieces of wood still attached. Such coffin fragments are reported from all tomb types. Excavation reports state that each loculus in the hypogea of Dura Europos contained a wooden sarcophagus. Pitgraves at Homs yielded elaborately constructed examples with lead, silver, iron, and gold fittings and ornaments, and an example from a hypogeum in Hama was covered with stucco decoration. Terracotta coffins were also popular throughout the Roman period, and are found in cist- and pit-graves, funerary enclosures, and hypogea. Few studies exist of these containers, yet they appear to follow two regional stylistic trends. Rectangular shapes with lids of terracotta tiles were common on the Mediterranean coast. Analysis of terracotta coffins from Beirut has identified eastern Cilicia as a place of production. The second type adhered to Mesopotamian-Parthian traditions. Rectangular examples with rounded edges,

TOMB TYPES

or a trough shape, came from Dura Europos. Eastern Syria has also yielded tub-shaped coffins, but none could be securely dated to the Roman period. This shape was common in Mesopotamia and, for instance, discovered in the Hellenistic cemeteries of Uruk.2 Gawlikowski notes that the pit-graves of the Northeast Cemetery at Palmyra contained so-called “slipper-shaped coffins,” another characteristically Parthian-Mesopotamian type. No images are known, but if this is correct, they may represent the westernmost attestation of such coffins.3 Stone coffins, or sarcophagi, were popular across Syria (Figure 17). They are found in all types of tomb, with the exception of tumuli and cist-graves. A significant portion in the sample, more than 350, originated from the funerary enclosures of the al-Bass Cemetery at Tyre. Sarcophagi arose as popular burial containers late in the 1st c. CE, a trend that accelerated after the mid-2nd c. CE. By this date, the collection of locally produced coffins made of limestone, basalt, and sandstone was augmented by imports. Marble sarcophagi originated from Proconnesos in Turkey and Attica in Greece. Several came from Egypt, made in pink granite and gray porphyrite, and one group of gray-purple coffins may have been imported from Assos in Turkey.4 Imports are usually restricted to Syrian coastal sites with easy access to the sea. Some travelled further inland, such as the Pentelic (Attic) marble sarcophagus at Bosra (T. 16) or a Proconnesian sarcophagus at Restan, near Homs.5 Popular decoration of imported sarcophagi consisted of moldings, garlands, animal heads, floral motifs, fruit, gorgon faces (gorgoneion), and framed spaces for epitaphs (tabellae ansatae). Occasionally, reclining figures carved in the round were placed on the lid. Elaborate reliefs with depictions from Greek mythology, such as the life of Achilles and Bacchic scenes, adorned a small number of imported coffins in Tyre (cf., Figure 40). Many of the locally produced sarcophagi were plain; others followed the decoration schemes of the imported coffins or developed different styles. This latter group displayed strong regional preferences, with depictions of lion heads in Bosra, and disks and pelta-shaped shields at Douris and Baalbek. Group reclining scenes come from Palmyra, as do busts, camels, and other features typical of life in this desert city. Lead coffins represent a late addition to the assemblage. They appear in the mid–late 2nd c. CE, and retained their popularity well into the Byzantine period. Their production centers presumably stood along the Levantine coast, which explains their popularity in the western parts of the province. Lead coffins were found in hypogea, pit-graves, and funerary enclosures. Several examples from the Deb’aal Tomb close to Tyre were placed inside a stone 2 3

East Syria: Geyer & Monchambert 2003, 164; Novák 2000. Uruk: Boehmer et al. 1995. 4 5 Gawlikowski 1970, 34. De Jong 2010, 607. Restan: Gatier 1997–1998.

53

54

THE TOMB

17. Sarcophagi. A: S. 1133 (Tyre, al-Bass, Complex 16). B: Lower sarcophagus from Pit 2 (Baalbek, Douris). C: S. 2 in Hypogeum 1 (Apamea)

TOMB TYPES

18. Rock-reliefs. Qatura. A: Niche of Augaios. B: Niche of Barathes. C: Rock-reliefs in cliff

sarcophagus. The rectangular or trapezoid-shaped coffins were elaborately decorated on the lid and box. Common were geometric and floral motives, depictions of architecture (temple façade, columns), masks, sphinxes, lions, and standing or seated persons.

Unknown Shapes: Rock-Reliefs (Chart 2) A cliff face at Qatura on the Limestone Plateau contained at least thirteen reliefs depicting seated or standing people in rectangular frames, dating to the 2nd c. CE (Figure 18). They feature in the introduction of this book. The framed reliefs were not associated with graves, although Butler believed that pit-graves were located at an undiscovered location in the bottom of the ravine under

55

56

THE TOMB

the niches. Others considered them cenotaphs, i.e., commemorative memorials without graves.6 The type of decoration, combination of figural relief and text, and content of the inscriptions find close parallels with funerary material from the Limestone Plateau and elsewhere in the province, and confirm their commemorative nature. Several other relief groups associated with cemeteries come from Roman Syria, but their publication record is poor, and few are dated. Examples have been found elsewhere on the Limestone Plateau and in the steppe lands directly to the east, as well as in the vicinity of Apamea, Damascus, Byblos, and Tyre, and in the Beqa’ Valley. They are described in detail in Appendix 2. The reliefs depicted seated and standing people, busts, and stylized stelae or cone-shaped projections. The latter are often considered aniconic depictions representing the spirit of the deceased, or nfs/nefesh – a discussion to which we will return in Chapter 5 (p. 159). The rock-reliefs and associated tomb structures require extensive research, but some preliminary conclusions can be drawn. In several regions of Syria, cliff walls depicted iconic and aniconic scenes. They likely date to the Roman period or to the preceding Hellenistic centuries. Their spatial association with tombs, pit-graves, and hypogea, and the inclusion of epitaphs, point to a funerary nature. A final word about the fragmentary materials from the second-tier assemblage (Chart 4). The largest group is formed by inscriptions published in epigraphic corpora. Such corpora sometimes add a few lines about the shape of the stone that held the inscription (e.g., stele, altar, lintel, column, or cippus), but little else is included about the associated tomb, decoration, and placement. The chart also contains fragments of sculpture that, according to the publications, belonged to funerary monuments, such as eagles and busts (T. 17–24). Without images or justification, this identification is difficult to assess.

DIVERSITY AND UNIFORMITY IN FUNERARY ARCHITECTURE

Despite the gaps in the publication record and decontextualized preservation, the remains tell us a great deal about the choices that guided the construction of the final resting places of Syrian communities. Most of the time, they chose pit-graves and hypogea, two types that often appeared in the same cemeteries. In some cases, multiple pit-graves surrounded a single hypogeum. Syrians also turned to a variety of other tomb types, and the burial grounds of the province reflect this mixture of shapes. One of the main overarching principles of Syrian mortuary practices was diversity. People were not buried in similar types of tombs or containers, nor were they commemorated with the same epitaphs or 6

Butler 1920, 249–250; Griesheimer 1997a, 170; Peña et al. 1999, 154–163.

DIVERSITY AND UNIFORMITY IN FUNERARY ARCHITECTURE

reliefs. Instead, Syrian cemeteries display an eclectic mix of tombs, coffins, and forms of decoration. The cemeteries of Apamea illustrate how the variation in the burial grounds developed. The oldest tombs here are jar-burials and pit-graves with cremations, possibly of the 1st c. CE. To the same period dates a stele with a Greek inscription commemorating a woman and her infant son.In next two centuries, hypogea were constructed, and numerous funerary stelae honored Roman soldiers and local inhabitants of the city. Groups of decorated sarcophagi stood on pedestals in the open air, and a brick mausoleum housing two stone coffins aligned the main north–south road. All the while, pit-graves and cist-graves continued to be constructed. By the middle of the 3rd c. CE, the older communal tombs (hypogea, mausolea, sarcophagus groups) were likely still operational and some new cist-graves were constructed, but overall, the diversity in the Apamean burial grounds had diminished. The final stele dates to around 252 CE. Elsewhere in Syria, the urban cemeteries also witnessed an increase in variety of tomb types and decorative styles, with a peak in the 2nd c. CE. Toward the late 2nd c. CE, diversity decreased, following the cessation of the construction of tower-tombs and possibly cist-graves. A further slimming down occurred after the 3rd c. CE with the disappearance from the funerary record of jar-burials and funerary enclosures, and a reduction in popularity of stelae. This pattern of increasing and decreasing diversity in funerary architecture and decoration was mostly an urban phenomenon. The cemeteries of smaller and rural sites were less diverse. Most sites in the Hauran, for instance, yielded a single mausoleum and a handful of stelae, whereas on the Limestone Plateau, single hypogea were common, sometimes in combination with pit- or sarcophagus-burials. Larger settlements in the Hauran, such as Qanawat, Shahb¯a, and Si’, displayed more diversity in tomb types. It seems that the larger the site and, to some extent, the longer the time-span of usage, the more varied the types of tomb in the associated cemeteries. Even the urban cemeteries display considerable differences in levels of variation. The Apamean examples yield at least six tomb types, whereas the alBass Cemetery of Tyre hosted solely funerary enclosures. Four cemeteries at Palmyra included hypogea, mausolea, and tower-tombs, but a fifth – the Northeast Cemetery – held only pit-graves with stelae. At this point, it is useful to start unpacking the tomb-type categories of the previous section, and to look more closely at how people were buried. For instance, a lining of stones in a pit was the only thing that distinguished cist-graves from pit-graves at Apamea, and the differences between other tomb types were perhaps not always so pronounced, either. The al-Bass Cemetery included only a single type, and its rows of funerary enclosures give an impression of homogeneity, yet this picture changes quickly once one looks inside the tombs. Here, we find a great

57

58

THE TOMB

variety, with people buried in stacked loculi, pit-graves, and local and imported, plain and elaborately decorated coffins. In other words, when zooming in on regions and cemeteries, the strong patterning of increasing diversity in funerary architecture dissolves and appears more random. In order to assess, therefore, whether people in Roman Syria were buried in similar or distinct ways, the focus should move away from (only) formal tomb types. Minor architectural differences aside, does the funerary record suggest that Syrians made the same choices in the construction of tombs? The answer to this question is no, and to a lesser extent, yes. Great variation existed within the cemetery and within the tombs, in terms of shape, degree of adornment, materials used, and so on. People were placed in single pit-graves or communal hypogea. In the communal tombs, we find the deceased deposited on the floor, in loculi, or in coffins, which could be made of marble, lead, or wood. Some burial spots were elaborately decorated with busts of the deceased or with myths carved in expensive high relief. Others were left plain. Even simple pit-graves could contain stone coffins or a cover of marble or terracotta. This does not give an impression of homogeneity in burial types, and clearly indicates that Syrians were not buried in similar fashion. The next chapters address whether it is possible to attach certain identities (i.e., status, gender, profession) to these distinctive practices. The notion of eclecticism within cemeteries of Syrian cities stands in contrast to the simultaneous increasing standardization of location of the burial grounds, as discussed in the previous chapter. This adds further support to the claim that the new concept of space was not intended to present members of the burying community in similar ways, but that it was related to a new or stronger emphasis on the links with the urban landscape. In other words, with the overhaul of the urban landscape, the suburban area, where the cemeteries were found, followed suit. Nevertheless, and here we come to the second answer to the question of variability in funerary architecture, in spite of the great variation in how people were buried, certain overarching principles seemed to guide the construction of many tombs. These concern regionalization, compartmentalization, single and co-burial, visibility, embellishment, and display. It is here that the divided communities and their choices for burial appear more unified. They are discussed next.

Regional Styles Roman cemeteries in Syria flourished in regional styles. Every region, by which is meant geographically defined territories encompassing multiple settlements, developed particular tomb shapes, decorative styles, or other signature aspects. This point can be best illustrated by summing up the regional

DIVERSITY AND UNIFORMITY IN FUNERARY ARCHITECTURE

trends: characteristic of the Limestone Plateau were hypogea marked by double columns (2nd c.CE),hypogea with decorated façades (2nd c.CE–Byzantine period), elevated sarcophagi (2nd/3rd c. CE–Byzantine period), tetrastyle mausolea (3rd c. CE–Byzantine period), and pit-graves with sarcophagus lids (mid 2nd c.CE–Byzantine period).In the Syrian Upper Euphrates region,large storage jars covered pit-graves in the 1st–2nd c. CE. Distinct types in eastern Syria included pit-graves with trough- and perhaps tub-shaped terracotta sarcophagi (Parthian period–3rd c. CE?), as well as a dearth of funerary stelae. Funerary enclosures arose on the Lebanese coast between the late 1st and the 4th c. CE. Imported stone coffins reached the Mediterranean coast and West Syria in the late 1st–3rd c. CE, and the same area produced lead sarcophagi in the mid to late 2nd c. CE–Byzantine period. In the hills of southern Lebanon, painted hypogea were in vogue between the late 1st and the 3rd c. CE. Rectangular mausolea gained popularity in Central–West Syria in the 2nd c. CE, perhaps around the same time as the use of rock-reliefs to mark graves. Tower-tombs were characteristic of Palmyra and the Middle Euphrates between the late 1st c. BCE and the early/mid 2nd c. CE. Tower-tombs with hypogea (1st–2nd c. CE), large T-shaped hypogea (late 1st c. CE–2nd c. CE), mausolea with elaborate façades (mid 2nd–mid 3rd c. CE), and stelae depicting curtains (1/50–150 CE) were typical for Palmyra. The Hauran stood out for its tumuli (100 BCE– 100 CE) and circular tombs (1st c. CE?), and had distinct sculptural traditions expressed in sarcophagi and stelae. Sarcophagi and funerary reliefs in Palmyra also represented distinct sculptural traditions, as did perhaps the sarcophagi in the Beqa’ Valley. This long list illustrates regional distinctions in funerary architecture in Roman Syria. From largest to smallest, these regions are: East Syria (east of the Euphrates), West Syria (including the Middle Euphrates region), Central– West Syria (Limestone Plateau, the Orontes Valley, and the Hauran), the Middle Euphrates region, the Coastal Levant, the Limestone Plateau, the Hauran, South Lebanon, and Palmyra. Regional styles, for instance, in figural sculpture and stele decoration, extended across the provincial borders into southeast Turkey and northwest Palestine, and, in the case of the tower-tombs and terracotta coffins, beyond the imperial boundaries into Parthian territory. With signature tomb shapes and portrait busts, the Palmyrenes arguably developed the most distinct funerary culture. The flourishing of regional funerary styles can be clearly traced in time. Tumuli and circular mausolea in the Hauran and the aforementioned Palmyrene tower-tombs are early examples of regional expressions, dating between the later 1st c. BCE and the 1st c. CE. By the end of the 1st c. CE, funerary enclosures and imported sarcophagi appeared in the coastal Levant, painted hypogea in southern Lebanon, and figural sculpture in Palmyra. Regional diversity had reached its fullest expression in the first half of the 2nd c. CE, and continued on a diminished scale in the Byzantine period.

59

60

THE TOMB

The following chapters illustrate that pronounced regionalism also played a role in sculpture and epitaphs, but not so much in the way the body of the deceased was treated or the gifts that were selected to accompany them. Regionalism became a key element in constructing local identities in the Roman province, but did not infiltrate all aspects of the mortuary ritual.

Spatial Divisions If we move away from the formal tomb types, and look at single and multipleoccupancy tombs, considerable differences start to emerge in the sample. The layout of pit-graves, cist-graves, and jar-burials (Figures 8, 13) was designed for one burial at the time. Communal tombs held more than one spatially delineated burial spot, such as loculi, pits, and coffins, and could accommodate multiple simultaneous burials. Hypogea, mausolea, funerary enclosures, towertombs, and tumuli are all communal tombs (Figures 9–12, 15). Within the sample, communal tombs make up about a third more than the single tombs, but this may not represent a real pattern. Many single tombs, simpler in design, likely remain unpublished. It is also unclear to what type the funerary stelae, displaced sarcophagi, and rock-reliefs belonged. What is important here is that both types were common, and they were popular around the same time. Sometimes, they are grouped: single pits surrounded hypogea on the Limestone Plateau and a mausoleum at Homs. Elsewhere, they were physically separated, such as in the Northeast Cemetery of Palmyra, which contained only pit-graves. At Selenkahiye, a cemetery with single tombs occupied the mound, whereas possibly contemporaneous communal tumuli extended at the bottom of the mound. Urban cemeteries of Beirut and Tyre may have held only communal tombs, and at Tell Kazel, only single types were discovered. Few cross-provincial patterns emerge from the sample, except that burial grounds consisting only of single tombs were rare and mostly restricted to non-urban regions. This picture may change rapidly as a result of current excavations in the urban setting of Beirut, where burial grounds consisting of pit- and cistgraves are currently under excavation.7 The following chapters illustrate that the distinction between single and communal tombs concerned more than just a choice of architectural types. It involved also different social classes, as well as different concepts of ritual space and display of kinship. In the communal tombs, there is little evidence of spatial hierarchy within, i.e., the placement of certain graves in more prominent positions. Many of the hypogea had burial niches along three walls, directing axial attention to the loculus in the back wall. Yet no other means were employed to mark these graves, e.g., through sculpture or painted decoration. Sarcophagi were often 7

Personal communication with Vana Kalenderian.

OLD AND NEW IN FUNERARY ARCHITECTURE

prominently and centrally placed inside a tomb, to such an extent that they blocked access to other burial spots, but this was not always the case, and no cross-provincial patterns emerge.Axiality was more important at Palmyra,especially in the hypogea (Figure 9d). Here, the elongated spaces focused attention on the back wall, where the tomb of the founder was located, often behind an elaborate sculptural relief. But even in Palmyra, this was not uniform in every tomb, and side-niches and lateral rooms sometimes yielded equally lavish sculptural scenes and decorated coffins. A different kind of ordering of space did occur in the Syrian tombs, in both communal and single types. Across the various types of tomb of Roman Syria, people were buried in relatively narrow rectangular spaces: coffins, pits, and niches. Exceptions are hypogea where multiple individuals were placed on the floor of a chamber, such as at Beirut (T. 25) and Hama (T. 26). Yet, in both cases, we cannot be certain that the corpse was not placed in a wooden coffin, no longer preserved, and that we are not looking at the result of later disturbances. At Dura Europos, the earliest hypogea of the Hellenistic period contained benches on which corpses were placed, but over time, the tombs grew and loculi were added. The compartmentalization of burial space indicates the existence of strong sentiments about where a corpse was placed in relation to the space of the tomb. The compartments also give an impression of individualizing the space of the tomb: each person lies in his or her own loculus, pit, or sarcophagus. Chapter 4 demonstrates that in practice, this was not upheld. Even within compartments with the smallest dimensions, multiple individuals were buried. OLD AND NEW IN FUNERARY ARCHITECTURE

Thus far, this chapter has concentrated on trends in funerary architecture in the Roman period. To what extent did they follow older traditions? We start this discussion by turning to two pre-Roman sites that yield relatively high numbers of tombs (Figure 19). In Beirut, a cemetery occupied the older burial ground to the west of the town (see Appendix 1). This Iron Age burial ground contained hypogea and shaft-graves, i.e., pit-graves that are accessed by deep vertical shafts. Several of the tombs were still used in the Hellenistic period (4th–1st c. BCE) and perhaps the early Roman decades, to be abandoned only when the city expanded. On the other side of town, east of the city wall, extended a field containing twenty pit-graves dating to the 2nd c. BCE (Bey 152). A nearby third location (Bey 045) yielded several 4th–3rd c. BCE pitgraves. Throughout the Hellenistic period, pit-graves and hypogea seem to have been the preferred mode of burial in Beirut. Moving east to the Khabur region in the Syrian steppe, an extensive cemetery was excavated at Tell Sheikh Hamad. At least 140 tombs were found here,

61

62

THE TOMB

19. Pre-Roman graves. A: Hypogeum 8 (Beirut). B: Pit-graves Bey 045 (Beirut). C: Jar-burial (Tell Sheikh Hamad, Grab 84/03). D: Sarcophagus burial (Tell Sheikh Hamad, Grab 93/122). E: Cist-grave (Tell Sheikh Hamad, Grab 84/101)

dated between the 2nd c. BCE and the 2nd c. CE, comprising the Hellenistic and Parthian periods in this region. Pit-graves and cist-graves lined with mudbricks represent the most common types. In some cases, the deceased were placed in oval and tub-shaped terracotta coffins, while others were covered by or put inside jars.8 One hypogeum was discovered at a different location, dating to the 1st–3rd c. CE. In Beirut and Tell Sheikh Hamad, hypogea and pit-type graves, e.g., shafts, jars, coffins in pits, and cists, represent the most common modes of burial. This is also true elsewhere, as illustrated by Chart 5. The continued popularity of these types in the Roman period, save for stylistic 8

Kühne 2005; Novák 2000.

OLD AND NEW IN FUNERARY ARCHITECTURE

60 50 40 30 20 10 0

63

pit-grave (120) cist-grave (63) hypogeum (50) jar-burial (20) tower-tomb (7) Beirut (32)

Dura Jebel Jebleh (3) Palmyra (2) Tell Kazel Tell Sheikh Europos Khalid (48) (45) Hamad (66) (65)

shaft-grave (1)

chart 5. Distribution of tomb types – Hellenistic and Parthian periods

alterations, highlights the deep-rootedness of these forms of burial. Communal and single tombs occurred side-by-side, although more cemeteries contained only single types when compared to the Roman assemblage. Chart 5 also illustrates another point. Hellenistic and Parthian cemeteries displayed far less variation than their Roman successors. This is directly related to the introduction of new types in the period between the 1st c. BCE and the 2nd c. CE. Appendix 2 describes the evolution and chronological development of each of the tomb types, as well as the possible pre-Roman antecedents. Here, a summary suffices. Pit-, cist-, and jar-burials, as well as hypogea, represent older types. No mausolea can be securely dated before the 1st c. BCE. Debate exists on this matter, and scholars have pointed to a 28 m-high tower at Hermel in the northern Beqa’ Valley as a Hellenistic example (p. 320).9 Yet, both the date and the funerary nature of this tomb remain uncertain. The funerary enclosures of the Lebanese coast also had no predecessors before the Roman period, although different components of this type, such as burial in rock-cut pits or loculi, pre-dated the Roman centuries. The same is true of the groups of stone sarcophagi raised on a pedestal, as we find them in the Syrian province of the 2nd and 3rd c. CE. The construction of tumuli in the Hauran started in the 1st c. BCE. No examples have been identified from the preceding centuries. The fact that a much older (Chalcolithic and Bronze Age) tradition in tumulus building existed in the region may indicate that the inspiration for this new type was drawn locally. The last section of this chapter discusses more examples of this phenomenon, whereby Syrians may have mimicked funerary architecture of times long past. The origin of another newcomer, the tower-tomb, has received more attention. Several predecessors have been identified, yet the evidence is slim and hardly convincing (see discussion later). To this date, the tower-tombs at Palmyra and the Euphrates sites, constructed between the second half or final quarter of the 1st c. BCE and the early 2nd c. CE, form a distinct group without obvious forerunners. It is important to note that both the tower-tombs 9

For instance, Fedak 1990, 148; Cormack 1997b, 348.

64

THE TOMB

and the tumuli were not restricted to Roman territory per se. Both emerged late in the 1st c. BCE, in a period when provincial boundaries continuously moved (p. 11). The Hauran and Palmyra fell in and out of Roman control, whereas the Euphrates region of Dura Europos and Baghuz was in Parthian hands. The introduction of new tomb types did not, necessarily, accompany the establishment of the Roman rule. In fact, a close examination of the chronology of change reveals an important point. The new types appeared between the 1st c. BCE and the 2nd c. CE. In this period, more than 150 years after the creation of the Roman province in 64 BCE, we can truly speak of eclectic burial grounds as just described, where older tomb shapes stand side by side with new types. This timing of change also applies to the so-called “unknown shapes.” Funerary stelae occurred in Hellenistic contexts, but this practice was not widespread. The majority date to the 2nd and 3rd c. CE. Stone and terracotta coffins were common in pre-Roman periods, although very few can be placed in the centuries directly before the 1st c. BCE. They become increasingly visible in the archaeological record of the late 1st and 2nd c. CE. It is possible that rock-reliefs drew on older, or at least Hellenistic, practices, yet it was only in the 2nd and 3rd c. CE that inscriptions with dates accompanied the reliefs. The available evidence from Syria points to an overhaul of funerary architecture there during the first centuries CE. The degree of regional variation in pre-Roman cemeteries is more difficult to assess. To some extent, regional variation always exists in the material record of funerary practices. In pre-Roman Syria, the most pronounced distinction was between East Syria and West Syria/Lebanon. The irregularly shaped hypogea and deep shaft-graves found in Hellenistic Beirut are characteristic types of the Levant, whereas the tub-shaped and oval terracotta coffins of Tell Sheikh Hamad are a peculiar feature of East Syria and further east in Mesopotamia. The use of stone as a building material, the import of sarcophagi, and the erection of stelae were largely restricted to the Mediterranean coast and West Syria, whereas mudbricks, pits with tub-shaped terracotta coffins, and jar-burials characterize cemeteries further east. A cultural border is not clearly defined, but it may have lain slightly west of Palmyra and the Balikh Valley. A lack of precisely dated tombs does not allow for ascertainment of whether this east–west distinction was characteristic of the Hellenistic period or only appeared after the Parthian conquest of eastern Syria in the late 2nd c. BCE. Tomb types and materials in the pre-Parthian cemeteries of northern Iraq suggest that similar differences already existed before the Parthian expansion.10 10

For instance: cemeteries at Nimrud (250–145 BCE), with cist-graves of baked bricks, troughshaped sarcophagi in terracotta, and jar-burials (Oates & Oates 1958); Seleucia-on-the-Tigris (ca. 307–141 BCE), with similar cist-graves and baked brick hypogea with vaulted roofs (Graziosi 1968–1969; Hopkins 1972; Invernizzi 1967; Negro Ponzi 1970–1971, 1972, 2002; Valtz 1986, 1988; Waterman 1931; Yeivin 1933); Tell ed-Der (325–200 BCE), with cist-graves of baked brick (Gasche 1991, 1996); and Uruk (400–200 BCE), with tub- and reversed-tubshaped terracotta sarcophagi (Boehmer et al. 1995).

A NEW CONCEPT OF FUNERARY SPACE

As we shall see, this distinction persisted throughout the Roman centuries, with Palmyra combining in its funerary architecture and decoration elements from both East and West Syria. In spite of the strong east–west differences, however, Hellenistic and Parthian burial grounds never attained the degree of regionalism that characterized the Roman examples. This has much to do with the contexts in which we find this regional expression. New forms of decoration, such as sculpted sarcophagi, busts, and painted walls, and built portions of the tomb, most notably those aboveground (i.e., funerary stelae, decorated façades, and mausolea), became canvasses on which regional styles were prominently expressed. Hellenistic and Parthian tombs were relatively plain, and rarely contained aboveground makers. It is here that Roman tombs started to part with older traditions. A NEW CONCEPT OF FUNERARY SPACE: VISIBILITY AND MONUMENTALITY

What characterizes the new tombs of the Roman period? As discussed, one can question the differences between the tomb types identified in this book, and point to the somewhat arbitrary nature of these categories as developed by archaeologists. Many practices remained the same, such as burial in rectangular compartments, and the co-existence of communal and single tombs. This section, therefore, takes a different approach and addresses how the new tomb types involved a different concept of funerary space. The tombs shared a number of features regarding visibility and elaboration, and, importantly, this was also introduced in the older architectural forms. In other words, there was a general trend toward greater spending of resources on the construction and decoration of funerary architecture. Sometimes, it took monumental shape. The architectural types introduced in the 1st c. BCE and later centuries were visible aboveground, either by the addition of an aboveground portion, such as tumuli or rock-reliefs, or by raising the burial space itself above the surface, as in the case of the mausolea, elevated sarcophagi, tower-tombs, and funerary enclosures. An increasing emphasis on visibility was also noticeable with regards to older architectural forms. Hypogea in the Roman period could be marked with columns, stelae, or other standing architecture. The reliefs on the exterior cliff walls of the hypogea drew attention to the façades and entrances of the tombs. Visibility was enhanced by elevated placement. The double pillars of the distyle tombs, in themselves already reaching 10 m in height, stood on the side of hills (cf., Figure 45). Griesheimer mentions that the example from Turin was visible in at least five of the surrounding (ancient) villages (T. 27).11 Towertombs stood over 20 m tall on elevated spots in the landscape, and sarcophagi were raised on pedestals. Even simpler graves were marked: pit-graves received a 11

Griesheimer 1997a, 186.

65

66

THE TOMB

stone, gabled sarcophagus lid, or a stele. We have already seen that many tombs were aligned with the main roads leading to and from the urban centers, a location that drew further attention to the tomb buildings. Close to half of all the tombs in the sample were, at least in part, aboveground (this number excludes the stelae and sarcophagi of which the original location is uncertain). Research biases likely overemphasize the popularity of grave markers, which are easier to find in the archaeological record and were thus more commonly recorded. The percentage of the total tombs that were visually marked remains unclear. However, the fact that the trend can be discerned across all types of tomb, both communal and single, and in every region of the province, suggests that it constituted a widespread phenomenon. An epitaph from the Hypogeum of Zabda in Palmyra (T.28) honors the founders who built the tomb and “the construction that is above it.” No trace remains of this construction, but the inscription serves as a reminder that even underground tombs could be marked by structures long gone.12 Tomb markers existed before the Roman period, but with notable differences. Amphora rims (Tell Sheikh Hamad) or tumuli (Dura Europos) marked Parthian tombs, and cairns were associated with pit-graves in Hellenistic Beirut (Bey 045). A handful of funerary stelae originated from the Levantine coast in the Hellenistic centuries. None were found in situ, but they may have marked the locations of underground burial spots. A possible trend in relief carving of cliff façades existed in the Lebanese mountains, although, as described already, both the funerary nature and date of the reliefs remain uncertain. None of the pre-Roman tombs had funerary space above the ground, and, with the possible exception of the Hermel tower, the existing markers did not draw attention to the tombs by large dimensions in the way they did in the Roman period.

Resources: Size, Building Material, and Decoration An increase in the amount of resources spent on the tombs represents a second characteristic of the Roman funerary assemblage, after visibility. Construction, size, and decoration are examples of this trend. Building materials now more often included imports, mostly in the form of coffins made of marble or other stones from Turkey, Egypt, or Greece. These imports started around the late 1st c. BCE and increased in the 2nd and subsequent centuries. Tombs also grew larger, although sizes varied widely. Some hypogea reached twice or three times the size of the average Parthian and Hellenistic hypogea. The largest tomb in pre-Roman Dura Europos (55 m2 ) could easily fit inside the Hypogeum of Artaban at Palmyra (89 m2 ). The new tomb types for the most part also stood 12

Henning (2013, 15) also notes that new research in Palmyra identified aboveground markers by the entrances to hypogea.

A NEW CONCEPT OF FUNERARY SPACE

tall. The average rectangular mausoleum measured ca. 11 × 11 × 5 m, and funerary enclosures covered on average 200 m2 . Perhaps the increase in resources directed toward funerary architecture appears most clearly when considering the levels of embellishment, i.e., the addition of decoration. Almost half of all Roman tombs in the sample received some sort of decoration, or non-structural elements used for adornment; this number excludes the sarcophagi (Figures 3, 12, 14, 18, and 20). Decoration was sparse in the cemeteries before the 1st c. CE, and came in the form of reliefs on stelae from the Levantine coast, white plastered walls on the hypogea of Dura Europos, and the previously mentioned Lebanese rock-reliefs.13 The earliest new types, tumuli and tower-tombs, also received minimal decorative embellishment. It was only in the later 1st and 2nd c. CE that relief, paint, and sculpture in the round appeared in large numbers in the burial grounds, including in the later tower-tombs and the rectangular mausolea that replaced the Hauranite tumuli. Chapter 4 describes a similar chronology for the adoption of another form of embellishment: the funerary inscription. What motifs were used in Roman tombs? Most forms of decoration were in relief, but sculpture in the round, as well as plaster and paint, is also attested. Some Tyrian funerary enclosures had mosaic floors. Most popular were forms of architectural decoration, such as pediments, columns, cornices, and the like. Geometric patterns and floral and faunal elements such as eagles, ram’s heads, and bucrania also occurred. The common occurrence of figural sculpture in the form of busts and full-length statues is described in Chapter 4. Less frequent were painted scenes from Greek mythology and reliefs of deities.14 References to Graeco-Roman gods and myth, as well as reclining scenes, were uncommon in Syria before the Roman period. During the Roman period, early forms of decoration were floral designs, such as palms, rosettes, and garlands. Funerary busts were introduced in Palmyra in the second half of the 1st c. CE, and elsewhere in the 2nd c. CE. In the same century, mythological scenes were painted on hypogeum walls, and the use of decorated stone coffins spread across the province. The trend toward increased embellishment is evident in equal measure in the newly introduced tomb types and the older shapes. The ratio 13

14

Meurdac & Albanèse (1938) report the find of a hypogeum of Hellenistic date with paintings of an individual standing in an aedicula and of garlands with leaves,fruit,and masks at Helalieh, east of Sidon, but provide little evidence in support of its early date. Online Appendix Palmyra 1, #101 (2nd c. CE): Achilles at court of Lycomedes and Abduction of Ganymede; Online Appendix Tyre 1, #1 (75–150 CE): busts with personifications of the winds; Online Appendix Tyre 1, #8 (75–150 CE): Psyche bust, Tantalus, Heracles and Alcestis, abduction of Persephone, Priam and Achilles, and the Sirens. Depictions of Victories were common in Palmyrene tombs. Other depictions of Victories came from the Limestone Plateau. Less common motifs are gorgon or Medusa heads, wreaths, cartouches, baskets, and the hanging curtains typical of the Palmyrene stelae, although these are perhaps also depicted in the hypogeum of Hama (Online Appendix Hama 1, #15).

67

20. Plaster and painted decoration. A: Plastered ceiling in Tomb of ’Abd’astor (Palmyra). B: Painted ceiling in Hypogeum 1 (Djel el-’Amed, Tyre). C: Painted entrance of Hypogeum 1 (al-Awatin, Tyre)

DISCUSSION

and visibility depended on the type of tomb. Decoration inside the communal tombs usually drew attention to the individual deceased, through figural sculpture or elaboration of the individual burial spot, the loculus or coffin. Exterior ornaments focused on the façade and often the area around the door.

Billboards in the Provincial Landscape Despite the continuity with earlier architectural traditions, tombs in Roman Syria demonstrate new attitudes toward funerary architecture. Not only did greater variation exist in types, but also more emphasis was put on the visibility of the tomb, and extensive resources were directed toward its construction. Based on their dimensions, visibility, and decoration, the mausolea, towertombs, tumuli, and funerary enclosures can be considered display tombs. Some of the hypogea reached equally monumental size and lavish embellishment. The widespread occurrence of these features, and the fact that they are found on both new and old tomb types, suggests that the change was structural. In other words, it was not (merely) a change in fashion, but an alteration in the way tomb buildings functioned in society. By the 2nd c. CE, most areas of Syria were filled with visible funerary markers.We have already seen that many tombs were positioned alongside roads. Travelers through provincial Syria could not fail to notice the cemeteries, prominently placed and filled with structures that screamed for attention. Chapter 1 made clear that there was little evidence for spatial differentiation within the cemetery, indicating that people or families were not singled out by the location of their tomb. The cemeteries represented the communal identity of the deceased urban dwellers. This chapter demonstrates an opposite trend. The regional, local, and intra-tomb diversity portrayed little unity among the users of the cemeteries, but rather a heterogeneous community. The inhabitants of Roman Syria sought to distinguish themselves from one another when selecting a type of tomb, although the level and degree of this distinction varied by region, site, and cemetery. This was a feature found both in the cities and in the countryside of the province. DISCUSSION: THE USERS AND MODELS OF THE NEW TOMB TYPES

If the tombs reflected a new concept of funerary architecture, what portion of the population participated in their creation? This question is relevant, as the tombs discussed in this book housed only a tiny fragment of the inhabitants of Roman Syria. In other words, the new tomb shapes may represent exceptions rather than a pattern. Chapter 5 elaborates on this topic and discusses how a portion of the provincial population likely never received a grave that left archaeological traces. Furthermore, not every tomb followed the new trends in

69

70

THE TOMB

architecture. Within the cat. 1 sample, perhaps about a third of the tombs continued to adhere to pre-Roman traditions. Monumental tombs, those that were significantly larger than other tombs, elaborately decorated, conspicuous in the landscape, and often built with expensive materials, constituted around 30% of the cat. 1 assemblage. This percentage is surely an overestimation. Scholars have concentrated on the tall and visible tombs of Roman Syria at the expense of smaller and less visible examples. The sample is thus slanted toward aboveground, inscribed, and tall tombs, particularly in areas where little excavation took place, such as in the Hauran and the Limestone Plateau. Even in the case of better-studied Palmyra, the only cemetery with non-monumental tombs, the North-East Cemetery, was never published. Despite these limitations, the numbers are informative about changes in the size of the participating group. Although perhaps not making up 30% of the assemblage, monumental tombs were not uncommon either. The cat. 2 sample includes at least 250 examples, not counting tomb doors and inscribed lintels that could have adorned large standing tombs. The contrast is stark with the pre-Roman period, which has yielded perhaps only one monumental tomb, found at Hermel. It is also significant that monumental types stood outside cities and villages, and isolated in the countryside, occurring everywhere in Syria with the exception of the steppe. Their construction represents a widespread phenomenon across the province. Urban cemeteries usually contained multiple examples, and the smaller sites often had a single monumental tomb. Both the popularity of monumental tombs and the general increasing emphasis on visibility and adornment followed clear chronological patterns. Initially, only a small collection of tomb-designs departed from earlier forms. The al-Bass Cemetery at Tyre had perhaps only five funerary enclosures in the late 1st and 2nd c. CE. In the region of the Hauran, about twenty-one monumental tombs arose between the 1st centuries BCE and CE, and around thirty-six tower-tombs at Palmyra in the period between 50 BCE and 50 CE. Henning comments that the years 70–80 CE mark the start of a building-boom in tower-tombs at Palmyra.15 Approximately 140 new towers arose here in the period between this date and 128 CE. Rectangular mausolea spread in popularity across the Hauran, the number of funerary enclosures at Tyre rising to thirty-nine in the 2nd and 3rd c. CE. What had started on a small scale had, by the 2nd c. CE, turned into a widespread phenomenon across the province. As we have seen, this is also the period that witnessed the popularity of creating regional styles and adding aboveground portions to the tomb. In other words, starting in the mid-late 1st c. CE, more people were involved in the construction of visible, decorated, and regionally distinct tombs, some of monumental 15

Henning 2003, 98; see also al-As’ad & Schmidt-Colinet 1995, 31.

DISCUSSION

dimensions, others marked by simple stelae. Diversity and regionalism peaked in the 2nd c. CE. The end point of these developments is harder to pinpoint, and varied per region. The 3rd c. CE tombs at Palmyra were smaller than the earlier examples, and elsewhere a decrease in size can be detected in the 4th c. CE. After the 3rd c. CE, no new tombs appear to have been built in the al-Bass Cemetery at Tyre, although older enclosures were still used. On the Limestone Plateau, by contrast, the construction of large tombs continued in the 4th and subsequent centuries, and this was also true for rural areas such as the Orontes Valley and Hauran. Due to the abandonment of several tomb types in the 2nd and 3rd c. CE, the cemeteries became less diverse after the 3rd c. CE, but there is no indication that the visibility and funerary adornment decreased. Pronounced regional styles existed on the Limestone Plateau in the Byzantine period, particularly in the construction of mausolea. Whereas urban cemeteries did not alter significantly, it seems that rural areas witnessed a period of continuity or even increased elaboration of the tomb in the Byzantine period.

Sources of Inspiration: Achaemenid, Hellenistic, and Roman Models Where did Syrians find the inspiration for the new shapes of tomb, the aboveground markers, imported building materials, and forms of decoration? The question of architectural models is complicated but important.Were the sources of inspiration purely local, entrenched in Syrian traditions, or were they related to foreign influences? In this section, I discuss possible models by way of comparison with the funerary assemblage of Iron Age Syria and that of the surrounding territories in and before the Roman period. This analysis is cursory, as it awaits the publication of Syrian funerary materials of the 1st millennium BCE, as well as comprehensive studies of other parts of the Hellenistic and Roman Near East. One can microscopically search the Hellenistic world, the Roman Empire, and Parthian Mesopotamia for stylistic parallels and likely unravel a complex mosaic of influences. It is, however, not the purpose of this book to trace the origin of each new element, even if such an endeavor were possible. Rather, it aims to reconstruct the general cultural milieu in which the creation of the funerary styles of Syria took place. The Achaemenid period (6th–4th c. BCE) produced tombs with stylistic parallels to those in Roman Syria, both in the Persian homeland and in the provincial territories. The so-called Tomb of Hiram found ca. 6 km southeast of Tyre consisted of a large sarcophagus resting on a 3 m high stone platform marking a hypogeum, and possibly dates between 550 and 330 BCE (Figure 21).16 This tomb was still visible when the sarcophagi were hoisted on pedestals and platforms in the nearby al-Bass Cemetery in Roman Tyre. 16

Jidejian 1996, 24–28; Renan 1864–1874, plates 47, 48.

71

72

THE TOMB

A

B

21. Tomb of Hiram (Tyre region). A: Artist’s impression. B: Section of aboveground and belowground portion

DISCUSSION

Elevated sarcophagi can be found in other Achaemenid provinces, such as 5th c. BCE Lycia, and in Persia itself, most famously represented by the Tomb of Cyrus at Pasargadae. Both regions have also yielded hypogea with elaborately sculpted façades, which share features with those on the Limestone Plateau in the Roman period, such as pediments and columns framing the entrance.17 To the Achaemenid era date the aboveground funerary monuments at Amrit in northwestern Syria (Figure 22). Three tombs here, known as Meghazil, consisted of a hypogeum topped by a square pedestal with monolithic markers reaching 7–10 m in height.18 They were constructed between the 6th and mid 4th c. BCE. Although their shape and decoration have little in common with Roman tomb-markers, the idea of adding a monumental and decorated marker to an underground tomb was repeated in the Roman examples. The same was true for the tradition of burial in monumental stone coffins on the Levantine coast. Sarcophagi imported from Egypt and the Greek world were common between the late 6th and mid 4th c. BCE, and could be adorned with figural busts and (mythological) scenes in high relief.19 The Achaemenid period examples mentioned in this section originated out of various regional traditions and represent a diverse collection. None of them were closely replicated in the Roman period. Yet, inspiration was likely drawn from these predecessors, still highly visible in the landscape, in terms of shape, dimensions, aboveground construction, decoration, material, and coffin burial. Tower-shaped mausolea and tumuli perhaps also had local predecessors, but this evidence is circumstantial.Clauss and Will have demonstrated that the form of a tower-shaped mausoleum on a podium, topped by a pyramid-shaped roof, appeared in Phoenician territories in North Africa and the western Mediterranean in the Iron Age. They hypothesize the existence of a tradition of burial in towers in Phoenicia itself, which in turn could have served as inspiration for the tower at Hermel, the Tomb of Sampsigeramos at Homs, and the towershaped mausoleum at Serrin in the Syrian Upper Euphrates region (73 CE).20 The fact that no tower-tombs have been reported in Iron Age Phoenicia makes this thesis uncertain (see also Appendix 2). Tumulus burials raise similar questions, and it is striking that in areas with Roman-period tumuli there also stood much older examples. In the Hauran and neighboring Golan, tumuli date to the Bronze Age, and one Chalcolithic example from Der’a was reused in the 17 18

19

20

For an overview, see Fedak 1990 and Nunn 2001. See also Cormack 2004, 17–27. The shapes of the monoliths were cylindrical with a rounded top and square with a pyramidshaped upper part. A double geometric frieze covered one of the Meghazil, and sculptures of lions were attached to its base (Dunand 1953; Dunand et al. 1954–1955; Elayi & Hakal 1996; Renan 1864–1874, 59–90; Saliby 1989). Tombs in Sidon (Lebanon): Hamdy Bey & Reinach 1892; Jidejian 1971, 121, 131–137; Smith & Ertu˘g 2001, 71–90. Anthropoid coffins in local basalt and imported marble were discovered in Arwad and Arados on the Levantine coast (Elayi & Haykal 1996; Lembke 2001, 6–16). Clauss 2002; Will 1949b. Serrin tomb: Gogräfe 1995.

73

74

THE TOMB

A

B

22. Tomb at Amrit. A: Reconstruction and artist’s impression. B: Section of aboveground and belowground portion

Bronze Age and possibly again in the 1st c. CE.21 In the Euphrates region of Selenkahiye, Bronze Age tumuli dotted the landscape.22 There is no evidence 21 22

Nasrallah 1950. Other tumuli in the Hauran: Sartre-Fauriat 2001, vol. II, 52; 2007b, 239. Porter 2002.

DISCUSSION

for uninterrupted use of these tumuli, and their construction in the Roman period was a phenomenon of the 1st c. BCE or CE. Yet, they perhaps referred to older local traditions. Like the Tomb of Hiram and the Amrit Meghazil, the older tumuli were still prominent in the landscape and possibly held special meaning, as evidenced by the reuse of the Der’a example. The first mausolea of the Hauran copied the circular shapes from the tumuli. This copying of elements of older tombs may have been a conscious choice by the provincial population. In Chapter 6, I come back to this issue when discussing strategies for embedding local traditions in times of social change. Other models for Roman tombs in Syria were those constructed in the surrounding regions. In particular, elite tombs in Anatolia and Palestine displayed similarities in architectural type, decoration, and monumentalization. The most obvious example is the mausoleum, a monumental, rectangular, and heavily decorated tomb type whose construction began in 4th c. BCE Asia Minor and subsequently spread through the Hellenistic world. Its rectangular shape, podium, and architectural and figural decoration on the exterior façade were copied in the mausolea of Roman Syria. The late 2nd and 1st c. BCE mausolea in Jerusalem, with pyramid-shaped roofs, architectural decoration, and sunken courts, also bear similarities with the later Roman examples. These tombs, too, combined underground burial space with an aboveground marker. Sculpted façades covered the rock-cut tombs in Nabataean territory, most famously at Petra between the 1st c. BCE and 1st c. CE. The type of decoration of the Petra tombs bore only limited resemblance to the sculpted façades on the Limestone Plateau, but both the eclectic nature of decoration, borrowing from local, Alexandrian, and possibly Achaemenid repertoires, as well as the use of cliff walls to draw attention to the tomb, were copied in the funerary architecture of Roman Syria. Colledge describes 1st c. BCE tombs of local royalty in Commagene as topped with a double column.23 This feature is reminiscent of the later distyle tombs of the Limestone Plateau. One source of inspiration for the new tombs in Syria, therefore, came in the form of monumental display tombs of the Hellenistic period in the neighboring territories. These tombs belonged to elite members of society, and often local royalty, and perhaps portrayed a dynastic feel that was aspired to by Roman Syrians.24 Another area of inspiration was the Roman world. Stone sarcophagi, for instance, decorated with garlands, bucrania, gorgoneion, and mythological and Bacchic scenes,were produced on a large scale in marble quarries of Asia Minor, Greece, and Rome for wealthy classes of the Roman Empire. The mausolea in Palmyra and the Limestone Plateau followed Roman traditions in sacral 23

24

Jerusalem: Kloner & Zissu 2006. Petra: McKenzie 1990. Commagene: Colledge 1977, 46–47. See also Wagner 2000, 18–23. Cf., de Jong in press.

75

76

THE TOMB

architecture and perhaps copied elements of theater design. Busts were a common phenomenon in the funerary sculpture of the Roman world, and in Syria, they occurred on stelae, closing slabs of loculi, painted medallions, and freestanding sculpture (see Chapter 4). The same was true for depictions of reclining individuals on sarcophagus lids, relief carvings, and stelae. Whereas the origin of these elements lay in different areas of the eastern Mediterranean and Italy, it was their large-scale production and consumption that was typical of Roman elite culture. In this form, the new elements made their way to Syrian tombs as well. We discuss this incorporation of Syrian communities into the globalized networks of the empire in Chapter 6. Here, is it important to note that in addition to earlier local Achaemenid and Hellenistic models, the population of Roman Syria also turned to contemporary styles in sculpture and architecture, consumed by empire-wide elites, for inspiration.

CHAPTER THREE

GIFTS FOR THE DEAD: FUNCTION AND DISTRIBUTION OF GRAVE GOODS

I

n 1961, a farmer from the village of deb’aal discovered a hypogeum while digging a cistern by his house. This hypogeum (T. 29) is one of the few tombs in the region that was left untouched by grave robbers. It yielded a vast array of coffins and grave goods, and provides a unique insight into the customs of placing items with the dead. Figure 23 illustrates the items put in one of the graves of the Deb’aal tomb, a lead sarcophagus placed in a loculus. This chapter analyzes the items that accompanied the dead. Such artifacts were not merely personal belongings given to comfort the dead, or parts of his or her wardrobe. They represent conscious selections of the burying community, and this chapter aims to unravel the meaning behind these choices. Doing so allows us to tie the discussion of grave goods to the themes of this book. What were the main features of mortuary practices in Roman Syria; and what are the patterns of continuity and change? Chapter 2 drew attention to another theme, that of distinctions in the burying community. People were not buried in the same types of tomb or even container. This chapter asks if they were accompanied by similar sets of artifacts. The first section discusses distributions across the main categories of grave goods, as represented in Figure 23: items of personal adornment and vessels, as well as coins and lamps (not pictured in the image). The second section focuses on the patterns of variation and standardization of grave good assemblages, and traces their placement over tomb types, and over time. Concluding this chapter is a discussion of the possible function of the items in the tombs, in practical and 77

78

GIFTS FOR THE DEAD

23. Finds from loculus 1a in the hypogeum at Deb’aal (glass vessels, golden earring, golden necklace, golden ring)

in spiritual terms. In many ways, the grave good assemblages illustrate patterns that diverge from those established in Chapter 2. In the selection of gifts for the dead, Syrian communities kept close to older customs. TYPES OF GRAVE GOODS

Less than a third of the cat. 1 tombs contained artifacts. Most yielded only a handful of items, but this number could vary considerably. One of the fullest graves was a loculus in Tyre, filled with a pair of golden earrings, bracelets of gold and iron, bronze pendants and bells, glass bottles, bronze coins, iron nails, and a terracotta oil lamp, amounting to more than 30 finds (T. 30). The

TYPES OF GRAVE GOODS

items in this tomb reflect the common grave good types across Roman Syria: items of jewelry, clothing fragments, vessels in glass and ceramic, coins, and lamps. Together, the tombs in the sample yielded close to 9000 artifacts (8871, to be precise). This collection is, however, plagued by various sample problems, and cannot be taken at face value. The most problematic issue is the coarse and short descriptions of many of the items, and the absence of illustrations in the excavation reports. At times, all we learn is that the deceased was accompanied by “vessels” or “jewelry,” without any information about the type, material, and number of finds. Rampant grave robbing further impedes the calculation of total numbers. Looters frequently entered (and enter) the tombs to carry off the items reserved for the deceased. Most of the tombs in our sample followed this fate, and we can assume that the amount of grave goods was originally much, much higher. A group of relatively undisturbed tombs provides some insight into these original assemblages. These 46 tombs, excavated at Beirut, Homs, Nawa-tell Umm al-Hauran, Palmyra, and Selenkahiye, yielded 607 finds.1 The in situ context illustrates that averages of around three artifacts per grave were likely, but that the actual numbers ranged from one to thirtytwo per grave. Some deceased, whose graves seemingly were not disturbed, never received goods, at least not any made of materials that preserve in the archaeological record. The dry climate of the Palmyrene desert did preserve some such artifacts, made of textile and basketry. Overall, the assemblages from the undisturbed tombs did not differ significantly from those in other tombs, although they included more rings and golden face covers, and vessels in materials other than glass and pottery, such as bronze and wicker. Such vessels, rings, and golden items were, thus, more likely to have been robbed, decayed, or overlooked. The publications of Syrian tombs often omit to note the physical and chronological connection of the grave goods with the burials. It remains unclear where the artifacts were placed or discovered, and with which burial they should be associated. This latter point is important, since many of the tombs contained multiple burials. The assemblages from the al-Bass Cemetery in Tyre are a case in point. The tombs here yielded the greatest collection of finds in the sample (6526), deposited sometime between the late 1st and the 5th or 6th c. CE. In the absence of a stratigraphical report and full publication of the finds, it is difficult to link these artifacts to individual burials or

1

Online Appendix Beirut 1, #1, #2; Online Appendix Homs 1, #1, #3, #7, #8, #9, #10, #11, #12, #15, #16, #21; Online Appendix Hauran 1, #6, #7, #8, #9, #10, #11, #12, #13, #14, #16, #17, #18, #19, #20, #21, #22, #23, #24; Online Appendix Palmyra 1, #111, #112; Online Appendix Selenkahiye 1, #2, #4, #11, #15, #24, #26, #33, #35, #38, #39, #41, #43, #45.

79

80

GIFTS FOR THE DEAD

1200 1000 800 600 400 200 0 personal vessels (866) adornment (947)

oil lamps (987)

coins (303)

other (306)

chart 6. Distribution of grave goods over main categories

date their placement. It remains uncertain whether an object was deposited in the first construction phase, during the next centuries, or in the Byzantine period. Many of the finds from Tyre may thus belong to a period well out of the chronological boundaries of this book. As a result of the aforementioned limits and untrustworthiness of the data, this book concentrates its discussion of grave goods on those tombs that provide enough reliable information for analysis. The total number of finds was 3409, discovered in 143 tombs.2 This includes 1064 objects from the al-Bass Cemetery that could be dated more precisely to the Roman period. Tombs from all sites and regions yielded artifacts, except those on the Limestone Plateau. Looters most likely emptied these graves. The discussion that follows divides the finds into four functional types: items of personal adornment (jewelry and clothing), vessels, coins, and lamps. A fifth group includes the unusual, one-ofa-kind items. The breakdown in the categories is represented in Chart 6. As with the discussion of tomb types in the previous chapter, there is a degree of artificiality in these divisions. Coins, for instance, could be worn as jewelry, while amulaic pendants may not necessarily have been items of jewelry. In the concluding section of this chapter, we move away from these functional categories to look at how, as sets and as individual artifacts, grave goods played a role in the funerary ritual. 2

This number is based on counting collections of beads as a single find, i.e., belonging to a necklace. Collections of iron nails are generally not counted, as they most likely come from wooden sarcophagi. Other elements of wooden coffins, such as hinges and decorative plaques, are also not included. Single iron nails are added to the total number of grave goods, and those belonging to shoes are counted as a single find.

TYPES OF GRAVE GOODS

24. Items of personal adornment. A: Gold applique (4 × 3 cm) from Pit 6 (Baalbek-Douris). B: Golden earring from S. 3881 (Complex XXIV, al-Bass Cemetery, Tyre). C: Golden earring from M. 4202-l.2 (Complex XXX, al-Bass Cemetery, Tyre). D: Gold leaf of wreath (3.7 × 2.5 cm) from Pit 6 (Baalbek-Douris). E: Golden bracelet with turquoise inset (2.1 cm high) from Tomb 1 (Homs). F: Golden face-mask (19 × 18 cm) from Tomb 1 (Homs)

Personal Adornment Earrings in bronze, silver, and gold, and necklaces with beads of various sorts of stone and metal, represent the most regular finds in the graves (Figures 23 and 24). Together with other items of jewelry and pieces of garments, such as

81

82

GIFTS FOR THE DEAD

table 2. Items of personal adornment earring ring bracelet necklace pendant bell pin amulet golden face-cover gold leaves (set)

229 125 83 115 56 40 76 20 19 12

golden mask golden wreath golden headband textile shoe (single) textile appliques (set) belt buckle button other jewelry Total

2 4 2 21 19 5 14 4 101 947

belt buckles and shoes, they form a common category of grave goods (Table 2). When their find location is recorded, they were found on the body, indicating that the deceased wore the jewelry and clothing. The exception was an engraved gold ring found in a wooden box placed next to the deceased in a pit-grave at Nawa-tell Umm al-Hauran (T. 31). Within the category of personal adornment, earrings were most common. Their proportional number dropped over the centuries. Necklaces and pendants made up the next group, and included bronze and silver bells, which may have been part of a necklace or garment. One example from Tyre was still attached to the silver chain of a necklace. Rings made up 13% of the assemblage, and their number increased after the 2nd c. CE. Some artifacts were identified as amulets or talismans, although the basis for this identification is not always clear. They took the shape of human and animal figurines, bronze inscribed sheets, a stone miniature of Bes, and a miniature glass amphora. Occasionally, sheets of gold covered the eyes and mouth of the deceased. Complete face covers in the form of masks in gold and silver were most common in the 1st c. CE, and particularly in the western part of the province of Syria. Examples come from Baalbek, Beirut, Homs, Deb’aal, and perhaps Tyre. The discovery of gold sheets in tombs at Dura Europos and of a face-mask at Halabiye on the Syrian Euphrates points to similar practices farther east.3 The dry conditions of Palmyra allowed for the preservation of textiles, and several mummies wrapped in linen, wool, cotton, and silk originated from the tombs. Two buckles from a hypogeum in Hama still had the remains of leather and linen attached. Items of personal adornment became less common over time. The number of tombs containing such items was higher in the 1st c. BCE and CE (around 80%) and dropped in subsequent centuries (around 30%). In the later centuries, items of personal adornment also constituted a smaller portion of the entire grave goods assemblage per tomb. In general, we can say little about 3

Online Appendix Dura Europos 1, 2. Halabiye: Curtis 1995, 229.

TYPES OF GRAVE GOODS

distributions across tomb types, as few significant patterns emerged. Tombs in the Euphrates region (Selenkahiye, Dura Europos) and the rural south (Hauran) included items of personal adornment less often than tombs in other regions. Particularly rich and unusual collections came from the pit-graves in Homs and Nawa-tell Umm al-Hauran, where we find, among other things, jewelry with turquoise inlays and silver belt buckles.4 Scholars generally interpret items of personal adornment as having an ornamental purpose, decorating the body of the deceased and being worn at the time of burial. It is possible that they held additional, amulaic or magical, functions. In the Near Eastern, Egyptian, Greek, and Roman worlds, precious stones held various magical powers, and the same is true for precious metals. Lead had negative associations, but was also popular for its apotropaic characteristics, as indicated by lead amulets. Amulets served to protect a person from evil and to ensure good health and prosperity. Garments or necklaces tied with bronze bells could have had the purpose of protecting the wearer from evil spirits that fear the sound of metal.5 The occurrence of masks and face covers in the eastern Mediterranean, Near East, and Egypt has prompted various interpretations, ranging from conservation to protection to solar symbolism.6 In more general terms, they signal the importance of covering the face, particularly the orifices. Very little is known about magic or the use of apotropaic artifacts in Roman Syria, or about the influence of Mesopotamian, Egyptian, Greek, and Roman practices. However, a large portion of the items of personal adornment in Syrian graves possibly carried with them magical properties, through either material (precious metal and stone) or function (bell and amulet). We come back to the possible function of these and other items in the tomb in the final sections of this chapter.

Vessels A glass or ceramic vessel accompanied many of the deceased. They amount to about a fifth of the total assemblage and, before the 2nd or 3rd c. CE, were found in 80% of all tombs with finds. Afterward, the number of tombs with vessels, as well as the number of vessels per tomb, fluctuated. The assemblage included comparable quantities of pottery (437) and glass (392) items, but the chronological distributions of these materials were not the same. In the 1st c. BCE, glass vessels were rare, and a century later, they made up more than half of the assemblage. By the 3rd and 4th c. CE, glass had almost entirely replaced 4 5

6

Online Appendix Hauran 1, #6–24; Online Appendix Homs 1, #3, 7–12, 15–16. Oettel 2000, 113–114. For a wider discussion of metals, precious stones, and sound, see Luck 2006, 49, 218–220. Curtis 1976, 1995; Fick 1999; Seyrig 1952b; Theodossiev 1998.

83

GIFTS FOR THE DEAD

250 200 150 100 glass 50

pottery

unknown shape

water pot/basin

cooking pot

incense burner

jar

pithos

bowl

amphora

jug

plate

pitcher

cup

goblet

small jar

small vessel

flask

amphoriskos

bottle

0 unguentarium

84

chart 7. Vessels

pottery. The popularity of glass is likely directly related to the invention of the technique of glass-blowing on the Levantine coast in the middle of the 1st c. BCE, perhaps in Roman Syria.7 Glass was now produced on a larger scale and was more widely available. Burial grounds close to the possible production sites – Beirut, Tyre – yielded particularly high quantities of glass in relation to pottery. Those farther from the coast, such as at Dura Europos and Palmyra, continued to contain more ceramic vessels. Occasionally, vessels in other materials, such as metal, wood, bone, or basketry, were discovered in the tombs (37 in total).8 Chart 7 distinguishes the vessels by functional type (e.g., cups, amphorae, bottles) and material. As a result of the inconsistent terminologies used in the excavation reports and the absence of drawings and images, this categorization remains crude and includes a high number of unknown shapes. It was not possible to standardize the entries in the database according to a single typology. Nevertheless, clear patterns emerge when we look at the types. The majority of the ceramic vessels consisted of small shapes, such as flasks and unguentaria (Figures 23 and 25). In the larger range, pitchers and amphorae were popular. Typical for Palmyrene tombs were plastered (water) pots and incense burners, the latter often made from the base of a broken vessel. In the single tomb types, the pit- and cist-burials, pottery vessels were placed by the body. In the communal tombs, vessels more often stood in front of the loculi, on the floor,

7 8

Butcher 2003, 201; Jennings & Abdallah 2001, 237. Lead (2), bronze/copper (12), silver (2), wood (2), faience (1), bone (1), wicker (2), stone (1), bronze incense burner (1), incense burner unknown material (4), vessel unknown material (9).

TYPES OF GRAVE GOODS

25. Vessels. A: Two glass unguentaria from M. 837–l.12 (Complex XII, al-Bass Cemetery, Tyre. B: Glass vessel from M.4888–l.2 (Complex XXXIII, al-Bass Cemetery, Tyre). C: Glass jar from Tomb GXV (Hama). D: Pottery from pit-graves (Shahb¯a). E: Pottery bottle from Tomb 19 (Apamea)

or on benches in the central area of the tomb. Clear distinctions thus emerge, relative to the type of tomb. Compared to the pottery, the glass assemblage covered a smaller range of functional types. The majority consisted of small vessels designed to hold scented liquids, creams, and make-up. The unguentarium or long-necked bottle was the preferred shape. Glass bottles often stood by the feet of the deceased and, less frequently, close to other parts of the body. Although some were found in the central areas of communal tombs, it is not certain whether this was their

85

86

GIFTS FOR THE DEAD

original placement or the result of later disturbances. In Palmyra, glass cups were discovered in a plastered basin in the main room of a hypogeum (T. 32). No evidence exists elsewhere for the intentional placement of glass outside the burial spots, i.e., outside the loculus or coffin in the communal tomb. With regards to placement and type-range, therefore, differences existed between pottery and glass vessels. Occasionally, graves yielded vessels in other materials: bronze, silver, lead, faience wood, bone, and basketry. Single tombs, and in particular pit-graves, held the highest numbers of such vessels. Tombs constructed after the 2nd c. CE no longer appear to have contained such items. In fact, we can detect a degree of standardization in the vessel assemblage, where small glass shapes became the dominant vessel type. Tombs with unusual assemblages, such as the storage vessels in a hypogeum at Djel el-’Amed (T. 33) and cooking pits, tablewares, and storage amphorae in a pit-grave in Shahb¯a (T. 34), were early in date. In the 2nd or 3rd c. CE, the range of materials and types of vessel in the Roman tombs decreased. The discussion of the purpose of vessels in tombs varies for each functional category. The largest group from the assemblage can be characterized as containers of valuable liquids and unguents, such as perfumes or scented oils, creams, and make-up. This identification is primarily based on the shape: small and closed forms, such as bottles and unguentaria. Residue analyses of similar vessels found outside Syria confirm that they held oils, moisturizing creams or skin balms, white and reddish facial powders, ochre, and compounds associated with myrrh and incense.9 The contents of these vessels can therefore be interpreted as luxury substances and personal cosmetics. They may also have been related to ritual activities, as perfumed oils played a role in the anointment of the body that occurred after death. The reasons for the presence of these items in a grave are rarely investigated. Adornment and embalmment of the body were directed at preservation, cosmetic beautification, purification, and possibly pacification of the spirit of the deceased.10 Perhaps the bottles in the graves were the remnants of these activities. Perfumes could also be offered as part of the funeral, as discussed later. A small collection of vessels, around 11%, was related to food preparation, serving, and dining. Such vessels were usually ceramic, and were more popular

9

10

See, for instance, Anderson-Stojanovi´c 1987; Khalil 2001; Pérez-Arantegui et al. 1996; Ribechini et al. 2008; Sachet 2010; Welcomme et al. 2006. See articles in Dudley & Rowell 1993.See also Green 2008,159,163.The connection between precious perfumed liquids in small vessels and the practice of anointment was also mentioned in the New Testament (Mark 14:3–9). See for instance Dalley 1993, 29; Fappas 2011; and Lucian On Mourning (Luc. Luct.). Anointment was also connected to healing practices and offerings. Luck (2006, 218) notes that pleasant scents were thought to have scared of evil spirits. For Scheid (1984, 120), they masked the smell of the corruptible dead.

TYPES OF GRAVE GOODS

in communal tombs. Scholars connect grave goods associated with food to two ritual practices: banquets and offerings. Tomb-side banquets are well known from Roman Italy and occurred during the funeral and yearly commemorative festivals. As discussed in Chapter 5, however, there is little direct evidence for funerary banquets inside or next to tombs. The second interpretation was perhaps more current. The vessels could hold offerings to the deceased and to divinities associated with death. Offers of sacrificial meat, vegetables, and fruit, as well as libations of milk, honey, perfumes, water, and wine, are attested in literary and archaeological sources across the Roman and Greek world. Incense could also function as an offering, and the burning of incense was connected to religious ceremonies. Here, we have sources closer to home, mentioning the burning of incense at funerals in the Levant. Some authors point to the fact that incense would clear the stench of decomposition more effectively than perfume in closed bottles, and that smoke may have had magical properties.11 Syrian tombs yielded several incense burners.

Lamps Terracotta oil lamps constituted the largest group of grave goods in terms of total numbers,but came only from a relatively small group of tombs (Figure 26). Some contained large amounts, such as a hypogeum at Hama with 123 lamps (T. 35) and the Tomb of Yarhai in Palmyra with 259 examples (T. 36). In fact, more than half of the lamps in the assemblage originated from only four tombs. This type of grave good was,therefore,far less common than vessels and jewelry. The number of tombs with lamps remained relatively stable until the 4th c. CE, when it appears to have declined. The quantity of lamps per tomb grew over time, reaching its peak in the 3rd c. CE. Most lamps were ceramic; one bronze example originated from Nawa-tell Umm al-Hauran, and three made out of glass from graves in Tyre. The production of glass lamps was possibly a late (4th c. CE) development.12 Some lamps were placed in single tombs, but the majority originated from various locations in the communal tombs, predominantly from hypogea. Many stood inside a loculus, whereas others were discovered in the central areas of the tomb, such as the main chamber or side-niches (cf., Figure 37). Good evidence comes from Palmyrene hypogea, where lamps were placed on the floors, platforms, inside a water vessel, in front of loculi, on top of pit-graves in the floor, in front of funerary sculpture, and in used and unused loculi. The function of oil lamps in tombs is understudied, rather surprisingly given their preponderance. It is often assumed that they were used to light the dark 11 12

Luck 2006, 480–481. See also Green 2008, 165 and Levison 2002. O’Hea 1993.

87

88

GIFTS FOR THE DEAD

26. Terracotta lamps from Tomb C (Palmyra)

tombs or symbolically referred to darkness.13 This leaves unanswered why they were left behind in the tomb. The evidence from Roman Syria does not confirm a common interpretation that the lamps lit funerary meals that took place in the tombs. Or rather, those tombs with pottery related to banquets did not necessarily also contain lamps. The placement of lamps in the communal areas 13

Scheid (1984) has connected the use of lamps during a Roman funeral to reference to the night, an inversion of reality that happens at the time of death and funeral. In his argument, much of the funerary ritual reverses regular practices, and lights refer to the night, the opposite of the day, and place the mourners in an opposite place when compared to regular people.

TYPES OF GRAVE GOODS

of the larger tombs does suggest that they functioned as part of the tomb furniture, to be lit when a funeral or other activities took place. Yet only a portion was placed in these areas, and many originated from locations that were barely accessible after burial had taken place. One would have to dig out the pit-grave or jar-burial, or remove the closing slabs of sarcophagi and loculi, to retrieve the oil lamp for a later funeral.

Coins Coins were fairly uncommon in the early Roman assemblages. Their numbers jumped in the 2nd or 3rd CE, with a heavy concentration in the 3rd c. CE. This chronological distribution primarily reflects patterns in the al-Bass Cemetery in Tyre, which yielded more than half (59%) of all coins. Without the Tyrian tomb, coin finds were more equally distributed over the Roman centuries. They appear to have been less common in single tombs such as cists and pits, and none of the jar-burials yielded examples from this find category.Most coins were bronze issues; gold, silver, billon, and gilded bronze examples were rare.14 The mints, although rarely recorded, included local production centers such as Tyre and Palmyra, as well as imperial mints. The non-bronze coins stem mostly from Tyrian tombs, but silver may have been popular more extensively on the Levantine coast. Very little information is available about the find spots of coins. A few examples were found in or near the skull; two other coins were discovered close to the chest and in or near the hand of the deceased. A collection of thirteen examples lay between the legs of an individual in a tomb at Selenkahiye (T. 37). The dominant interpretation in Classical studies links the placement of coins in tombs, particularly those put in the mouth of the deceased, to a belief in Charon’s crossing and Greek mythology of the underworld.Stevens has pointed out that coins in graves should be placed in a wider interpretative framework than just influence from Greek beliefs, and the Syrian examples support this statement. Only a few were found in the mouth of the deceased, and examples from Achaemenid tombs in Kamid el-Loz in the Beqa’ Valley indicate that the placement of coins on the body was part of an older tradition. Stevens ties the deposition of coins in tombs in the eastern and western Mediterranean to the religious-magical significance of coins, based on the intrinsic value and potency of money.15 As we have already seen in the section on jewelry, metal may also have held magical qualities. In Roman Syria, furthermore, some coins were worn as jewelry, as evidenced by suspension holes. 14 15

Bronze: 192; silver: 33; gold: 13; billon: 11. The material of the remainder was not published. Stevens 1991. Van Andringa et al. (2013, 923) interpreted coins in tombs in Pompeii as helping fix the dead in the next world. Kamid el-Loz: Oettel 2000, 110. More examples from the Achaemenid period are listed in Nunn 2001, 402.

89

90

GIFTS FOR THE DEAD

table 3. Miscellaneous finds spindle whorl/loomweight bronze/copper mirror bronze medical tools bronze ladle bronze spatula bronze needle spoon iron strigiles iron knife iron scissors iron arrow head lance umbo sword bronze dagger bronze helmet silver helmet cuirass mortar/pestle wetstone lead curse tablets/envelopes bronze key

22 6 6 1 2 1 5 1 4 1 2 4 2 3 3 2 1 1 3 1 4 1

bronze weight bronze prism bronze tripod bronze musical instrument sculpture terracotta figurine bone comb ivory fish flute nail bronze object iron object lead object silver object bone object glass object wooden object stone object shell pottery object unknown material Total

1 1 1 1 1 5 3 9 2 33 46 42 9 1 7 23 2 17 12 2 12 306

Other Finds The remainder of the grave goods in the Syrian tombs consisted of unidentifiable objects, such as fragments of bronze or glass, and objects that were rare or even unique. In this latter group, several functional categories can be distinguished (Table 3, Figure 27). Spinning implements included spindle whorls and loom weights in a variety of materials. Other tools, such as spoons and spatula, are listed in Table 3, many of which came from pit-graves in the Hauran. The same tombs, together with the pit-graves from Homs, also yielded most examples of armor and weaponry. Finds in this category date to the 1st and/or 2nd c. CE. A particularly spectacular example comes in the form of a silver helmet with a face-mask attached, weighing over 2 kg (T. 38). The helmets found in these tombs, as well as some of the other bronze parts, were likely part of military parade gear (see p. 134). The next group can be considered magic in nature. Four folded lead envelopes were found in Beirut,and a possible fifth originated from Tyre.These examples have not been opened, but possibly served as curse tablets. Similar envelopes and tablets inscribed with curses and prayers come from several sites in the Levant. According to Heintz, they were buried in the grave of a person who had experienced an untimely or violent death and whose ghost was roaming without finding peace. These spirits could perform the task inscribed

TYPES OF GRAVE GOODS

91

B

A

27. Miscellaneous finds. A: Silver/iron helmet and mask from Tomb 1 (Homs). B: Bronze helmet from Tomb 1 (Naw¯a-tell Umm el-Haur¯an)

on the tablet. Curse tablets were often excavated in graves close to a circus, possibly placed there to condemn the racers and their horses on the racetrack.16 Both the cemeteries of Beirut and Tyre lay in the vicinity of a circus. Metal nails may also have carried magic properties. Usually, nails were remnants of wooden coffins or shoes, but occasionally, they occurred alone or in a pair. In this light, Walbank’s remark on the use of nails in graves in Roman Corinth is interesting. She comments that nails could have held magical significance and that the presence of an iron nail in 2nd c. CE graves was regarded as a means of protection against the supernatural. This practice was also attested in earlier times in Greece.17 The fact that six tombs with single nails were found at the military colony of Jebel Khalid may provide evidence for a particularly Greek practice in Hellenistic Syria.18 Figurines or miniature sculpture form another group, and shells were placed in graves at Si’, Hama, and Palmyra. Pit-graves,hypogea,and particularly funerary enclosures included more finds from this category than the other tomb types.Enclosures and pits also contained a higher number of these finds relative to other finds, resulting in more idiosyncratic assemblages. Several sites had unique finds. The funerary enclosures in Tyre, for instance, included two small bronze prisms, a bronze key, weights, an eggshell encased in lead sheet, a mortar and pestle, and a glass loom weight. 16

17 18

Heintz 1998, 337–342. See also Luck 2006, 48–49; Wilburn 2012, 238. On the discovery of a curse tablet from the area of the circus at Beirut, see Jidejian 1993, 108–111. Bodel (2001, 23) points out that lead was used because of the negative connotations of the properties of this material. Corinth: Walbank 2002, 277. Greece: Kurtz & Boardman 1971, 216. Littleton & Frohlich 2002; Jackson & Littleton 2002.

92

GIFTS FOR THE DEAD

The pit- and cist-graves at Nawa-tell Umm al-Hauran held spoons, medical sets, weapons, a set of strigiles, ivory fish with Latin numbers and Greek letters, and possible fragments of a musical instrument. The cat. 2 tombs of Dura Europos, of possible Roman date, yielded iron arrowheads, decorated bone plates, bronze mirrors, and bone combs. GRAVE GOOD ASSEMBLAGES IN SPACE AND TIME

The distribution of artifacts in the tombs of Roman Syria shows that people were not buried with similar sets of grave goods. A relatively undisturbed funerary enclosure in Beirut illustrates this point neatly. This tomb (T. 39) contained eleven niches, each with multiple compartments stacked on top of one another. Of these, twenty-three were excavated, and sixteen were empty of grave goods upon discovery. Several showed signs of later disturbance and may have been looted. However, the closing slabs of at least seven of the empty compartments were still in their original location. People in these graves were likely intentionally buried without accompanying artifacts. The remaining six compartments contained two or three finds on average (range of one to six). The collection consisted of items of jewelry (seven), glass vessels (four), and coins (six). One grave held the skeleton of a young adult with a glass unguentarium placed at the feet. The others contained more than one individual, and here again variation appears. Three adults were buried in a single compartment, but grave goods accompanied only one of them: three coins and glass vessels. A particularly full grave with nine individuals placed on top of one another yielded only three earrings and the same amount of coins. Highly valuable finds come from one primary adult burial, with a gold ring discovered by the right hand and a golden face-mask near skull. In sum, in this tomb complex in Beirut, people were buried with or without grave goods, and when they did receive gifts, variation existed in their number, type, and placement. Our data suggest that this pattern repeats itself in the other tombs of the province. Certain artifacts were often found together, such as vessels and coins. Items of personal adornment often appeared with vessels and coins or unusual artifacts. The frequency of these combinations, however, was only slightly higher than that of other combinations. In other words, strong overarching trends in combinations of grave good types were absent. People in Roman Syria never received homogeneous sets of objects.

Distinctions in the Burying Communities The variation in grave good assemblages, however, was not random, and it is possible to detect central principles guiding the selection of grave goods. This becomes most visible when we move from the level of individual graves to

GRAVE GOOD ASSEMBLAGES IN SPACE AND TIME

distributions across time and space. There was a slight increase in the number of tombs containing finds in the 1st c. CE, and this trend continued in the 2nd–3rd c. CE. The average of eight objects per burial spot in the 1st c. BCE– 1st c. CE, on the other hand, decreased to around three in the later periods. In other words, over time, more graves contained finds, but they contained fewer of them. The variability in quantities of artifacts per grave possibly also declined, as indicated by the fact that before the 2nd or 3rd c. CE, the numbers of finds per burial spot varied more than after this period, when the majority included between one and four grave goods. The earliest tombs contained few coins and glass vessels, and high quantities of items of personal adornment and pottery. In the 1st c. CE, glass vessels and lamps grew in numbers. A century later, the number of lamps and glass vessels had further increased, as had the quantity of coins. Tombs of the 2nd c. CE yielded the greatest variation in artifacts. Afterward, the supremacy of glass over other vessel materials and the diminishing of vessel shapes led to a reduction in diversity. These distributions tell us a number of things. Over time, more burials contained finds, and these assemblages were smaller in size and more similar in range. This standardization may have been a feature in particular of urban cemeteries. Rural tombs generally displayed greater variation in artifact collections. Regional trends are apparent when comparing the assemblages in the tombs on the Levantine coast and those from inland sites. We have already mentioned that glass was predominant in the coastal areas, and reached the hinterland more slowly. Typical for the Levantine coast were (silver) coins, figurines, and perhaps lead curse tablets. Away from the coast, the assemblages become irregular and diverse, as can be seen in the collections of Hama, Homs, the Hauran, and Palmyra. Otherwise, few regional distinctions emerge from the sample, and the types of grave goods are fairly similar across the Syrian province. This is not to say that within the categories strong regional preferences are apparent, e.g., with certain types of hooped earrings or long-neck bottles restricted to particular regions. The incomplete nature of the data prohibits such analyses at this moment. It remains unclear whether the pronounced regionalism characterizing funerary architecture of the first centuries CE, discussed in the previous chapter, also involved the gifts for the dead. Certain tomb types yielded high-number finds: cist-graves, funerary enclosures, jar-burials, tower-tombs with hypogea, and tumuli. Again, it is difficult to assess the significance of these patterns. Lacking inscriptions, tumuli, cist-graves, and jar-burials were dated primarily on the basis of grave goods, meaning that those with artifacts were more likely to be included in the cat. 1 assemblage. The funerary enclosures in Tyre were used for reburial on a greater scale than tombs elsewhere, potentially increasing the number of finds with each burial. All tomb types included artifacts from each category,but there are some distinctions. The most common types, hypogea and pit-graves, included relatively few

93

94

GIFTS FOR THE DEAD

items of personal adornment, and more vessels and artifacts from the “other” category. This vessel collection also included a wider range of shapes. Funerary enclosures stand out in their high quantities of coins, vessels, and unusual artifacts. Perhaps we can conclude that people buried in the less common tomb shapes, and in particular the large tombs, received richer and more standardized artifact collections. However, it cannot be ascertained whether this represents a true pattern or reflects excavation and publication biases inherent in the sample. More robust patterns arise from the comparison between single and communal tombs. We have already seen in the previous chapter that spatial concepts and levels of monumentality and elaboration varied between the two types. Distinctive practices with regards to grave goods indicate that this was not only a matter of architecture. The single tombs contained on average five to six artifacts. Compared to communal tombs, they held fewer of the common objects (personal adornment, coins, and lamps) and more unique items and vessels, especially those made of ceramic and other materials. Grave goods were placed with the burial inside the grave, with the exception of two pitgraves at Nawa-tell Umm al-Hauran, where objects were positioned inside the pit on top of the basalt slabs closing the smaller burial pit (T. 40, 41). The loculi, pits, or coffins in communal tombs included 6.2 artifacts per burial spot. This number is distorted by the full graves in the al-Bass Cemetery of Tyre. Without the Tyrian tombs, the average in communal graves was 4.3 grave goods per burial spot. Communal tombs often had finds from each category, and the assemblages appeared less idiosyncratic than in single tombs. More items in precious metals and stones originated from the communal tombs. The placement of grave goods, in particular, points to an important distinction between the two tomb types. Artifacts in communal tombs were either located in a burial spot, such as a loculus or sarcophagus, or placed in the central areas, in front of a loculus, by the entrance, or in side rooms. Lamps were most frequently found outside the grave, which one would expect, due to their function. Yet vessels, jewelry, coins, and other finds also originated from find locations outside the burial spots. Undisturbed tombs in Palmyra suggests that this positioning was intentional and not (always) the result of later disturbances. The placement of finds in communal tomb suggests that they played more varied roles in mortuary practices when compared to the grave goods in single tombs, which almost always accompanied the body. We will return to this issue in the discussion of ritual practices. Inside the communal tombs, possible distinctions emerge between modes of burial. The loculi of the al-Bass Cemetery contained fewer items of personal adornment and more lamps and pottery vessels compared to the sarcophagus burials in the same funerary enclosures. In Beirut (Bey 104), loculi that were filled with terracotta coffins held the highest quantity of finds per buried

GRAVE GOOD ASSEMBLAGES IN SPACE AND TIME

individual, and on average more jewelry. The same pattern emerged from the Deb’aal hypogeum, where the sarcophagi placed in the loculi contained aboveaverage quantities, particularly of jewelry and coins (T. 29, Figure 23). If one can take these patterns as representative for the province, or at least the Levant, it follows that within communal tombs, sarcophagus burials yielded the highest numbers of artifacts, particularly items of jewelry and, perhaps, coins.

Pre-Roman Customs The patterns just described appear to have strayed little from older practices. Pre-Roman tombs yielded similar finds and find categories. They point to high degrees of continuity in the principles guiding the selection of gifts to accompany the dead (Table 4). Hellenistic and Parthian graves usually contained between two and four finds. The placement of these items by the body or in the central room of communal tombs was similar to practices of the subsequent centuries. Coins and lamps were uncommon in Hellenistic tombs, but coin finds increased in the Parthian period. Examples come from eastern Syria and Mesopotamia.19 The most pronounced difference was the lack of glass artifacts. As indicated, this was probably related to the fact that a new technology developed in the 1st c. BCE increased the availability of glass in the Roman period. Later (1st–2nd c. CE), Parthian tombs at Dura Europos and Tell Sheikh Hamad did include vessels in blown glass, indicating that the popularity of this material crossed the imperial borders.In western Syria,small bottles designed to hold valuable liquids also dominated the earlier pottery shapes, indicating that the introduction of glass here signaled not a functional change, but a change in material. Kitchen and storage wares were found more frequently at inland sites such as Dura Europos and Palmyra. As we have seen, this trend possibly continued in the Roman period, until the mid–late 3rd c. CE. Gold, a frequent occurrence in the Roman tombs, was almost completely absent from the Hellenistic assemblage. The Parthian tombs of the 1st and 2nd c. CE did include gold beads, pendants, face covers, and earrings. Starting in the 1st c. CE, therefore, gold turned up in tombs more often, both in the Syrian province and in Parthian territory. With the exception of one bead in a tomb in Tell Kazel, no lead objects were found in the Hellenistic and Parthian tombs, and the same was perhaps true for ivory. Parthian graves included a wide range of vessel materials: basketry, bronze, silver, wood, and bone and numerous pre-Roman tombs yielded alabaster vessels. Thus far, Roman graves have not yielded alabaster vessels. 19

Abu Qubur: Gasche 1996; Gasche et al. 1989, 1991, 5–6; Warburton 1989, 14. Nimrud: Oates & Oates 1958. Nineveh: Curtis 1976. Seleucia-on-the-Tigris: Hopkins 1972; Valtz 1986; Yeivin 1933. Sheikh Hamad: Novák 2000.

95

table 4. Finds in Hellenistic-Parthian tombs Beirut (59) earring (209) bracelet (70) ring (160) pin (35) amulet (2) necklace (112) pendant (70) bells (17) gold leaves (17) other jewelry (16) textile appliques (3) textile (27) belt buckle (9) button (18) glass small bottle (70) glass goblet (2) glass plate (2) glass pitcher (2) glass bowl (2) glass unknown shape (3) pottery small bottle (43) pottery plate (8) pottery cup (5) pottery jug/jar (46) pottery pitcher (18) pottery bowl (28) pottery amphora (30) pottery pithos (2) pottery unknown shape (44) bronze/copper vessel (4) silver vessel (2) bone vessel (2) alabaster vessel (15) basketry (4) bronze coin (7) silver coin (8) coin unknown material (8) oillamp (29) bronze lamp (1) spindle implements (8) spoon (2) iron knife (9) bronze spatula (20) iron arrow head (10) weapon (13) iron dagger (2) tc figurine (5) bronze/copper mirror (21) single nail (11) unknown material/type (168)

Dura Europos (615)

Jebel Khalid (38)

Jebleh (26)

2 1 2 1

39 18 122 7

3 1 1 1

2

4 1

32 41 33 17 9 3 21 4 10 58 1

3

2

2

4 2 1 1

8

2

12

32 15 11 16

Palmyra (147) 30 11 7 1 1 9 2 2

Tell Kazel (48)

Tell Sheikh Hamad (531)

2 7 2 4

131 32 26 21 1 56 26 27

6

2

3

2 2

4 3 8 10

2 1

4

1 1 2 3

2 1

2 1 1 10 6 2 9 7

3

28 4 2 1 9 2

1 7

3 2

4 1 5 2

4 9 2 8

1 4 1

7 2 8

2 4 1

2

10

6 11

1 1

1 1 2

1 10

4 1 2 1 16

1 17 29

1 3 6 15

8

5

6 1 6 10 6 13 1 1 1 4 95

DISCUSSION

These patterns tell us that the grave good assemblages remained fairly similar to those in the Roman area between the Hellenistic and Parthian periods, although the latter yielded more diverse materials. This observation stands in sharp contrast with the developments in funerary architecture discussed in Chapter 2. The introduction of new tomb shapes and an increasing elaboration of the funerary space, often visible to the public, resulted in cemeteries that looked profoundly different from their predecessors and from burial grounds across the border in Parthian territory. By the 2nd c. CE, a whole range of new shapes and decorative forms filled the Syrian cemeteries. Yet, how people were buried, and the items that were chosen to accompany the deceased, continued to adhere to long-standing traditions. The changes that did occur, such as the introduction of glass and the increase in popularity of gold items, were not restricted to Roman territory but extended into the Parthian lands. Concepts of space transformed cemeteries of the province, yet the functions of grave goods, in magic practices, embalming, and offerings, crossed the imperial boundaries. It is to these roles that we turn next. DISCUSSION: GIFT EXCHANGES WITH THE DEAD

Parker Pearson has pointed out that grave goods were not simply personal trappings, but items bound up in gift exchanges with the dead.20 In this concluding section, I discuss these exchanges and the possible reasons for placement of these artifacts in the tombs. Many items in the assemblage were valuable, either made of precious and rare materials or holding expensive content. Small numbers of luxury items, thus, accompanied the dead. Occasionally, imported items entered the graves. The glassware was probably produced on the Lebanese coast and exported to the hinterland. Pottery vessels and lamps, when recorded, were produced locally, on in nearby regions such as Eastern Turkey and the Southern Levant. Incense and silk in the Palmyrene tombs were foreign materials, as were the precious metals and stones throughout the province. Items in these materials did not seem to have been solely produced for the funerary context. Jewelry and garments could be worn during life, and incense was burnt during non-funerary rituals. The same held true for the other grave goods: the vessels, coins, and lamps could have been used before ending up in the grave. It is, of course, possible that within the categories certain items were reserved for funerary purposes, e.g., a particular shape of glass vessel for the anointment of the dead or a distinct iconography for funerary lamps. Data for such a study are currently not available. At this stage, we can only identify three groups of artifacts that were (presumably) made exclusively for placement in a tomb, each of which was rare: face covers, shrouds, and curse tablets. This last group 20

Parker Pearson 2008 [1999], 85.

97

98

GIFTS FOR THE DEAD

should probably not be considered grave goods per se, as they could be placed in the graves by anyone, not just individuals involved in the funeral. The lead envelopes found in a grave in Beirut were 15 cm above floor level, indicating that they were placed in the fill after the last burial took place (T. 42). Thus, save these three exceptions, there was no specialized market for the production of grave goods. This stands in sharp contrast, again, with the tomb architecture discussed in Chapter 2, where a whole industry was dedicated to the production of funerary materials such as stelae, portraits, and coffins.

On the Function of Grave Goods Why were objects placed in the tomb? Studies of grave goods often describe them in purely functional terms: jewelry adorned the body, perfumes hid the stench of rotting human remains, and lamps lit the dark, underground spaces. Yet, the conscious selection and deliberate placement of these finds in tombs begs for more in-depth explanations. Why would lamps be placed in single tombs and inside burial spots where they would not be accessible or of use to light the tomb? Perfumes in a bottle with a long narrow neck probably did little to keep away unwanted smells. Why leave these items behind in the tomb after the funeral? The burial rituals to which the grave goods pertained are discussed in Chapter 5, in conjunction with information from the architecture and epigraphy. This section concentrates on the artifacts and reconstructs the possible reasons for their placement in the tomb. In doing so, it places question marks after commonly heard but rarely proven assessments of the role of grave goods. The analysis starts by addressing the exact location of grave goods in the tomb,and unfortunately this contextual information is not abundant.The grave goods appear to concentrate in two areas of the tomb: inside the burial spot and in the central areas. In the burial spots, most artifacts were placed on or near the body. Jewelry, coins, and clothing lay on the body, and glass vessels stood by the feet. Garments and jewelry were likely worn at the time of burial, and the other finds put by the body in the coffin, loculus, or single tomb shortly after burial. These items, therefore, were associated with the funeral, or more generally, with the period between the moment of death and the sealing of the grave. There is little stratigraphic evidence that graves were opened for the addition of new artifacts. This is less obvious for the objects in the central area of the communal tombs, which could have been deposited at the time of burial or at an earlier or later stage. The time of placement of these artifacts cannot be determined with any accuracy. Who owned these objects? Scholars often assume that many of the finds were personal belongings of the deceased, in particular the items of jewelry, clothing, and unusual artifacts such as weapons,figurines,tools,and spinning implements.

DISCUSSION

There is, however, little to reconstruct ownership, and these objects could also have been gifts from others, or purchased for the occasion of the funeral. Two artifacts in the Syrian assemblage carried a personal name: a gold bracelet in Beirut was inscribed “Claudia Procla,” and a bottle in a tomb at Dura Europos was painted with “Gadda, son of …” in Palmyrene Aramaic (T. 43, 44). More items with inscriptions remain unpublished.21 Yet,even in these cases,it remains uncertain the issue of ownership cannot be resolved easily. An often heard explanation is that the deceased needed certain items during a transitional phase or while residing in the afterlife. In the absence of a textual or visual record of an afterlife, this remains speculative. In fact, even when artifacts can be related to concepts of afterlife, such as the coin for Charon, their actual function in the grave is open to various and perhaps locally specific interpretations. Concepts of afterlife also do not explain the placement of objects outside the burial spot, where they were not associated with a single individual. In such cases, it remains unclear who would need such objects, and when. Perhaps some objects were polluted due to their involvement in funerary rites. Pollution concepts were discussed in Chapter 1, when considering the emphasis on extramural burial. Pollution could affect not only the corpses and participants but also the artifacts used in the funeral. This would explain the abandonment in the tomb of lamps and embalmment and banqueting/offering vessels. Such items could not be taken back to the house, as they were now polluted. Perhaps we can extend such interpretations even to personal belongings, such as garments, jewelry, and some of the idiosyncratic finds such as a comb or flute. Their association with the deceased individual rendered them unfit to be worn or used by others. They had to be removed from the sphere of the living and symbolically killed, in order to achieve purification for the living. At this stage, however, such an interpretation has to remain speculative. Despite the difficulties in interpreting grave goods, the assemblages in Syrian graves refer to two ritual practices: protection through magic and beautification and the provision of offerings. Many items, especially those placed on or close to the body, may have had magical properties. Dalley notes that in older Mesopotamian funerary rituals, ointments served both as toilette and as a means of pacification of the spirit, to ensure it would not come back to haunt the living.22 The items in Syrian tombs related to embalming, dressing

21

22

These include a golden ring with precious stone in which a chariot and Greek name were engraved (Online Appendix Hauran 1, #8), a lamp with engraved bottom (Online Appendix Tyre 1, #13), an inscribed golden ring (Online Appendix Tyre 1, #23), a golden ring with Greek engraving (Online Appendix Deb’aal 1), an engraved bead of black stone (Online Appendix Palmyra 1, #26), a golden ring with engraving of a royal bust (Online Appendix Homs 1, #3), and a storage jar with a painted text on top of a pit-grave (Online Appendix Selenkahiye 1, #7). Dalley 1993, 29.

99

100

GIFTS FOR THE DEAD

up, and application of make-up may have served not only to make the body presentable during the final stages of the funerary ritual but also to appease the soul. The practice of placing lead curse tablets in tombs points to a belief in spirits roaming the grave fields, who could be summoned to carry out curses. The magic items, thus, may have served a dual role of protecting the deceased and shielding the living from unquiet spirits. A small group of items was associated with offerings of liquids and incense: plastered water pots, incense burners, and sacrificial ladles or simpula. Both in Rome and closer by in the Nabataean world, perfumed oils could be offered as well.23 Plenty of evidence exists for such items in Syrian tombs. Objects in tombs may have been remnants from the ritual of offering that took place inside. In this case, it was not the object itself, but its contents (oil, incense, water) that were the primary reason for placement in the grave. There is no evidence for offerings of sacrificial meat or plant products. Here, it should be said that botanical and faunal remains were rarely sampled in Syrian tombs, save a few exceptions.24 In Chapter 5, we connect the evidence for offerings to the spatial configuration of the tomb and discuss the possible recipients for these gifts. Other than a focus on adorning and protecting the body and placement of offerings, there are no single purposes for the grave goods. This can be seen from the fact that the assemblages were highly diverse, even in the tombs that were never robbed. The grave goods do not point to internalized practices. Perhaps this changed over time, when artifact assemblages became more standardized and more common. Although it is hard to say what it was, perhaps a stronger coherence emerged in the role of artifacts in the tomb. In any case, we should resist the temptation to seek a singular purpose for the objects, which may have been adorning and at the same time protective, and may have served both as offerings to the dead and reminders that the family had performed the correct rituals.25

Funerary Rituals in Space and Time Chapter 2 directed attention to several distinctive practices in the spatial configuration of tombs. To some extent, these are also reflected by the grave good 23

24

25

Sachet 2010. Liquid offerings of perfume in Pompeii: Van Andringa et al. 2013, 921. Wine is also mentioned in relation to (non-funerary) libations in the context of Palmyra and Dura Europos (Kaizer 2002, 190–191). One bottle in the hypogeum of Djel el-’Amed reportedly held grains (Online Appendix Tyre 1, #2), and one grave in Selenkahiye contained a long bone of a large mammal (Online Appendix Selenkahiye 1, #34). Two hypogea in Hama yielded bones of sheep, fowl, horses, and oxen (Online Appendix Hama 1, #11, #15). It is uncertain whether these were original deposits or the results of later usage of the tombs as animal pens and garbage dumps. Cf., Van Andringa et al. 2013, 921–922.

DISCUSSION

assemblages. Take, for instance, the intense variety in types of tomb and modes of burial. This variation also prevailed in the grave good assemblages, and people were buried with different assemblages. Others were never accompanied by objects. We have already seen a distinction between communal and single tombs. Communal tombs rose in popularity in the Roman period, and it was particularly in the context of these that a greater elaboration of and energy expenditure on funerary architecture occurred. The analysis of grave good assemblages indicates that communal tombs yielded more valuable finds when compared to single tombs, although the differences are not very pronounced. Within the communal tombs, sarcophagus burial contained the richest assemblages. The placement of finds both in- and outside the burial spots in the communal tombs signifies that the grave goods performed more and different roles than in the single tombs, likely pertaining to different rituals. Although we cannot say that these rituals did not take place in the context of single tombs as well, it is clear that they did not take place at the tombs. The analysis of the space and the gifts in single and communal tombs suggest distinctive ritual practices, an argument that is further developed when we discuss who is buried in the next chapter. The selections of grave goods also changed little from earlier customs, and instead stress continuity in mortuary practices from the Hellenistic and Parthian periods to the Roman era. Although it is difficult to discuss changes over time within the grave good categories, the pre-Roman assemblages point to similar ritual practices. In other words, if protection, beautification, and offering explain the placement of artifacts in the Roman tombs, these were also important in older traditions. The composition of the grave good assemblages did not remain stable throughout the Roman centuries. The quantities of jewelry decreased after the 1st c. CE, around the same time as the tombs contained more coins. Perhaps coins took over some of the symbolic functions of jewelry, in terms of both valuable metal and magical or protective properties. Vessels connected to food preparation, serving, and dining seem to disappear in the 3rd c. CE. The associated rituals, therefore, no longer took place in the tomb. To come back to Parker Pearson’s statement at the beginning of this section, gift exchanges with the dead, as revealed by grave good assemblages in the tombs of Roman Syria, involved the need for adornment and protection of the deceased body, as well as the presentation of offerings. The items were part of a series of ritual activities that took place in the tomb, and may have involved offerings to the deceased. Chapter 5 discusses these ritual practices by combining the collection of artifacts with epigraphic and spatial evidence. First, however, we turn to the occupants of the graves in Chapter 4.

101

CHAPTER FOUR

THE DEAD: BONES, PORTRAITS, AND EPITAPHS

W

ho was buried? this chapter focuses on the builders and occupants of the tombs in Roman Syria.The previous analysis has highlighted distinctions within the burying group, in terms of funerary space and grave good assemblages. Here, I investigate the identity of these people. Identity here is understood as the broader groups that individuals and communities identified with, in as far as it is visible in the funerary record. It concerns gender, age, ethnic, professional, collective, and individual identity (see also Introduction, p. 15). Funerary portraits and epigraphy reveal the represented identities of the deceased and the burying community. How were they depicted in art and text? What messages were considered important to highlight, and what was omitted? Ultimately, these questions concern commemorative practices: how people wanted to be remembered and what the community thought was important to stress. This chapter also addresses the physical remains of the dead, which inform us about the age and sex of the buried community, as well as trends in collective and individual burial. The three categories of evidence – figural imagery, epigraphy, human remains – come with their own sets of interpretative issues, which form the topic of the first section of this chapter. Subsequent paragraphs explore the treatment of the body, the custom of communal and co-burial, the distribution of sex, gender, and age groups, and kinship and military identity. In the final sections, I analyze chronological developments and patterns of continuity and change. By the 1st c. CE, new modes of commemoration had emerged,

102

SOURCES

and direct references to the bereaved community as heirs and survivors now adorned the tomb walls.These were inserted into long-standing traditions concerning co-sharing the tomb and preserving the body. SOURCES

Human remains, epitaphs, and figural sculpture of the deceased provide explicit information about the occupants and sponsors of the tomb.Each offers valuable insights concerning identity, and comes with specific limitations, be they a lack of reportage or a decontextualized and partial publication. These sources represent different aspects of the mortuary ritual and cannot always be compared. Text and imagery focus on representing the deceased, and concern commemorative practices. Sometimes, the deceased him- or herself was involved in formulating the text of an epitaph or purchasing the portrait; in other cases, a family member or fellow soldier sponsored the tomb. In most cases, however, we do not know who was responsible. The practices of these two groups – the future deceased and the surviving community – cannot therefore easily be disentangled, and, with a few exceptions, are treated together in the chapters that follow. The human remains illustrate different aspects and inform us who was (actually) buried. Very rarely in the Syrian case are skeletal remains excavated in tombs that also yield inscriptions or funerary busts. The sources discussed in this chapter, thus, do not necessarily talk about the same people. As a result, this chapter moves back and forth between the represented and the buried community.

Human Remains The field of osteoarchaeology is relatively new to Roman archaeology in Syria. Human remains for a long time tended at best to be merely counted, or, at worst, to be discarded without any reportage. Most tombs in the sample were devoid of human bones upon discovery, or contained remnants that were never published. In about 40% of the tombs from the cat. 1 sample (203), a minimum of 1779 individuals were discovered. These data are not always appropriate for analysis, as the degree and trustworthiness of investigation varied per site. The age determination, for instance, sometimes seems conducted on the basis of the size of the remains. Such a questionable method fails to distinguish between adolescents and short adults. At Apamea and Homs, the age of the deceased was only recorded when a tomb belonged to a child, resulting in an overemphasis in the sample on burials of the young. Another example of problematic methodologies concerns the sexing of skeletal remains. At times, this appears to be conducted on the basis of grave goods, whereby jewelry was associated with female burials. I demonstrate the fallibility of this association later in this chapter.

103

104

THE DEAD

As a result of these methodological problems, the evidence used in this book is restricted to those sites where osteological research took place: Beirut, Hama, Palmyra, and Selenkahiye. Also included are the handful of well-preserved mummies of Palmyra, where the sex and age could be determined by visual inspection. Together, the evidence consisted of the remains of 629 individuals, the majority (431) of which were found in Palmyrene tombs. Even in these reports, inconsistent usage of terminology poses a problem. Common terms such as “infant,” “child,” and “adult” are not specified in terms of calendar age or age range. When possible, I make the distinction in this book between infants (under 1 year old), children (1–16 years), and adults (over 16 years). The data set is primarily used to investigate sex and age distributions. Data about diet, lifestyle, and disease are rarely available, and are not (always) related to mortuary practices.

Inscriptions The next category is extraordinarily abundant compared to human remains. Roman Syria has produced thousands of funerary inscriptions commemorating the deceased or the owners of the tomb. More than a third of the tombs from the sample were adorned with one or more epitaphs (430), and well over 1500 are listed in the cat. 2 collection. Funerary inscriptions come with a different set of drawbacks. Epitaphs were carved and sometimes painted on a variety of media: stelae, coffins, cippi, altars, closing slabs of loculi, and tomb entrances. The vast majority were found removed from their original context, and we know little about the tombs they once adorned. Strong disciplinary divides between those who study inscriptions (epigraphists) and those who study the stone or tomb that holds the text (archaeologists) further complicate the investigation. Whereas the texts tend to be well published, it is often the only part that is published. We learn little to nothing about the shape and type of stone it was carved in, or associated relief decorations. For this reason, a high number of inscriptions have ended up in the cat. 2 assemblage of this book. Their sheer number, nevertheless, makes them hard to ignore, and the epitaphs from Roman Syria provide persuasive indications for changes in commemoration of the deceased. As stated in the Introduction, this book is not an epigraphic study, but uses inscriptions to reflect on patterns emerging from the archaeological record. It is hoped that scholars in the future will initiate a comparative study of funerary epigraphy from Syria. In order to integrate the epigraphic data into the contextual analysis of the identity of the burying and buried community, I start with an overview of the chronological and regional patterns and the content of the inscriptions from the Roman province of Syria. Only a handful of epitaphs stem from the 1st c. BCE. Their numbers increase slightly in the 1st c. CE, and close to 90% date

SOURCES

105

table 5. Distribution of inscriptions (site and language) from cat. 1 tombs

Site

Aramaic/ Latin & Aramaic/ Nabataean Palmyrene Language Greek Latin Greek Nabataean and Greek Palmyrene & Greek unknown

Apamea (55) 15 Baalbek (4) 2 Beirut (1) Bosra (8) 2 Deb’aal (1) 1 Dura Europos (2) 1 Hama (1) 1 Hauran (8) 1 Homs (1) 1 Limestone 39 Plateau (43) Palmyra (298) 2 Tyre (8) 8 Total 73

39 1 1 6

1 1

1 1

2

4

1

48

3

3

1

2

272

24

272

24

to the 2nd and 3rd c. CE. They decrease sharply again in the 4th c. CE, but this may be the result of a bias in the database, which cuts off in the middle of the 4th c. CE. Numerous later examples come from the Beqa’ and Orontes valleys, the Limestone Plateau, the Hauran, Palmyra, and Tyre. Inscriptions were not distributed equally across the province (Table 5). Both rural and urban cemeteries yielded texts, but they were less common on the Levantine coast and in eastern Syria. Palmyra yielded the largest collection in the sample, and high numbers also originated from Apamea, Bosra, and Homs. The study of the distribution over tomb types illustrates high incidences of inscriptions on hypogea, stelae, and rock-reliefs. The cat. 2 examples predominantly were carved on stelae, and less often on sarcophagi and stones of various shapes (i.e., column drums, cippi, altars). Considering the fact that the presence of a legible inscription was often the sole reason for publication, and that the tomb type associated with the largest inscribed groups, the stelae, is unknown, these distribution patterns may be deceptive. The cist-graves and jar-burials in the sample, for instance, did not yield epitaphs, but could have been originally marked by an inscribed stone. Examples come from at Seleucia Pieria on the Syrian coast, where inscribed marble strips covered the cremation urns of sailors of the Roman navy.1 When inscriptions were discovered in situ, they marked the exterior portions of the tomb, usually close to the entrance, or burial spots such as sarcophagi and loculi inside the communal tombs. 1

Seyrig 1939.

7

106

THE DEAD

table 6. Distribution of inscriptions (site and language) from cat. 2 tombs Aramaic/ Latin & Aramaic/ Nabataean Greek & Language Greek Latin Greek Nabataean and Greek Palmyrene Palmyrene unknown

Site Apamea (97) Baalbek (18) Beirut (11) Bosra (241) Hauran (109) Homs (123) Jebleh (1) Limestone Plateau (4) Palmyra (1069)∗ Tyre (18) Total ∗

4 10 2 190 91 123 1 4

92 8 9 30

3

10

16 444

2 151

1

21 14

1

35

3 2

5

1

1000∗

56

1000

56

1

This number is an estimate based on PAT

Tables 5 and 6 list the preferences of languages in the cat.1 and cat.2 database, respectively. Greek and Palmyrene were most commonly used, with all of the texts in the Aramaic dialect of Palmyrene stemming only from their name site. The Aramaic/Nabataean texts came from the Hauran and Bosra. Debates exist as to whether this script was Nabataean or a local version of Aramaic.2 An unspecified Arabian dialect appears to be used for one inscription in the Hauran. Bilingual texts, usually combining Greek with an Aramaic dialect, were less common. Latin was reserved for Roman soldiers and their family members. Civilian epitaphs in Latin, all from the cat. 2 sample and found in Baalbek and Beirut, were likely associated with Roman colonists. The chronological distribution suggests that Semitic and bilingual Semitic–Greek inscriptions often belong to the earliest group of the 1st c. BCE and CE. Greek epitaphs were uncommon before the 2nd c. CE, when their number exploded. They dropped sharply again after the 2nd c. CE. Latin inscriptions were infrequent before the 3rd c. CE, although possible 2nd c. CE examples are listed in the cat. 2 collection from Bosra. Not all texts adorning tombs were strictly speaking funerary in nature. Sometimes, names identified mythological figures in painted scenes, such as the Hypogeum in al-Awatin (T. 45; see Figure 40). Participants in the Maioumas festival, furthermore, left a friendly note on a tomb in Tyre (T. 46): “those who celebrate the Maioumas in this place have nice days.” One group of inscriptions has a more indirect funerary association. These are the so-called “cession 2

MacDonald 2006, 285; Nehmé 2010, 453.

SOURCES

inscriptions” from Palmyra, recording the transfer or sale of a portion of the tomb. They appear to be copies of contracts of sale. In addition to naming the new owner(s), the cession inscriptions state that the owner’s offspring is meant to be buried in the newly acquired space. In that sense, they resemble the other funerary inscriptions that do the same thing, i.e., showcase the ownership of the tomb and identify rights of burial. The cession inscriptions were added to the tomb façades in Palmyra starting in the mid 2nd c. CE or slightly earlier. The remainder of the epitaphs can be subdivided into two types. Foundation inscriptions specify the dedicator(s) of the tomb, who paid for its construction, and sometimes the individuals who are intended to be buried there.The second group only names the dedicatee(s), for whom the tomb was destined or who was commemorated in the text. Whereas the founding texts were far more likely to be encountered on the exterior façade of the communal tombs, the majority of the examples from the second group were placed inside the tomb or on a grave marker (stele, cippus). The foundation inscriptions are some of the earliest in the sample, and occurred throughout the period under study in this book. The dedicatee inscriptions arrived in great numbers between the 90s and the 250s CE, with the majority falling in the 2nd c. CE. Their use diminished in the second half of the 3rd c. CE, but this may be related to the fact that two sites that yielded many examples stopped their production: Roman soldiers left Apamea after the Persian invasions, and in Palmyra, tomb construction mostly ceased after the revolt of the 270s. Pronounced regional trends can be detected in the content and language of the inscriptions. Extended records of lineage and the cession texts in the local dialect are typical for Palmyra. The Greek inscriptions from Bosra often repeat formulaic expressions (“courage,” “nobody is immortal”), and include the age of the deceased. The collection from Homs (cat. 2) provides specific calendar details, by giving not only the year but also the day and month of death or dedication. The trend of increasing regionalism that was already highlighted in Chapter 2 with the example of funerary architecture is equally pronounced in the epitaphs.

Visual Representations Sometimes, people were depicted on the tomb. Approximately 400 separate images depicting one or more individuals are listed in the tombs from the sample, carved in the living rock, or sculpted on stelae, sarcophagi, and the closing slabs of loculi. When they are coupled with inscriptions, we learn that they represent the deceased and their family members, the owners, and the future occupants of the tombs. The collection of figural imagery is heavily decontextualized. They were often removed from their original location, leaving open questions about

107

108

THE DEAD

placement and visibility. Unlike the epitaphs, the figural imagery is not extensively published and data about many examples are missing. In fact, the sample discussed in this book likely is only the tip of the iceberg, as illustrated by the collection from the Hauran and Palmyra. The Hauran has yielded no cat. 1 examples of figural sculpture, but the second-tier assemblage includes at least eight busts and two depictions of seated persons. In her publication on funerary material from the Hauran, Sartre-Fauriat mentioned fifty-two funerary busts, and the thousands of stelae and sarcophagi from this region may have carried figural representations as well. For Palmyra, Heyn counted 585 busts and 282 group scenes and figures in relief, yet only about 200 are included in the cat. 1 sample.3 We can expect that the total number of visual representations was much higher than the sample suggests. The collection of visual representations of figures from funerary contexts consists of reliefs, sculpture in the round, and painted depictions of busts and complete figures. Often, the term “portrait” is used, which in this book means a representation of an individual deceased. These are not to be taken as truthful depictions of what people looked like, in the sense of veristic sculpture. The main trends in figural sculpture from tombs in Roman Syria can be summarized as follows. Most portraits come from Palmyra, and the collection of this site is better studied than the rest. A longer description of this material is added in Appendix 1. Palmyrene figural sculpture is usually accompanied by an inscription identifying (some of) the represented persons. Famous are the relief portraits decorating the closing slabs of the loculi, usually depicting a bust or half-figure (Figure 28). The tombs also contained group scenes, known as reclining or banquet scenes, carved on relief slabs and sarcophagi. In Palmyra, figural sculpture was introduced in the mid 1st c. CE. Non-sculptural evidence comes in the form of painted portraits in medallions carried by Victories from two hypogea dating to the 2nd–3rd c. CE (T. 47, 48). Little is known about these, but they presumably depicted the deceased. The coffered ceiling of two tower-tombs also contained reliefs of portrait busts (T. 49, 50). Outside Palmyra, fifty-six unique figural representations were carved on stelae, sarcophagi, and rock-façades, or represented in portrait busts and statues, depicting a total of ninety-two individuals. This number excludes the few ambiguous cases, in which it is not certain if we are looking at the deceased or a mythological character. It is a diverse group, with clear regional and professional distinctions. To start with the latter, about half of the sample belonged to soldiers of the Roman army. We learn this from the Latin epitaph and the imagery of soldiers with their weapons and horses. The largest collection 3

Heyn 2010, 633. The results of the Palmyra Portrait Project directed by Andreas Kropp and Rubina Raja were not available in time to be considered for this book. Hauran: Sartre-Fauriat 2001, vol. I, 4, 217–218; see also Skupi´nska-Løvset 1999, 208–230.

SOURCES

109

B

A

28. Tomb C (T. 32), Palmyra. A: Plan of tomb. B: Back wall with figural reliefs

(twenty-five) originated from Apamea. In styles and types of imagery, the figural sculpture followed regional trends, with signature busts in Palmyra, schematized basalt carvings of the Hauran, and low reliefs decorating the cliff façades of the Limestone Plateau.4 Those from the Limestone Plateau mostly 4

Skupi´nska-Løvset (1999) and Wenning (2001) detected similar regional trends in nonfunerary figural sculpture.

110

THE DEAD

dated to the 2nd–3rd c. CE, and included busts and seated and standing figures. Sarcophagi with reliefs of busts and figures, as well as round-sculpture of reclining couples resting on sarcophagus lids, were found in the cemeteries of Tyre and Hama (150/170–250 CE). The more extensive collection of figural representations from cat. 2 tombs points to the popularity of their use in the hills and valleys of western and southern Syria. The Levantine coast and the steppe region, by contrast, yielded small quantities. The chronological phasing suggests a rather sudden introduction of figural imagery in the 2nd c. CE, and slightly earlier (mid 1st c. CE) in Palmyra. However, figural stelae and perhaps rock-reliefs were produced on a small scale in the pre-Roman centuries, at least around Antioch and the Lebanese hills (see Chapter 2 and Appendix 2). The sample, therefore, demonstrates the rapid increase in the use of such representations, and their expansion across the province. Parthian tombs have not yielded examples. The use of funerary sculpture in Palmyra ceased entirely after the revolt and invasion in the 270s CE. Elsewhere, visual representations of the deceased and their families remained popular at least until the mid 3rd c. CE. TREATMENT OF THE BODY

Having established the basic patterns in the three source categories considered for this chapter, I now turn to the discussion of the identity of the buried and burying community. We start with the treatment of the body. Death starts with a corpse, and its handling reveals the beliefs about what happened to the social persona of a deceased person. In Syria, inhumation was the common method of burial before, during, and after the Roman centuries. In combination with the fact that all burials appear to have been primary burials, this indicates a strong insistence on keeping the body intact. Aberrations from this standard practice also exist. Three sites have yielded cremation graves. Archaeologists discovered six urns and pits with cremations in the North Cemetery of Apamea, dating to the 1st–2nd c. CE. The cat. 2 collection from the same cemetery includes thirteen additional cremations in jars. At Dura Europos, three cremations in urns lay in between the other tombs of the main cemetery. The remaining instances stemmed from Palmyrene tombs, where cremated remains were sometimes intermingled with inhumation burials in loculi of hypogea.5 Given the excavation methods at many of the sites under 5

Online Appendix Palmyra 1, #2, #111, #112. The Hellenistic Baalshamin Tomb at Palmyra held two possible child cremations, in a jar and in a pit. According to the excavator, these were Roman intrusions, but no further indication is provided concerning the date (Fellmann 1970). Two cremations at Tell Sheikh Hamad could have been either Parthian or Roman in date (100–250 CE). Here, the cremation remains were placed in a pit (Novák 2000).

TREATMENT OF THE BODY

investigation, cremation graves could have easily been missed or discarded. Nevertheless, the practice does not appear to have been common. Scholars often connect the custom of cremation to soldiers of the Roman army, and the previous section mentioned an example from the Syrian province: at Seleucia Pieria, ashes of sailors of the Roman navy were buried in urns marked by inscriptions. The fact that cremation graves were discovered at sites known to have housed Roman military – Apamea, Dura Europos, and Palmyra – perhaps explains the presence of these burials in Roman Syria. On the other hand, inscriptions on two sarcophagi in a hypogeum at Apamea (T. 51) and on the entrance to a hypogeum at Qatura (T. 52) suggest that soldiers also practiced inhumation. Palmyra provides evidence for another uncommon type of treatment of the corpse. At least forty mummies were reported from five tower-tombs (T. 49, 50, 53–55). This number may have originally been higher, as the fragility of the mummified remains and their attraction for robbers and early travelers probably resulted in undocumented removal and destruction. The same tombs also yielded non-mummified inhumation remains, indicating that both practices existed, although their contemporaneity cannot be established. As mummies originated only from the earliest Palmyrene tomb types, the custom may have died out by the later 2nd c. CE or earlier. Recent study has demonstrated that the corpses were first dried, then wrapped in cloth and covered with several layers of myrrh paste.6 Exquisite textiles covered the mummies,which also wore sandals.7 The mummies included individuals from various age and sex groups. Overall, the body of the deceased was treated in manners similar to preRoman practices. We have already seen hints of this in Chapter 3, where grave goods point to strong continuity in ritual practices focused on the body.Protection and safe-keeping were important, and this section illustrates that keeping the body whole, or at least not cremated, was another main feature. Chapter 5 comes back to this custom when discussing the different stages of funerary ritual. The exceptions to the standard practice are equally relevant. Cremation burials of soldiers represent an entirely different concept of treatment of the corpse. This group will also emerge in the following sections as having distinct mortuary rituals, pertaining to commemoration and memorialization. Soldiers appear as a marked group within the burying community of Roman Syria. The same holds true,although on a different level,for mortuary practices in Palmyra. Residents of this city developed distinctive practices of commemoration.

6

7

Schmidt-Colinet et al. 2000, 56. See also Pfister 1934, 1937, 1940. A mummy with traces of bitumen, wrapped in silk and linen and wearing a face-mask, was discovered in 1848 in a tower-tomb at Halabiye (Curtis 1995 and Hoffmann 1878, 25–26). According to Stauffer (2012, 89), the mummies were otherwise not dressed.

111

112

THE DEAD

SHARING THE TOMB

Carved high on the columns of a distyle tomb on the Limestone Plateau was the following inscription: “in the year 201 [152 CE] Eisidotos, son of Ptolemais, made all for himself and for Markia, daughter of Kodratos, his wife; and he shall lie in his own sarcophagus, the third in the first arcosolium on the right as one enters” (T. 56). This tomb was clearly destined for more individuals besides Eisidotos and Markia, as corroborated by the fifteen burial spots in the hypogeum that lay underneath its double columns. We have already seen evidence for the practice of communal burial in Chapter 2, and the inscriptions confirm that tombs could be reserved for more than one person. A total of 108 epitaphs record the communal burial of parents, spouses, children, brothers, and perhaps slaves. At least sixty-nine group scenes were carved on funerary materials. In this case, we are not certain if they depict the deceased community or a single deceased person surrounded by, for instance, living family members. The shape of the tomb, the epitaphs, and the figural sculpture, nevertheless, point to a common practice of communal burial. The number of people buried often varied according to the source, as illustrated by tombs on the Limestone Plateau and Palmyra. A hypogeum at Beshindlaye was dedicated to the parents of Tiberius Claudius Phylocles, but included at least nine burial spots (T. 7). In other cases, inscriptions over the entrance to the tomb referred to a single man while the tomb held multiple graves (cf., T. 57). Sometimes, sculpture and text did not match. According to the epitaph over the entrance of a tomb in Qatura, it was reserved for Flavius Iulianus, his wife, and their descendants. The accompanying relief image only shows us a reclining male, presumably Flavius (T. 58). In many cases, neither the number of individuals in inscriptions nor the number in figural reliefs correspond with the space of the tomb, or with each other. One can conclude that the texts and reliefs were not intended to portray all those allowed to be buried inside, but rather were to commemorate the founder or the individuals in whose honor the tomb was built. There is only one tomb on the Limestone Plateau where the space of the tomb conforms to the sculpture: a relief above the entrance of Tomb 2 at Daousat with three loculi portrayed a man, a woman, and perhaps their child (T. 59). The Palmyrene evidence also provides inconsistent messages about who was commemorated and who was actually buried, and as on the Limestone Plateau, the discrepancies often center on gender and age. Palmyrene foundation texts claim the communal tombs for the male dedicator and often his male descendants. Up to several hundred individuals could be buried in the large towers and hypogea of this site. Sculpture and inscriptions identified the graves of the founder and his sons, but also their wives, mothers, and daughters.

SHARING THE TOMB

29. Tomb of Abedrapsas (T. 60), Frikya. A: Drawing of entrance and plan. B: Relief of Abedrapsas and Amathbabea. C: Row of busts with inscriptions

The foundation texts, thus, promote the ownership of the tomb and the heirs, rather than (only) commemorating the occupants. As we will see next, over time male ownership gave way in Palmyra to mixed forms of gendered ownership. The Palmyrene inscriptions also present evidence for the existence of hierarchy within the tomb. We have discussed this while investigating the shape of the tomb, which appears to direct axial attention to the back wall of the elongated tombs. The epitaphs confirm that there was a preference for the placement of the tomb of the founder in back of the central area. Figure 28 depicts Yarhai son of Lišamš, the founder of Tomb C, who was buried in the elaborately sculpted back wall of the tomb (T. 32). Such examples are rare outside Palmyra. The foundation text of Abedrapsas’ Tomb (T. 60; Figure 29) was inscribed in a tabella over the central grave in the back. Other than this example, and Eisodotos’ unusual epitaph mentioned at the start of this section, inscriptions did not specify the location of burial of the founder.

113

114

THE DEAD

Co-burial Syrians were not only buried together in a tomb but also placed together in a single loculus or sarcophagus. This practice of multiple-occupancy burial, or co-burial, was fairly common in the province, as about a third to half of all graves yielded such evidence. This practice was not evenly distributed over tomb types. Only in a few instances were single tombs, such as pit-graves, used for co-burial. These examples include a pit-grave and sarcophagus from the Limestone Plateau that held a spousal couple (T. 61, 62) and a relatively small sarcophagus from Bosra that was assigned to the three children and husband of the dedicator (T. 63). Twice, a double stele marked a pit-grave at Palmyra (T. 64, 65), at least one of which was dedicated to two individuals. The skeletal remains confirm that co-burial occurred in single tomb types, but that the practice was uncommon. Only four examples were recorded in the sample. Most single graves contained one interment. Co-burial was common in the communal tomb types. The evidence comes from human remains, inscriptions, and figural sculpture. Figure 30 depicts examples from the latter group. Inscriptions identified them as parent and child, and, less frequently, as husband and wife, brothers, or cousins. Such depictions are not always conclusive evidence for the practice of co-burial, and we are on firmer ground with the human remains. When information about skeletons in a grave is provided, about half contained more than one individual. On average, four individuals shared a burial spot, be it a coffin, loculus, or pit. This number is probably distorted by the heavily reused funerary enclosures of the al-Bass Cemetery in Tyre, where the loculi and sarcophagi held over thirteen individuals on average. Many of these burials date to the Byzantine era. Elsewhere in the province, the average was 1.8 individuals per burial spot, but with great variation: the total numbers of co-buried people in a single burial spot ranged from two to twelve. Co-burial was most common in funerary enclosures, a pattern largely the result of reuse of the al-Bass graves. On the other hand, the enclosures of Beirut, which were not reused in later periods, contained on average 2.7 individuals, almost one person more than the provincial average. In Tyre, the freestanding sarcophagi within the enclosures contained the highest numbers of interments. This pattern was mirrored in sarcophagi placed in the hypogea elsewhere in the province, which contained more burials compared to the other forms. Pits dug in hypogeum floors, by contrast, almost always contained only a single burial. The spread of co-burial over the architectural types indicates that the shape of the tomb was an important factor. A loculus or sarcophagus was easier to open than a jar-burial or pit-grave filled with earth. In the Palmyrene hypogea, the back wall of the main room contained more shared loculi than other parts of the tomb. This means that the most exclusive location in a Palmyrene hypogeum,

SHARING THE TOMB

30. Co-buried individuals. A: Reliefs from Tomb of Yarhai (Palmyra). B: Sarcophagi in Complex V (Tyre, al-Bass). C: Stele 4 (Baalbek, cat. 2). D: Relief from Tomb C (Palmyra)

the one that the inscription tells us was reserved for the founder, was in fact the most heavily reused of all graves. These loculi may have been destined for the head of the family, but he was not the only one ending up in there (Figure 28). Perhaps we can conclude that the most prestigious burial spots – the tombs of the founders and the sarcophagi – were the most popular choices for reuse.

115

116

THE DEAD

Some graves were quite full, whereas others in the same tomb contained only a single individual or were empty on discovery. This was also the case with the relatively undisturbed tombs from Beirut and Palmyra, suggesting a conscious choice of the burying community, rather than chance of archaeological recovery. In fact, minor distinctions are visible in the grave good assemblages as well. Within the communal tombs, those burial spots containing only a single deposition had three grave goods on average, whereas those with more interments held a single find. In other words, the people who did not share their grave were accompanied by more items than those who did. The types of finds were relatively similar, except that pottery vessels were slightly more popular in burial spots with a single individual. The practice of co-burial persisted throughout the Roman period. It cannot always be determined whether the multiple interments occurred contemporaneously or represent decades or even centuries of use. We have seen examples of the latter from the funerary enclosures found at Tyre. Grave goods and decoration of crosses indicate continued use into the Byzantine period in a hypogeum at Hama (T. 66). Yet, sometimes, burial may have taken place within a shorter time-span. Complex 1 in Beirut, a funerary enclosure (T. 39), was in use for a period of about 125 years, as evidenced by the artifacts. Of the undisturbed burial spots in this complex, at least three contained multiple individuals. The dead were placed on top of one another, and in other instances, the bones of previous burials were swept to the side to make way for the next interment. The twelve individuals in these three compartments were buried within the 125-year lifespan of the tomb, following the decomposition of earlier burials, when it was possible to sweep the bones to the side. One curious case in this tomb had nine individuals placed on top of one another. The low level of intermingling of remains suggests that at least some of the remains had not entirely decomposed when a new corpse was added. Considering the small dimensions of the compartment (2.25 × 0.64 × 0.72 m), one wonders how this all fitted. Other burials also provide indications for the timing of reuse. Contemporary burial seems likely in a cist-grave at Selenkahiye, where an infant was placed under the hand of an adult (T. 67). In a loculus in Tomb C (T. 32) in Palmyra, the remains of the original burial of an adult male were disturbed by the subsequent burial of a mature woman, who, according to the epitaph, was his mother. In the tombs of Roman Syria, therefore, multiple individuals were co-buried at the same time, within a few years or decades from one another, or over a period of several centuries. The epitaphs tell us that coburied individuals were members of the same family. This family connection, and the emphasis on kinship identity, is starting to emerge as a main feature of funerary commemoration in Roman Syria. We will return to this issue later.

SHARING THE TOMB

Co-burial in Text, Image, and Practice When comparing the biographical information provided by inscriptions and sculpture with the skeletal remains and the space of the tomb, discrepancies arise. In the Habbasi Tomb in Hama, for instance, shallow niches occupied the space above the hypogeum’s loculi (T. 66). Some held statuettes and busts of the deceased (Figure 31). Each loculus had such a niche, but five were not spatially associated with a loculus. They held portraits of four women and a boy. Perhaps this sculpture commemorated the individuals buried in the three sarcophagi, centrally placed in the tomb. There was also no congruence of the skeletal remains in the loculi and coffins with the sculpture. For example, one loculus below a niche with a male bust held the bodies of an adult man and a woman. The entire collection of figural sculpture from the Habbasi Tomb consisted of five women, three men, and one boy, whereas the skeletal remains included a minimum of six male and six female individuals, and twenty-two adults and eleven children. The sculpture was added in 2nd c. CE, but the tomb was in use until the late Byzantine period. Later reuse, therefore, may explain the discrepancies between image and skeletal remains in the Hama tomb. Yet, it is also possible that it illustrates burial customs,as suggested by the following examples. In Palmyra, the number and identity of deceased mentioned in the inscriptions usually corresponded with the figural relief. A bust of a woman is usually accompanied by an epitaph commemorating a female. But this was not always the case. Sometimes, an adult and a child were depicted, with the latter omitted from the text. Not every individual in the Palmyrene group scenes is named by an inscription, either. The figural and epigraphic evidence suggest that most burial spots in the Palmyrene tombs were reserved for one person. The space of the tomb seems to confirm this. Loculus slabs with a single-portrait bust closed a single burial spot. The group scenes, or so-called “banquet scenes” (cf., Figure 30) usually covered multiple loculi, with room for all individuals marked in the visual and epigraphic evidence. The human remains, however, tell a different story. Take, for instance, the previously mentioned Tomb C, where seven closing slabs were still in their original position (T. 32). Four of the slabs were inscribed with a single name, but the associated loculus held two or four individuals. In each case, children and women were not mentioned. Even Yarhai, the founder who built the tomb for himself and his sons, shared his grave with a seven-year-old child (Figure 28). Another closing slab portrayed a man and a small boy, whereas the inscription referred to a man and his mother, and the loculus it covered held two adult males, one adult female, and a child. In fact, only in two cases in Tomb C did the human remains correspond with the associated inscription and sculpture. Similar discrepancies arise from all Palmyrene tombs where such data are available.

117

118

THE DEAD

31. Habbasi Tomb (T. 66), Hama. A: Figural sculpture from niches in right wall of tomb. B: Right wall of tomb. C: Plan of tomb

MEN, WOMEN, AND CHILDREN

Elsewhere in Syria, osteological information cannot be paired with inscriptions or sculpture, but disparities appear nonetheless. The Tomb of Abedrapsas on the Limestone Plateau may have held only a single sarcophagus or rock-cut coffin (T. 60; Figure 29). It was a small tomb, 3 × 3 m, built by Abedrapsas and his wife, Amathbabea. Both are depicted in a reclining scene, together with their daughter, a female servant, and a male figure carrying a cornucopia. The latter person was likely a deity, but one wonders why the daughter and servant were depicted. Were they also buried in the tomb? In fact, the tomb walls were decorated with numerous busts of men and women, the names of several of whom were identified by an inscription. Altogether, the epigraphy of this tomb referred to twelve different individuals, and the visual imagery to seventeen, whereas the plan of the tomb only indicates a single burial spot. Did this small tomb serve as the final resting place for all of them, or was there another reason for the inclusion of their busts and names? The long foundation inscription is of no help, as it only refers to Abedrapsas and his exploits. The Abedrapsas tomb and the Palmyrene examples provide glimpses into the complexities of ownership, burial, and commemoration. A full analysis combining spatial information with epigraphic, visual, and osteological data is not feasible for most other tombs in Roman Syria. However, we should suspect that co-burial was more common than text and image suggest, and that the biographical information of the inscription did not necessarily correspond with the sculpture and the skeleton. This was true with regards to both the burying group of a communal tomb and those within a single grave. Epigraphy and sculpture often referred to fewer individuals than were actually buried or than there were spaces for in the tomb. Usually, women and children were omitted in the epitaph or relief. In other cases, more people were depicted in reliefs or epitaphs than were buried, perhaps pointing to their function as cenotaphs. We will come back to this issue in Chapter 5. MEN, WOMEN, AND CHILDREN

A total of 221 skeletons in twenty-one tombs were sexed based on osteological research or, in the case of four mummies in Palmyra, preservation. Sex determination for children is complex, and the sample from Syria does not include sexed individuals below sixteen years of age. The adult group consisted of 119 males and 102 females. This distribution was similar across the province, except at Selenkahiye, where females dominated, with ten out of twelve individuals. Men and women often shared the grave, and three female burials included children. One woman was buried alongside another female; the remaining burial spots contained a single female interment. Men were more often buried with children (eleven) or alone (twenty-eight). Two burial spots in Palmyra held two males.

119

120

THE DEAD

The numbers are too small to discover patterns in sex distribution in tombs, other than that all combinations occurred and that children more often accompanied men. Tombs holding single female burials were common in the interior of Syria (Palmyra, Selenkahiye) in comparison to the western region of the province. The grave good assemblages also failed to reveal strong patterns between sex groups. Perhaps women received objects more often, but fewer of them. Graves of both sexes contained jewelry, and the discovery of such items alone is not a basis for sexing the grave, as has been done in several studies of Syrian graves. Female burials more often contained earrings and perhaps necklaces, while rings and amulets were preferred with male interments. These finds were not exclusive to male or female burials, however. There were higher frequencies of lamps accompanying male burials, although fewer per grave when compared to those of females. The evidence, albeit limited, implies that biological sex did not necessitate differentiation in burial, either in type of tomb, co-burial, or selection of grave goods. That this was quite different when it came to gender groups is demonstrated in the next section.

Gender in Text and Image The human remains tell us about distributions across the sexes, but the epigraphic and sculptural sources inform us about represented or socially constructed gender groups. In this study, gender identification was primarily based on the name provided in the inscriptions, and in a few cases on familial position (i.e., son, mother) or profession (soldier). Figural representations distinguished between genders, but limited preservation does not always allow for such identifications. Because of the richness of the Palmyrene data, we discuss this material separately later. In the collection from other Syrian sites, most inscriptions (275) provided information about the gender of the individuals mentioned. In total, 161 men were mentioned, forty-one women, six male children, and one female child. The figural imagery portrayed sixty men and twenty-five women. Stelae of (male) soldiers (twenty-nine) dominate this group;if we exclude these, male and female depictions were closer in number. These distributions demonstrate two points: the epitaphs more likely mentioned men (80%), and, when not counting the soldiers, visual depictions represented men and women in more equal numbers. This pattern was largely repeated in the cat. 2 assemblage, with slightly more female inscriptions in Bosra (37%) and Homs (24%), and less sculpture of women on the Limestone Plateau. Based on the complete collection of inscriptions of the Hauran, Sartre-Fauriat found 65% dedicated to men and 35% to women. By the 3rd–4th c. CE, this number had reversed, with 66% of the Greek inscriptions representing dedications for women.8 Our 8

Sartre-Fauriat 2001, vol. II, 112–113.

MEN, WOMEN, AND CHILDREN

sample is too small to reveal a similar reversal of gendered commemorations elsewhere in the province, but the Palmyrene epitaphs discussed later offer similar insights into changing practices in female commemoration. The founder of a tomb was usually male; women only sponsored four tombs, and five were dedicated by mixed groups,such as parents or spouses.The female dedications were directed at their husbands and/or sons, whereas the male examples involved a wider group of people: parents, wives, children, freed persons, and fellow soldiers. Tombs in the Hauran could be sponsored by groups of up to twelve men, some of who were identified as brothers. It was highly uncommon in Roman Syria, although not impossible, for a woman to found a tomb; or, rather, for this fact to be recorded on the tomb. One exception from Baalbek may be significant (T. 68). The inscription on an architrave, possibly belonging to a 1st c. CE mausoleum, records that it was dedicated to Zenodoros, son of Lysanias the tetrarch, by his wife, whose name is not preserved. Debates exist about the identity of this tetrarch and his son.9 Yet they clearly belonged, or claimed to belong, to an illustrious family. In the context of high elites, it may have been more appropriate or necessary for a woman to dedicate a tomb. Women were not prominent on the receiving end of a commemorative text either. About 20% were dedicated to women. Such epitaphs usually centered on a burial spot inside a communal tomb, although the original location of many of the inscribed stelae and coffins remains unknown. The remaining 80%, dedicated to men, are found both inside the tomb and on the exterior façade. A small group of inscriptions was dedicated to mixed groups: wife and son, parents, parents and wife, husband and wife. These were generally placed on the exterior of a communal tomb, and occasionally on single tombs. Overall, the inscriptions and reliefs mark the exterior of the tomb as male space,reserved for announcements by the founder, heir, or owner. Inside the communal tomb, one encounters a mixed family group. The composition of the group of tomb founders and dedicatees varied across time and space in Roman Syria. In Tyre and the Limestone Plateau, very few tombs were solely dedicated to women, and the latter region included a high proportion devoted to a mixed gender group. In the Hauran, including Bosra, Nabataean/Aramaic and bilingual inscriptions were more likely to be dedicated to a woman than those in Greek and Latin. Latin inscriptions were predominantly dedicated by and to male soldiers, although a handful of epitaphs involved a wife as sponsor or dedicatee. Examples of mothers and, in one case, a sister sponsoring a soldier’s tomb come from the cat. 2 assemblage. Four Greek cat. 2 inscriptions recorded long and emotional dedications to a wife, commemorating her beauty, qualities in handiwork, chastity, and love 9

Kropp 2013, 205–207.

121

122

THE DEAD

for her husband (T. 69–72). Shorter inscriptions in Homs (T. 73) and Bosra (T. 74) mentioned a loyal and virtuous wife. Dedications to men did not refer to personal qualities, although Quintus Aemilius Secundus boasted about his own military victories and career on his stele found in Beirut (T. 75). His, however, is an unusual text in the collection from Roman Syria, and the only one referring to a Roman career, the cursus honorum. Changes in gender roles are clearly visible in the epitaphs from Palmyra. We take the example of the Hypogeum of Nasrallat in the Southwest Cemetery (T. 76). This large hypogeum has three rooms, with thirty-five niches in which loculi were stacked (Figure 32). Upon entry, the foundation inscription in Greek and Palmyrene tells us that the tomb was founded in June 142 CE by Nasrallat son of Malikho, for himself, his sons, and their male descendants (Figure 32Ai). More than a century later, in the words of Ingholt, an inscription was “crammed on” to the same lintel.10 This text tells us that Julius Aurelius Yedi’bel ceded two exedrae of this tomb, to the left and right when one enters, to Julia Aurelia Amate, daughter of Bolhazai Moqimu, and her descendants in June of 263 CE. A third inscription on the same lintel records how two years later, the same Julia bought another four rows, by which is meant the stacked burial loculi, for her offspring (Figure 32Aii). Female ownership had reached the tomb, and Julia was able to purchase burial spots for her descendants, no longer specified as male. Similar inscriptions recording the transfer of ownership from other tombs in Palmyra illustrate that women could also sell parts of the tomb (cf., T. 77, 78). The fact that freed persons were sometimes involved in the sale of tomb space, and that the transfer of tomb space could occur outside the family, indicates that ideas of ownership changed significantly in Palmyra.11 And this was announced on the tomb. The interior of the Palmyrene tombs was always a mixed space. Despite the fact that the foundation inscription might limit the tomb to male descendants of the owner, we find numerous portraits and full-figure depictions of women. The texts and images included 60% male and 40% female commemorations, which is close in number to those given by Heyn based on a more extensive collection of Palmyrene portraits.12 The stelae in the Northeast Cemetery had a higher portion of male dedications: forty-five men and nine women. The previous section demonstrated that the skeletal remains did not always match the epigraphic and sculptural data placed on the burial spots in Palmyra. The proportion of female skeletons was higher than the distribution among genders according to the texts and images. By now it is clear that, across the Syrian province, men and women had different roles in commemorative practices. Female roles were less numerous, 10 12

11 Ingholt 1935, 110. Cussini 1995, 2005. Heyn (2010, 633) recorded 323 men and 262 women.

MEN, WOMEN, AND CHILDREN

32. Tomb of Nasrallat (T. 76), Palmyra. A: Entrance (i: inscriptions of 142 and 263 CE; ii: inscription of 265 CE). B: Plan of tomb

and women were more restricted to being dedicatees. They were always placed in the context of a larger group: the family. Their constructed identity thus revolved around spouses and offspring. Men had more diverse roles, as sponsors and recipients of burial space. Later, we will see that men also had

123

124

THE DEAD

identities constructed around military profession, and sometimes as priests or councilors. Gendered identities can also be gleaned from the visual representations. There is much work to be done on the study of Syrian figural imagery. Here, we provide general impressions. In terms of representation, men and women are to some extent represented in similar ways. That is, we see them in bust or half-figure form or similar sizes. Yet, there are notable distinctions (Figures 16, 28–31). Women wore a long garment with a veil over the head, with hand often reaching to the veil in the familiar pudicitia type of the Roman world. They were often seated, but rarely reclined. Men stood or reclined, and when not in military gear, wore long garments and sometimes a toga. The introduction of beards in the 2nd c. CE illustrates that the sculptors of Palmyra, Hama, and the Hauran were in touch with imperial trends. Some in Palmyra reclined in loosely fitted pants, the so-called “Parthian dress” (Figure 30). This would be an example of “native” dress, a type of garment that clearly had origins in local society. A burgeoning field in studies of the Roman provinces investigates these depictions of native dress, often in combination with gender groups.13 It is hoped that the material from Syria will be included in such analyses. If we can identify native elements in female presentations of Syria, it probably lies in the elaborate depictions of jewelry and headgear. Here, standardized depictions of women attained a local flavor. The study of artifacts held by the individuals in the portraits is also still in its infancy, and mostly restricted to Palmyra, where women carried spindles, distaffs, and keys, while men held scrolls and whips, or cups and bowls when depicted as priest. Children grasped birds or grapes.14

Age Distributions From the cat. 1 tombs, the ages of 275 adults and 150 children (under sixteen years, including infants) are recorded. Co-burial of different ages occurred. Children, for instance, were buried alone in a tomb, with other children, and with male and female adults. No tomb type was exclusively reserved for an age group or combination, but children were prominent in jar-burials and pit-graves, whereas single adult burials often appeared in loculi of hypogea. In Palmyra,pit-graves in hypogea contained single child burials and mostly infants. Small differences also emerged with regards to grave good assemblages. Single burials of adults had a slightly higher number of finds on average per body, and 13

14

Carroll 2013; Carroll & Wild 2012; Masséglia 2013; Rothe 2013. See also Schmidt-Colinet (1997, 163) and Stauffer (2012) on Palmyrene sculpture. Heyn 2010, 2012. See also Skupi´nska-Løvset 1999, 191. Finlayson (2002–2003) investigated the headbands of Palmyrene female portraiture and suggested that these were tribal markers. Sculpture in Hama’s Habassi Tomb (T. 62) had similar headgear. Headbands are also depicted on portraits from Belkis (Parlasca 1982).

MEN, WOMEN, AND CHILDREN

perhaps also a greater variation of finds, whereas tombs with multiple adults and tombs with infants contained finds more often (57–65%,compared to 35%), but fewer of them. As already noted, burials of multiple individuals included far fewer items of personal adornment. Children, and especially infants, received the most (77%), and high numbers of necklaces and beads. Infant burials did not contain vessels. The placement of coins varied, but may have been more common in child and infant burials. It is difficult to draw general conclusions from the distributions of grave goods according to age groups, other than that no exclusive patterns were apparent. Similar to the sex distribution, no find categories were exclusively associated with an age group. Perhaps we can say that artifacts in child graves emphasized aspects of adorning and protecting the body, rather than embalming or offering. Epigraphy provides a different picture. Two different “ages” can be identified in the inscriptions:years since birth and social age.15 The first category consisted of inscriptions (thirty-six) that mentioned the specific age of the deceased. This information was added on Latin inscriptions of Roman soldiers at Bosra and Apamea (thirty-three), but occasionally also found on non-military texts in Greek. Sartre-Fauriat comments that 70% of the funerary stelae in the Hauran, most of which were undated and not in the database, mentioned the age at time of death.16 This percentage presumably includes the cat. 2 inscriptions from Bosra, of which 103 examples included the age of the deceased. More cat. 2 examples are recorded from Baalbek, Beirut, Homs, and Tyre. Specific references to age were, thus, not restricted to Latin or military epitaphs. In the cat. 1 sample, the average recorded age was thirty-eight years (range twenty-one to fifty-four). This collection did not include children and mature adults, but these groups were better represented in the cat. 2 material, where the youngest was a newborn and the oldest reached 100 years of age. Children were underrepresented in this sample (0.3 children : 1 adult). Only ten epitaphs commemorated female children. Whether these ages reflected actual calendar years is not certain, and none of the texts could be compared to excavated human remains. Social age refers to life-phases mentioned in the inscriptions. For instance, soldiers, priests, and councilors can be considered adults, as can those who referred to their own offspring and those who were married. This is a relative category, as one can be buried by parents, whereby identity as a child is stressed, or by a descendant, and thus marked as an adult. Spouses were adults, despite possible young ages of marriage. These categories are primarily based on current ideas about what defines an adult or a child, rather than on the life-phases constructed by people in Roman Syria. As expected, the dedicators of tombs in Syria were always adults, but the dedicatees were also almost always of adult age. Only four out of seventy (cat. 1) epitaphs memorialized 15

Gowland 2006, 143.

16

Sartre-Fauriat 2001, vol. II, 104.

125

126

THE DEAD

children. Children seem equally underrepresented in the figural depictions, although this is a tricky issue as size and relation to other figures (i.e., part of a family group) are the main indicators of age, and are hardly trustworthy. Parents or other adults usually accompanied children. In Palmyra, among 214 closing slabs placed inside the tombs, the imagery and texts referred to nine male and three female children. Another thirteen stelae portrayed an adult with smaller figures, who were identified as their children in the inscriptions: mothers with son or daughter, and fathers with son. It seems that they died as children, i.e., before reaching adulthood. Similar distributions occurred in the Palmyrene group scenes: very few children were depicted, and about a quarter of these were female children. The mention and depiction of children was again smaller (10%) than the amount actually buried in the graves, which made up 41% of all the aged skeletons in Palmyra. Children were thus less likely to be individually commemorated than to be buried, and those who were commemorated were usually boys. Having said this, it should be noted that age cannot always be deduced from the inscription. Of a collection of forty-three inscriptions from tombs on the Limestone Plateau, for instance, roughly equal numbers are too fragmentary to reconstruct age groups (sixteen), include reference to age (fourteen), or do not include such references (thirteen). In the latter group fall inscriptions such as “Lucianus, farewell.” The actual number of epitaphs dedicated to children could have been higher. What we can deduce is that often the burying community did not find it important to add biographical information about the deceased, other than the name, and thus that child identities are rarely highlighted. It is noteworthy that when they are, commemorative practices hint at a special status of some children. Male children could receive rather elaborate epitaphs. Two sarcophagi in Tyre carried three long dedications referring to the Muses, and the sorrow of the parents of the deceased (T. 46). Combined with the cat. 2 examples, about 20% of all inscriptions in Tyre consisted of dedications to male children, a higher proportion than at any other site. Two epitaphs found at Shahb¯a in the Hauran commemorated the premature death of the son of Julius Priscus, the prefect of Mesopotamia, and, not coincidently, brother of Emperor Phillip the Arab (T. 79, 80). These inscriptions were erected by local notables of Shahb¯a instead of family members, which makes them unique in the Syrian assemblage. Perhaps the most remarkable and upsetting example in this category originated from the eastern frontier (Figure 33). A Greek funerary stele recorded the lament of the parents of Lucius, who died at the hands of sinful murderers. Found at Qamishly, close to Nisibis, a region where inscriptions were rare, the stricken family was likely military.17

17

Cumont 1933, 385–386.

MEN, WOMEN, AND CHILDREN

33. Stele from Qamishly

Demography and Practice The age, sex, and gender distributions in the tombs of Roman Syria were inconsistent with one another and do not reflect standard demographic patterns. The most conspicuous case is that of child burials. Due to high mortality rates in the ancient world, children should form a large proportion of the total assemblage, much more than 35% of the skeletal remains in the Syrian graves. This gap is probably partly the result of archaeological recovery, especially considering the fragile and small bones of very young children. Earlier, we demonstrated that when the biographical information on the epitaph or image did not correspond to the skeletal remains, children and women were most often omitted. Epigraphy and sculpture are, therefore, not good indicators regarding child burials. Nevertheless, carefully excavated tombs in Beirut and Palmyra also yielded few children, illustrating that their absence was also the result of practices in the past. Children less often received a burial that was archaeologically visible, and they were unlikely to be included in the group of honorees. Female children were least visible in the burying group: this was not an identity that was considered important to stress on the tomb. The next chapter provides more examples of a special status of children in Roman

127

128

THE DEAD

Syria, either through different mortuary practices or through lack of proper burial. The skeletal remains of adults showed a slight majority of male burials,which is demographically incorrect, but aligns with the overrepresentation of this group as a result of limitations in osteological sexing.18 Other than minor differences in grave good assemblages, there were no obvious differences in burial of sexed males and females.Differentiation between genders,on the other hand, was pronounced. Men were almost always the founders of a tomb, and were more often commemorated in text and image. Biological sex is not the same as gender. That is, the gendered features of a name and image do not necessarily correlate with the biological sex of the body. Biologically sexed females could have been commemorated under a male name or familial position (i.e., father), which would explain the discrepancies in the Syrian tombs. Such practices do not seem to fit our understanding of Syrian society, but they cannot be excluded. The inconsistencies are perhaps better explained by the fact that text and image did not exclude co-burial of others. Texts more often commemorated men but they did not limit burial to these men. Instead, the epitaphs specified who had rights of ownership, and the imagery honored the main founder of a tomb (see also p. 142). The cession inscriptions from Palmyrene tombs indicate that even these practices did not remain stable over time. FAMILIES AND INDIVIDUALS

In addition to gender and age groups, several other identities have emerged from the analysis of the funerary epigraphy and imagery. Most prominently highlighted are familial, individual, and military identity. These three characteristics of the deceased and the burying community are prominently highlighted in art and text. We start with a discussion of the familial identity, or kinship. When specified, people were buried by and with members of their family. The dedicator(s) of tombs could be fathers, husbands, mothers, wives, parents, spouses, and sons. The group of dedicatees consisted of fathers, parents, husbands, wives, and sons. Only a handful of epitaphs mention daughters and sisters. Overall, references beyond the nuclear family (parents, spouses, children) are rare. On the Limestone Plateau, a small number of inscriptions concerned brothers and grandchildren.More examples of fraternal sponsorship are listed in the cat. 2 collection. The nuclear family appears as the main unit of commemoration, but other positions, and in particular male siblings, could also play a role. Positive evidence for commemoration within wider kinship groups comes from Palmyra. The foundation texts presented the male dedicator and several generations of descendants. The dedicator could be alone or could co-sponsor 18

Parker Pearson 1999 [2008], 96.

FAMILIES AND INDIVIDUALS

the construction of the tomb with his brothers. Texts highlight the burial of brothers, half-brothers, and cousins in these tombs, illustrating that the tombs were destined for larger descent groups than just the immediate family. Abundant evidence exists for tribal structures in Palmyrene society, and the tombs possibly housed such clans. The epitaphs indicate that in Palmyra, extended families were buried along patrilineal descent: parents, unmarried daughters, brothers, their wives and children. Adult sisters and aunts, when married, would end up in the ancestral tomb of the spouse. There is an exception: one relief in the Tomb of Artaban included a family member from the maternal side, the brother of the wife of the founder (T. 81). On the other hand, if husband and wife were cousins, both the wife and her brother lay in the correct (patrilineal) tomb. Occasionally, freed persons occurred in the Palmyrene cession inscriptions, selling or leasing part of tombs. Epitaphs from the Limestone Plateau, Tyre, and perhaps Baalbek included enslaved people as sponsors and dedicatees. Slaves dedicated tombs to their masters or themselves, and the previously mentioned Tomb of Abedrapsas depicted a female slave. It seems that inscriptions referring to slaves or freed people mostly dated to the second half of the 2nd and early 3rd c. CE. As can be expected, mention of slave or freed status often occurred in the context of the (former) masters. The military stelae represent a different category. Whereas fellow soldiers erected the majority of the military stelae, a handful were (co-)sponsored by a family member (wife, son, mother, or brother). A sister and mother dedicated a tomb to a soldier in Bosra (T. 82). Freedmen of deceased soldiers (co-)erected a handful of stelae, and more examples of freedmen epitaphs come from Bosra and Palmyra. The military commemorations are discussed in more detail later. The collection of epitaphs from Syrian tombs demonstrates that, when specified, civilian dedications were almost always done in the context of the family. Usually, these were members of the immediate family, be they parents, sons, or spouses. The extended family was also invoked, including brothers and grandchildren, or non-related dependents such as slaves and freed people. The large number of unknowns in the sample complicates any simple statistical method of assessing the commemorative patterns. The data hint at a preference for nuclear families, but this may have been only slightly greater than that for brothers or slaves. Regional preferences existed, with Palmyra as a major outlier. Practices of co-burial also seem to have occurred in the context of the family. People were buried with their family or household members, at least according to the epigraphy. The instances of co-burial outside the family group are revealing. We have already seen that Palmyrene cession inscriptions recorded men and women ceding parts of the tomb to relatives or to people without any obvious familial relation. By the time of the sale of these parts, therefore, generally between the mid 2nd and later 3rd c. CE, non-family members appear to

129

130

THE DEAD

have occupied portions of the funerary building. Over time, the kinship grips on the ancestral tomb loosened. A hypogeum in Apamea held two centurions and their wives (T. 51). Although their names are not fully recovered and their familial relation cannot be ascertained, it is possible that these people were not related but co-owned a tomb. This would be the only instance in Syria of communal burial of people associated with the military, save for the veterans discussed later. We should briefly draw attention to aspects that are absent in the inscriptions: co-burial of members of religious groups, funerary associations, and professional organizations known from other parts of the Roman world.19 Perhaps people from the same religious community shared a grave, but this particular aspect of their identity was not highlighted in the text or represented in sculpture.

Individuals Epitaphs and figural sculpture on Syrian tombs also drew attention to the individual deceased. We learn their name, gender, age, and sometimes their position as founder or heir, as well as their (formerly) enslaved status. Often the sculptures should be considered portraits. They represented a specific person, although the agelessness of many busts suggests that there were not realistic or accurate representations of the deceased. Caution is necessary to avoid applying current – Western – ideas of individuality and free choice.20 On Syrian tombs, individuals were not necessarily represented as independent actors, but often commemorated as part of a larger whole. Text and image usually placed the deceased within the context of a larger group: the family or the military. The communal space of many tombs also did so. They do not show us single persons, but a collection of individuals. Why place such individual markers on a tomb? The trend is largely new in Syria, and is tied to a general popularity in the 1st–3rd c. CE of inscriptions and figural sculpture. Putting names and faces on tombs was not only popular in the Syrian province but also found in many regions of the Roman Empire. In other words, perhaps the individualization was tied to fashions crossing the Roman world, rather than necessities in Syrian funerary beliefs. This issue is the topic of Chapter 6, where I develop a framework in which both empire-wide fashion and local application can be considered. Here, we draw attention to space. The names and faces allowed precise spatial positioning of individuals within a cemetery and a tomb. The living could return to the grave and point to particular deceased members of their community. In Chapter 2, we discussed the overhaul in funerary architecture that occurred in Syria in the 1st–3rd c. CE. Epitaphs and portraits, introduced in 19

Patterson 1992, 19–24; Van Nijf 1997.

20

Cf., Gillespie 2001, 78–85.

FAMILIES AND INDIVIDUALS

the same period, further point to new concepts of space, firmly fixing family members in the landscape. They may also have become a sign of distinction. Individualization was not for everybody. Even in the same tomb, many burial spots were not marked. Some deceased were singled out, and most often they were adult males.

Religious and Professional Identity Aside from names, lineage, and association with the army, we learn little else about the burying group. Occasionally, epitaphs included profession or function. Three priests, two from the Limestone Plateau and one from Palmyra, were identified as such in the text (T. 81, 83, 84). Palmyrene figures sometimes wore a modius, or priestly hat. A stele from the Hauran depicts a man with a pointed headdress, usually interpreted as priestly; and a female statue from a tomb in Hama perhaps wore Isiaic garb.21 Every now and then, other functions were specified: a hypogeum belonged to Zoilos the legislator (T. 57). The Palmyrene inscriptions sometimes added information about the father or (great) grandfather of the dedicator or dedicatee: Nurbel the physician, Dion Malku the decurion, and Maththa the councilor of Palmyra. Dromon’s sarcophagus in Tyre proclaimed that he had been the servant of the procurator of Syria Phoenice (T. 85). We have already encountered the brother of Emperor Phillip the Arab, who served as prefect of the province of Mesopotamia. Such biographical notes were rare, and almost always referred to religious and (high) administrative functions.None pertained to women. The cat. 2 collection adds a handful of other professions, such as a young doctor in Tyre and a mosaicist at Bosra. Such mention of non-religious or political position is reminiscent of the Byzantine inscriptions in the al-Bass Cemetery dedicated by purple-dyers, lobster fishers, and wine sellers. A handful of Latin inscriptions from Baalbek and Beirut (cat. 1 and 2) did not mention any military affiliation and were probably associated with colonists. These stelae honored five men and three women. Sometimes, the reliefs depict people partaking in activities, such as offering or reclining. Both are infrequent, although reclining groups of men and women were more popular at Palmyra. Scholars interpret such scenes as banquet scenes, and some of the Syrian examples included elements associated with a meal, such as cups and a table (Figure 29; cf., Figures 30, 34). Two images of female recliners come from the Lebanese coast: one from Sidon, accompanied by a possible incense burner, and the other from Beirut, accompanied by various dining implements and a servant. Parlasca lists three additional stelae with reclining scenes of two men and a couple from the 21

Online Appendix Hama 1, #66. Hauran stele: Sartre-Fauriat 2001, vol. I, 249–250.

131

132

THE DEAD

Antioch region (perhaps 2nd c. CE), also with tables and servants. According to Skupi´nska-Løvset, reclining scenes may have been popular in the coastal area.22 In Chapter 3, we discussed the evidence from the grave goods, which does not point to a common practice of sharing tomb-side meals. It is not likely that the banqueting reliefs depict activities that took place at the tomb, although such banquets could occur elsewhere. The meaning of these reclining scenes is not easily disentangled and was probably not uniform. In Palmyra, the scenes bore similarities with the religious banquets that featured in epigraphic and sculptural sources (see discussion p. 152). Other examples belonged to military graves (Apamea, Dura Europos, Qatura), whereas the examples on sarcophagi from Tyre were carved in Asia Minor and Attica before being shipped. They did not include banqueting equipment such as cups and tables. Perhaps in more general terms, the reclining scenes represented the deceased in a religious or festive banquet, or added an association with the luxurious lifestyle of the wealthy in the Near East and Eastern Mediterranean (see also Chapter 5). SOLDIERS AND VETERANS

Text and image often highlighted the military identity of the deceased (Figure 34). Military stelae formed a distinct epigraphic and sculptural collection in Roman Syria.23 These texts differed in style from the non-military inscriptions, as they listed double or triple names, legion, rank, age, and the name of the person who erected the gravestone. Most also invoked the chthonic deities or Di Manes. The practice of adding place of birth or origin was restricted to the military epitaphs. The soldiers originated from Dacia, Gaul, Greece, Italy, North Africa Pannonia, Thrace, Turkey, and possibly Syria. Many military stelae included an image of the soldier with armor, weapons, and horses, but they could also be seen reclining or partaking in a sacrifice. Roman soldiers were usually commemorated in Latin. One or more male heirs erected the epitaphs, presumably fellow soldiers and sometimes freedmen. Unlike civilian groups in Syria, the soldiers appear not to have shared a grave or burial spot, at least not according to the inscriptions. They occupied single graves. There is more to say about soldiers’ identities. Sartre-Fauriat speculates about the existence of a separate military area in the Northeast Cemetery of Bosra, where many Greek and Latin stelae referred to the army. Seyrig describes two separate grave fields at Seleucia-ad-Pieria on the Syrian coast, one for officers 22

23

Skupi´nska-Løvset 1999, 100. Parlasca 1982, 17–18. Female recliner from Beirut: Online Appendix Beirut 2, #32 (Sarcophagus 7). Sidon: Dahari 1998, 172–173. The assemblage included forty-seven stelae from soldiers in active service, discovered in Apamea, Baalbek, Beirut, Bosra, Dura Europos, and Qatura. Among the cat. 2 examples, at least fifty-one originated from Baalbek, Beirut, Bosra, Homs, Palmyra, and Tyre.

SOLDIERS AND VETERANS

34. Military stelae. A: Stele 1 (Dura Europos). B: Stele 6 (Apamea). C: Stele 23 (Apamea)

and one for sailors of the Roman fleet, although he does not clarify the exact find spot of the remains.24 Perhaps the fact that the soldier stelae of Apamea were reused in one section of the city wall of this town indicates that they 24

Sartre-Fauriat 2007b, 331; Seyrig 1939.

133

134

THE DEAD

came from a single location. In each of these cases, however, the existence of physically separate burial grounds cannot be proven with certainty. At Bosra, other burial grounds also yielded military inscriptions. At least two military tombs, one at Apamea and one at Qatura, were surrounded by civilian tombs (T. 51, 86). Scholars sometimes use grave goods to argue for an association of the deceased with the Roman army. The pit- and cist-graves at Nawa-tell Umm al-Hauran,for example,contained unusual assemblages,including spoons,medical sets, weapons, a set of strigiles, ivory fish with Latin numbers and Greek letters, and possibly fragments of a musical instrument. Two bronze helmets found along with fragments of a breastplate and a harness likely belonged to military parade gear. Pit-graves in Homs also yielded weapons and armor, as well as a parade helmet (Figure 27). Perhaps these graves belonged to Roman soldiers, yet such links between types of grave good and professional identity in life remain ambiguous. Different people could have been buried with military gear, including family members of soldiers, local warriors who managed to beat the Roman army and brought their loot into the grave, and young men who never reached military age.25 If the graves held soldiers, it would also need to be explained why they were buried with their gear. In any case, the discovery of these items was extremely rare. At Homs, the other finds in the pit-graves do not have a military character; they include unusual bracelets, a torque, and other ornaments in gold and turquoise. The excavator linked these items to Iranian, Central Asian, Celtic, and steppic traditions.26 If grave goods and spatial location do not necessarily single out military tombs, tombstones and modes of burial do. The soldier epitaphs name a wider range of dedicators and dedicatees, i.e., fellow soldiers and freedmen instead of (only) family, often explicitly referring to the heir. Soldiers are one of the few groups who announced their professional identity on their tombstone, and they often added visual depictions in support. As mentioned earlier in this chapter, the handful of cremation burials originating from Roman Syria perhaps belonged to the military. Setting up a pyre and burning a corpse entailed a completely different set of activities than carrying a coffin or bier to the grave. If these were soldiers, this further emphasizes the different practices between military and civilian burials. The Roman military also took its customs to East Syria, an area that was culturally distinct from other parts of Syria, and generally devoid of epitaphs. The small collection of funerary stelae originating from this region was associated with the military.27 25 27

26 Cf., Roymans 1996, 18–19. Seyrig 1952a, 231–237. Examples come from Dura Europos (Online Appendix Dura Europos 1, #11), Sura (perhaps second half of the 1st c. CE; Konrad 2001, 9), and Nisibis/Qamishly (probably early 3rd c. CE; Gatier 1988, 227–229). A Greek stele commemorating a murdered child was also found in Qamishly (Cumont 1933, 385).

SOLDIERS AND VETERANS

The modes of commemoration and the forms of burial of soldiers in Syria are similar to those in other places in the Roman world. Military tombs, or at least stelae, represent a relatively well-studied group in the empire, and show a remarkable degree of uniformity in visual and textual representation.28 The soldiers who served and died in Syria, thus, were part of an empire-wide military community with its own cultural traditions, and as such diverted from Syrian practices. While in service, the soldiers remained a separate group and followed distinct commemorative and possibly burial practices. These soldiers came from all corners of the Roman Empire and did not, seemingly, maintain their native burial customs. Or rather, the identity that the tombstone stresses is one of profession and not homeland or ethnic origin. The evidence does not entirely support the existence of a “total institution” whereby the institutional identity of the army replaced ethnic identity and separated soldiers from civilians, since family members did accompany the soldiers. Endogamy may have occurred, an indication that ethnic identity did still play a role. Nevertheless, commemoration of the soldiers was standardized despite their different ethnic origins and differed from the practices of those not associated with the military.29 The invocation of the Manes illustrates that this also extended to religious practices. Whether locally recruited soldiers followed the same practices is unclear, but we have a few tantalizing clues. A mother and sister dedicated a stele to a soldier in Bosra (T. 82), which specified Philadelphia as the family’s place of origin. The epitaph followed military conventions, but it remains uncertain whether the birthplace was nearby modern-day Amman or another Philadelphia. Soldiers buried in Palmyra, on the other hand, sometimes combined a conventional Latin text stating (double) name and position in the army with a Palmyrene-looking portrait bust (T. 87–92; Figure 35). Their names certainly sound Roman – Flavius Julianus, Julius Bassus, Vibius Apollinarius – and nothing of the inscription reflects Palmyrene Aramaic and Greek epigraphic practices. These are curious cases: are we dealing with local men or soldiers inspired by what they saw around them? Did the sculpture workshops of the town have nothing else on offer? Palmyrene practices are unique in this sense, underlined by the fact that Palmyrene soldiers abroad, in Britannia and Dacia, maintained some of their own practices, in particular the style and iconography of tomb reliefs and the use of Aramaic.30

28 29

30

Coulston 2007; Hope 1997, 2000, 2001, 2003a, 2003b; 2011, 221–234; Saller & Shaw 1984. For a discussion of the Roman army as a “total institution,” see Shaw 1983. See also James 2001 and Pollard 1996. Carroll 2013; Smith 2013, 165–172. An earlier military epitaph from Palmyra (27 CE) was composed in Greek and belonged to an auxiliary unit known as the cohort of Damascenes (Online Appendix Palmyra 2, #48).

135

136

THE DEAD

35. Tombstone of Vibius Apollinarius, Palmyra

There are several different factors behind the development of a strong military identity in death and the visibility of this aspect. In general, as men far removed from their land, kin, and other social groups, they were prime candidates for the formation of a new identity. Hope has argued that, at least in Roman Britain, tombstones were a sign of distinction, and perhaps served to indicate territorial control.31 The audience was thus as much the other military as the civilian population. Furthermore, if burial was the prime role of family members, new “families,” i.e., fellow soldiers, had to be formed to perform this role. Burial clubs could also do this, but there is no evidence for their existence from (Roman) Syria. The new “family” is often explicitly present in the epitaphs as the designated heir. Some military epitaphs did explicitly refer to family members. They were dedicated to and by wives, children, and mothers, and in one case a mother and sister. If we leave out the brothers, who may have been soldiers as well, 38% of the military stelae refer to families. This percentage is much higher than the 14% that Saller and Shaw calculated on the basis of epitaph collections from 31

Hope 1997, 255.

SOLDIERS AND VETERANS

other parts of the empire.32 The strong group identity of Roman soldiers could extend to their families. In a few cases, and only in the context of child burials, the soldiers diverted from their regular practices and dedicated an epitaph in Greek. Examples come from Bosra and Qamishly.33 Distinctive commemorative practices within the soldier group, as found in other regions of the empire, such as between non-citizen Roman subjects in the auxiliary forces (alae) and the citizen legionaries, or between centurions and lower-ranked soldiers,are not immediately apparent in the assemblage from Syria. The majority of the military epitaphs are in the second-tier category, and lack information about date, iconography, and original context. In the betterpublished group from Apamea,most tombstones belonged to legionary soldiers (twenty-four), and only nine to those serving in the auxiliary forces. Those of the first group date mostly to between 200 and 230 CE, whereas the latter are dated to the middle of the 3rd c. CE. The epitaphs are similar for both groups. The Syrian assemblage provides better evidence for distinctions between ranks. In Apamea, of the five centurions mentioned, only one was found in the same location as the lower-ranked soldiers. Two centurion families were buried in a hypogeum in the North Cemetery. This could point to separate treatment of centurions in terms of location and type of tomb, although the numbers are too low to detect any significant patterns. In the entire assemblage of military epitaphs from Syria, centurions made up about a quarter. The remainder was dedicated to lower-ranked soldiers or soldiers of unspecified rank. Relative to this number, centurions more often dedicated to family members or received a stele from family members, mostly wives. It was unusual to mention origin, and a relatively larger proportion of those that did belonged to centurions. In other words, if ethnic markers (origin or homeland) were present, they were more often found among centurions. The story changes when we consider the veteran inscriptions. Greater variation is visible in this, admittedly small, group. Veterans’ epitaphs were in Greek, Latin, or both. In the Limestone Plateau region, Valerius Romulus constructed a simple pit-grave with his wife in 310 CE, covered by a Greek inscription on the sarcophagus-shaped lid (T.61).In the same area,another couple was responsible for a grand hypogeum with a sculpted front adorned with a Greek and Latin text (175–200 CE, T. 58). A Greek text over the entrance of a mausoleum in the Hauran tells us that Kelesteinos built this place for himself, his children, and his beloved wife, from his income from military service (T. 93). According to this impressive CV in Latin, Quintus Aemilius Secundus, the Roman colonist in Beirut, held, in addition to a string of public offices, the position of 32 33

Saller & Shaw 1984, 133–134. Online Appendix Bosra 2, #169 (IGLS 9337), 192 (IGLS 9358), 224 (IGLS 9413). Qamishly (Nisibis): Cumont 1933, 385–386.

137

138

THE DEAD

Praefectus cohortis (T. 75). A veteran stele from Apamea followed more closely the formula of the soldiers in active service at the same site (T. 94). The cat. 2 examples, from Qanawat and Bosra, confirm the variation in commemorative practices of veterans. While in service, soldiers and their families copied standardized forms of commemoration, whereas those discharged from service had little reason to continue these customs. BURIAL AND COMMEMORATION ACROSS SPACE AND TIME

This is a good moment to take a step back and consider the material in light of trends that emerged from the previous chapters. The first one concerns diversity. People in Roman Syria were not commemorated in similar ways, as they were not buried in the same types of tomb or accompanied by similar sets of grave goods. In fact, multiple options existed: one could lie alone in a single tomb, together in communal tombs, or share the grave with many others. Within the same cemetery or tomb, some deceased were memorialized in script and relief, whereas others were just put in a hole in the ground. Even the most standardized collection, the military stelae, portrayed each soldier in a slightly different way and specified personal aspects such as name, military rank, and years served. The result of these practices often was the individualization of funerary space. There was always a group that did not participate in the new trends and continued to be buried without visual markers stressing kinship or other relations. On the Limestone Plateau, thirty-eight of seventy-seven tombs did not yield inscriptions. Since the number of unprovenanced epitaphs was low from this area, unlike in other areas such as Apamea and Palmyra, the act of inscribing a tomb was less common on the Limestone Plateau than the quantitative evidence based on cat. 1 suggests. Regional differences are apparent. Palmyra represents the most obvious case. Mummification existed here, and the trend of individual commemoration in text and image started several decades earlier than elsewhere in the province. Throughout the centuries, Palmyra retained its distinctive character in sculptural styles and epitaphs. The detailed recording of the sale or lease of portions of tombs is also only known from Palmyra. Elsewhere in the province, regional variation in sculpture and epitaphs largely followed the trends in architecture and decoration described in Chapter 2. I will summarize the most recognizable features. The figural sculpture of the Hauran and Limestone Plateau was regionally distinct, although the epitaphs from both regions displayed few internal commonalities. The content and sometimes language of the inscriptions could vary per cemetery and tomb. Greater standardization of funerary texts is visible in the collection from Bosra. Here, Nabataean texts included a double name, and Greek examples had a double name, age, and consolation term such as “courage” or “nobody is immortal” (cat. 2). Across the province,

BURIAL AND COMMEMORATION ACROSS SPACE AND TIME

epitaphs recorded the name of the deceased and/or sponsor of the tomb. Dates were added in Apamea, Palmyra, Homs, and the Limestone Plateau, and sporadically at Tyre and Baalbek and perhaps in the Hauran. Consolation terms (“regretted,” “farewell”) are found everywhere. Apamean inscriptions often held only a single name, as did a third of the sample from Homs. References to age were common on military stelae. Examples from Bosra concerned military and non-military tombs, and Tyre yielded several memorials to children. As we have already seen, funerary stelae were uncommon in the region east of the Euphrates, the Syrian steppe. Next comes the question of continuity and change. To what extent were the trends discussed in this chapter continuing from older periods? The placement of bodies in the graves displayed the highest levels of continuity. Inhumation was and had been the dominant form of body treatment. Co-burial occurred in the communal tombs of Hellenistic Beirut and Hellenistic-Parthian Dura Europos. People shared single burial spots in loculi as well as cist- and pit-graves in Beirut, Dura Europos, Palmyra, and Tell Kazel (see Appendix 1), as well as in Tell Sheikh Hamad.34 In Palmyra’s older Baalshamin tomb, the skeletal remains were swept aside to make space for a new interment, whereas in Hellenistic Tell Kazel, multiple individuals were placed on top of one another. Very few reliable data are available with regards to human remains. Osteological analyses exist for Hellenistic tombs from Beirut, Dura Europos, Palmyra, and Jebel Khalid.35 Only a handful of skeletons were sexed: eight male and four or five female. The tombs held eight children and forty-one adults. These numbers are not high enough for the establishment of sex and age distributions. The most extensive osteological report stems from the cemetery at Tell Sheikh Hamad, dating between 200 BCE and 220 CE. Here, between 28 and 62% of the tombs belonged to female individuals and 16–38% to males. This possible low number of male burials could be found in all centuries, except the latest – Parthian/Roman – period (100–220 CE).36 Children at this site made up 59% of burials, of which 15% were infants. The proportion of child burials at Tell Sheikh Hamad, and possibly that of women, was higher than in the Roman period. It is not possible to establish whether this was part of a general Parthian trend or specific to Tell Sheikh Hamad. However, in the Roman tombs at Selenkahiye, also in the Syrian steppe, females occupied 83% of the adult graves, and almost half of all individuals were children. Child burials were also more present in Roman Palmyra when compared to western Syria. These numbers imply, but by no means prove, that women and children received a

34 35

36

Novák 2000. The material of Beirut, Dura Europos, and Palmyra is described in Appendix 2. For Jebel Khalid, see Jackson & Littleton 2002; Littleton & Frohlich 2002. In this period, 17–67% belonged to male burials and 33% to female burials (Novák 2000).

139

140

THE DEAD

more (archaeologically) visible burial in eastern Syria before and during the Roman period. Commemoration of the deceased through art and text was highly unusual in the centuries before Roman rule. We have already touched on this topic when discussing tomb types and decorative forms (p. 58). The Parthian region of Syria did not produce sculpture depicting the deceased and their family, and only a single inscription, consisting of a name and date, in a hypogeum of Dura Europos. The material from Hellenistic Syria was limited to the coastal regions, which produced a handful of stelae, rock-reliefs, and inscriptions (see sections on stelae and rock-reliefs in Appendix 2). They depict seated and standing single individuals and groups of mixed gender. The pre-Roman assemblage was not only small in comparison to Roman material and regionally restricted but also less diverse. Only a single set of military inscriptions comes from Sidon, dedicated to male soldiers by members of their hometown or polity, and in one case each by companions and a wife and brother. Civilian stelae from the Levantine coast included a dedication by a father to an adult son and one by a son to a father-in-law, as well as three individual dedications, commemorating a woman and two men. Extended family members, children, and (formerly) enslaved persons did not feature on pre-Roman epitaphs. None of these inscriptions included dates or other legal terms used in the later examples. Professions mentioned are suffete, chief of porters, and soldier. One man was depicted as a priest. The numbers are small, but they suggest that distinction between genders in image and text were similar to those of the Roman period. Women were more likely to be depicted in relief than named in text. To summarize, the small collection of sculptural, skeletal, and epigraphic evidence for Hellenistic and Parthian Syria followed the trends of the later Roman material, including communal and co-burial, sex, gender, and age distributions, and the importance placed on family and military identity. The most profound change was the use of inscriptions and sculpture. The vast majority of pre-Roman tombs were never marked by text and image. Until the 2nd c. CE, the use of inscriptions was limited; most stemmed from the 2nd and 3rd c. CE. Their employment is attested into the Byzantine period, although perhaps on a smaller scale, and restricted to a few regions. Foundation inscriptions were common throughout the Roman period, whereas those commemorating individuals were largely confined to the period between the 90s and the 250s CE. Figural depictions were not common until the mid 1st c. CE, when they appeared in Palmyrene tombs. Elsewhere in Syria, their popularity rose early in the 2nd c. CE. Their usage declined around the same time as the inscriptions. The prevalence of texts and figural images was, thus, chronologically aligned with the general introduction of tomb decoration, as discussed in Chapter 2, as well as with the increased variation of tomb types. In other words, they played a role in the diversification of funerary architecture and

DISCUSSION

the new functions that occurred alongside continuity in communal or single burial.

DISCUSSION: FEAR, MOURNING, AND COMMEMORATION

The rising popularity of both figural decoration and epitaphs is not limited to Roman Syria, but is traceable across the Roman Empire. In particular, the chronology of the use of epitaphs follows the trend known as the “epigraphic habit.” This describes the spectacular rise of the use of inscriptions in most areas under Roman control in the 1st and 2nd c. CE, peaking around the turn from the 2nd to the 3rd c. CE. Bodel estimates that funerary inscriptions account for perhaps two-thirds of all Greek and Latin texts of this period.37 Even regions where the use of epigraphy long pre-dated the coming of Rome, such as Turkey, witnessed a significant rise in popularity and geographic expansion between the late 1st and the 3rd c. CE. These inscriptions highlighted various aspects of the dead and the burying community, usually identified the deceased by name, and often added bibliographic details. Saller and Shaw have demonstrated on the basis of several epigraphic collections in the provinces that commemoration was regularly carried out by the nuclear family and that children were often underrepresented. In other words, as we already noted with regards to military burial, both the idea of using an epitaph and the modes of commemoration reveal how Syrians were situated within empire-wide trends.38 This can also be argued with the busts and full-length figures placed on tombs, which likewise make their way across the Roman world. Chapter 6 addresses this issue and places Syrian families in the context of the Roman world. Here, I discuss how the new messages on the tombs highlight new attitudes toward the space of the tomb and the mourning process of the surviving community. Text and image drew attention to the communal identity of the burying group and, at the same time, to the individual. The communal identity is that of the family and the army, whose members were commemorated as part of the group. Part of the funerary ritual, therefore, as well as the choice of grave-site, was governed by lines of kinship and profession. Before the Roman period, lineage and perhaps military background may have held equal importance, but these aspects were rarely highlighted in text or sculpture. The architecture of the Roman tombs also directed focus to lineage. The aboveground construction of communal tombs visually stressed communal family identity. We have little evidence about the shape of military tombs. Most soldiers appear to be buried in single graves. 37 38

Bodel 2001, 30. The term “epigraphic habit” was coined by MacMullen 1982. Saller & Shaw 1984.

141

142

THE DEAD

At the same time, the epitaphs and portraits also emphasized individual identity. Single tombs and burial spots in the communal tombs were marked by the names and images of the deceased. This trend started in full force between the mid 1st (Palmyra) and the 2nd c. CE. Again, we cannot argue that the association of a single grave with a deceased individual was completely new in the Roman period. Older tomb types also allowed for individual burial. Yet, in the mid 1st–2nd c. CE, names and faces appeared on the grave. Most often, the external surface of the tomb highlighted family connections, and inscriptions placed in the interior denoted individual burial space. Tombs as a whole thus represented the family, the different members of which would be encountered inside. The family roles were not static, as is best illustrated by the sale of portions of the Palmyrene tombs outside the family. This was advertised on the exterior façades. The distinction between family-exterior and individual-interior did not hold true in each case, as families were sometimes represented in text and image inside the tomb, and, conversely, some external epitaphs only drew attention to the founder or dedicatee. The purpose of funerary texts and images can be summed up as legal, protective, and commemorative. To start with the first: the inscriptions specified legal ownership and the obligations of the heirs. This is explicit in the case of the military stelae naming the heir, who, by erecting the memorial, had fulfilled his or her obligations to the deceased. The addition of a calendar date perhaps reflects the importance of the text in legal terms. One can assume that the cession texts in Palmyra were copies of contracts stored elsewhere.39 Inscriptions in general may reflect contracts or references to legal ownership. The reasons why, in the mid 1st–2nd c. CE, it became important to specify the legal terms of a contract on the actual tomb building are less easy to unravel. Uncertainty about ownership and inheritance had perhaps increased as a result of rapid demographic change, migration, and sedentarization. This is further explored in Chapter 6. The inscriptions indicated that ownership of a communal tomb was usually in the hands of male members of the community.Women dedicated a stele or individual inscription to husband and/or sons, whereas male foundations included wider family groups. Contrasts existed, therefore, between the commemorative practices of men and of women, at least until the mid 2nd c. CE in Palmyra and the 3rd–4th c. CE in the Hauran, when women were prominent as owners and renters. The differences in gendered space remained: the exterior portions of the tomb usually displayed men, whilst female dedications centered on a burial spot inside. The outside of a tomb belonged not only to the family but also to the male founders. The interior represented a mixed space. Children, who could not be owners, were rarely included. 39

Cussini 1995. She notes that aspects of the texts were known from earlier Aramaic legal documents.

DISCUSSION

The discrepancies between text, image, architecture, and skeletal remains demonstrated in this chapter reveal that the inscriptions were not meant to specify or limit the burying group. Many more individuals ended up in the tomb. Nevertheless, and this was the second function of the text and images, there was an element of restriction. Not everyone could use these tombs. Names and portraits communicated who was allowed to be buried and, as a consequence, who was barred. The audience of these messages consisted of members of the same group and unwanted visitors. A small group of inscriptions explicitly referred to fears of unwanted access to the tombs. A priest in Brad, on the Limestone Plateau, for instance, stated that no one besides his son could be laid aside him, nor could the tomb be sold or leased (T. 83). Resale was also a main concern for the owners of a tomb in Me’ez (T. 95) and Qatura (T. 58). An inscription on a 2nd c. CE sarcophagus in Lattakia (Loadicea) warned against sale and specified that if somebody purchased or took a mortgage on the tomb, that person should pay 5000 denarii to the imperial treasury and 5000 to “my homeland.”40 We have more of these inscriptions referring to fines, each of 5000 or 10 000 denarii to be paid to the imperial treasury. A sarcophagus in a funerary enclosure in Tyre (T. 85) was reserved for Dromon, servant of Arellius Carus the procurator of Phoenice. The text specified, “whoever dares to do something hostile (?) to this sarcophagus, finds himself prevented by virtue of law (?) and will not pay less than 10 000 denarii to the imperial treasury.” It is possible that this warning was for grave robbers rather than against illegal reuse, like the inscription on a sarcophagus in Jebleh, which indicates that the deceased “does not rest among riches.”41 Curses are also attested, such as on a hypogeum in Palmyra: “whosoever sells the security, which provides against the tomb, is sinning against his soul.”42 Reuse of burial space by unrelated or other unwanted individuals may have been common, which would explain some of the inconsistencies between skeletal remains, individual inscriptions, and sculpture. Perhaps the placement of names and faces on graves performed a similar function. They communicated who was allowed inside and, therefore, who was not (see also Chapter 5). Warning inscriptions are perhaps connected to ownership, inheritance, and the effort of the sponsor to keep the tomb within the family long after death. The specific mention of heirs and the engagement of family members through texts may have served 40 41

42

IGLS IV, #1254 (pp. 14–15). Online Appendix Jebleh 2, #9 (Sarcophagus 1). Other inscriptions mentioning fines from Palmyra and Homs are mentioned by Gawlikowski (1970, 203–204) and in IGLS VI (2652). Fines frequently feature on tombs in Turkey; see, for instance, Harter-Uibopuu 2014. Online Appendix Palmyra 1, #93 (Hypogeum of ’Abd ’astor). More Palmyrene curses are listed in Gawlikowski 1970, 219. An early Syriac epitaph found near Urfa reads: “ … whoever removes my bones, may he have no afterlife and be cursed” (Healey 2009, 248–250). Such inscriptions warning against misuse and disturbance are attested in the Achaemenid period (Healey 1995, 190–191; Röllig 2004, 27, 31).

143

144

THE DEAD

to remind these people of their duties in keeping the body safe. The function of the copy of the contract was not the same as that of the original, which recorded ownership and inheritance. Rather, it highlighted the duties of the heir. It was placed on the family tomb, where heirs would return for future funerals, and finally for their own.43 Perhaps the most important function of the inscriptions and sculptures was the memorialization of the founder of the tomb and the commemoration of the deceased. The founder advertised (usually) his role in providing a (family) tomb and his access to the resources required to accomplish this task. The living community commemorated the deceased in epitaphs and sculpture. A cat. 2 inscription from Bosra (T. 74) recorded the good qualities of a deceased wife and mother, and added a greetings to passers-by. A stranger passing this grave now also learns of Aurelia’s prudence and love for her husband. Many inscriptions included terms such as “rejoice,” “carefree,” “farewell,” and “missed,” or specified that the departed was loved, indispensable, or died before any of her children. This addition of consolation terms is interesting, and requires study outside the scope of this book. Such terms appear to address the deceased and the bereaved community, who could find consolation in the carefree status of the dead. It was this community, after all, that was left behind and that needed to initiate the separation of the deceased from the world of the living. The longer inscriptions mentioning the qualities of a wife or the sorrow of parents more clearly reflect the consolatory nature of the epitaphs. They assisted the mourner in grieving. Perhaps the imagery performed similar functions. Portrait busts and reclining sculpture commemorated the deceased in happier times or as a participant in essential group activities. The timeless nature of the figural sculpture, not aged or on the brink of dying, depicted the idealized form of the departed. The question remains why these elements were introduced in the Roman period. The dead had always been mourned, and one can hardly imagine that the surviving community experienced deeper grief in the Roman times than before. Perhaps text and image were initially transported to the grave due to trends of individualization in the Roman world and greater fear of illicit reuse (see Chapter 6). Once there, they provided new avenues through which to engage the mourning of the survivors. This chapter ends with two reservations. First, inscriptions and figural imagery were less common on the Lebanese coast and in the steppe lands of East Syria. The function of funerary text and image was not as relevant in these areas. Perhaps it was performed by other aspects of the tomb. For instance, expressions of communal identity are clear in the large visible tombs of Beirut and Tyre. An inscription was not necessary here to draw attention to the important families buried in the funerary enclosures. Inside, the burial 43

See also Gee 2008.

DISCUSSION

spots were not individualized by text and portrait, but rather by great variation in type of burial and in the materials used. The second reservation concerns legibility. The epitaphs could not always be read, because of their location high on the walls of the tombs. The levels of literacy and bilingualism are unknown for this region or period, and it is unclear how many people were able to read. Although the text could be read out loud during graveside activities, this would have done little to bar unwanted visitors. It should be kept in mind, therefore, that the texts could have held a symbolic function,and that their readership may not have been only mortal. It is to this cosmological belief system – funerary beliefs – that we turn next.

145

CHAPTER FIVE

FUNERARY BELIEFS: DIFFERENTIATION, CONTINUITY, AND CHANGE IN RITUAL

T

ombs in roman syria displayed a variety of messages about their occupants and the people who buried them. The previous chapters have analyzed particular categories of funerary material: space, architecture, grave goods, and the deceased. This chapter combines these data to examine the norms and practices that guided the funeral. What was the belief system behind the decisions on how to bury and commemorate the deceased? What happened after death, and what steps were necessary to allow the community to deal with the loss of one of its members? The patterns in the funerary record are analyzed on three levels:practices that encompassed the entire buried community, those through which distinctions were made within the burying group, and those that changed in the Roman period. The first level uncovers the deep structures of Syrian funerary belief: the ones that applied to all burials and changed little over time. The second level addresses burial differentiation and the ways in which funerary ritual was used to mark out or ignore certain groups in society. What aspects of the social persona of the deceased resulted in distinctive funerary practices? The third level addresses whether and how funerary practices changed from the pre-Roman to the Roman period, as well as throughout the Roman centuries. Were the changes in tomb architecture and decoration that emerged so clearly in the previous chapters related to new concepts of how to bury and commemorate the dead?

146

THE DATA

THE DATA

The aim is, thus, to reconstruct funerary beliefs – what people thought happened after death – and the primary source is the material record. Yet, how does one move from a tomb to a set of beliefs? A common approach is to interpret funerary beliefs as an integral part of religion. That is, scholars assume that fixed ideas existed about what happened to a person after death and that these were tied to religious beliefs.1 An example is the notion of a journey by the deceased to an underworld or heaven, ruled by chthonic or celestial deities. Both Greco-Roman and Near Eastern literary and visual sources have given us examples of such stories, but their connection to actual funerary ritual is rather uncertain. In fact, better-studied contexts such as Classical Greece and Imperial Rome fail to provide evidence for fixed notions about the fate of the departed. Some dead may have travelled to an underworld or reached apotheosis, but many stayed close to the tomb. Their spirits could be asked for help, and sometimes they wreaked havoc on the living. Direct links between religious beliefs and graveside rituals are, at most, ambiguous. The divine world was not completely irrelevant in matters of death, and later I discuss the possible evidence for activities at the tomb that mirror religious ceremonies. The translation from the material record to funerary beliefs is much easier when contemporary literary sources describe the ritual actions performed at the tomb. For example, plentiful sources about funerary ritual around Rome in the Late Republican and Imperial periods clearly outline the moments of separation, liminality, and reintegration of the deceased and the survivors, the classical steps of a rite of passage. In Syria, such evidence lacks, and even the inscriptions remain fairly silent on the matter. The discussion in this chapter, therefore, occasionally looks outside Syria, to Rome, but also to Greek, Jewish, and Nabataean practices, to provide a framework for understanding Syrian material patterns. This is not to say that Syrians had beliefs similar to Romans or Jews.Rather,these better-known examples provide a starting point for interpretation of the patterns in the material record of Syria. The same approach is taken with the evidence from Palmyra. This site has yielded better evidence by which to reconstruct modes of burial and commemoration than has the rest of Syria. The analysis in this chapter, thus, runs the risk of projecting particular or exclusively Palmyrene practices on to other Syrian contexts. Parallels certainly existed, but it remains to the seen how similar Palmyrene funerary beliefs were to those prevailing in other regions of Syria (see p. 210). A final point regarding the evidence is the tomb focus of this book and the lack of information about commemorative rites that took place outside the tomb. Wakes, lying in state, processions, banquets and other celebrations could happen at various places and 1

See Tarlow 2013.

147

148

FUNERARY BELIEFS

locations in Syrian towns and villages, ones that are not under investigation in this book. The tomb only gives us the end point of the funeral, as well at the commemorative rites that occurred inside or on top of the tomb. PROTECTION, POLLUTION, AND THE BODY: BASIC FEATURES OF SYRIAN FUNERARY BELIEFS

The dead were separated from the world of the living in cemeteries that extended outside the boundaries of the settlement. When cities expanded in size, the burial grounds moved accordingly, underscoring the importance of maintaining a separation between the two spheres. Considering the ubiquity of notions of the polluting dead in the ancient world and the need for physical separation, it is reasonable to suggest that similar concepts applied in the Syrian context (p. 33). Chapter 3 entertained the notion that the concept of pollution encompassed more than the physical remains of the dead. Items associated with the deceased or used during anointment and other parts of the funeral often were left behind in the tomb. Perhaps their impurity forbade people to take these artifacts back to the settlement. It should be noted that here we are on less firm interpretative grounds, and the study of possible polluted artifacts has still to be commenced. Literary sources indicate that pollution centered on the physical remains of the deceased, and on people associated with someone who had recently died – the family members. The mourning family and professional undertakers were polluted, but strangers who saw or were physically close to the deceased also were at risk.2 Placing the dead outside the settlement thus ensured the continuous purity within the sacred boundaries of the dwelled space. Yet, in Syria, when a town expanded over an older cemetery, neither its tombs nor its contents were removed. People lived on top of the physical remains – the bones – of people long past. Such inconsistency cannot easily be explained by pointing to a lack of knowledge about the location of tombs, or a lack of care as demographic pressure overtook pollution fears. Rather, it suggests that pollution concepts were flexible. The few instances of intentional intramural burial discussed in Chapter 1 illustrate the same point. If the Philippeion in Shahb¯a indeed housed the remains of Emperor Philip’s father (T. 6), its location on the central plaza of the town would have sent clear messages to onlookers. The power of the emperor and the social standing of his father permitted the transgression of long-standing traditions. Socially agreed norms concerning pollution and purity, and the relationship between humans and the sacred, were challenged by such a burial. Sartre-Fauriat lists five inscriptions found at small sites in the Hauran that suggest intramural burial. These inscriptions state 2

Parker 1996 [1983].

PROTECTION, POLLUTION, AND THE BODY

36. Loculus M 1–1 in Tomb C (T.32), Palmyra

that a tomb was constructed in or next to a domestic context. Two of them belonged to important individuals, a diplomat who lived abroad and a beneficarius of the governor.3 Burial within the city may,therefore,have been possible for people of high status. This trend is well attested in Roman Asia.4 The practice of intramural burial, nevertheless, was uncommon in Syria, and perhaps a rural and relatively late phenomenon. All examples come from the Hauran and the Limestone Plateau, dating perhaps to after the 2nd c. CE. That rural communities had less stringent rules about where to place the dead was also demonstrated in Chapter 1. We return to the issue of flexible pollution beliefs later.

Whole, Beautiful, and Protected Bodies Mortuary archaeologists have drawn attention to the particular care taken by the community in dealing with physical remains. Post-mortem treatment of the body modified the social identity attached to the now deceased.5 Keeping the body whole was a key feature in Syrian funerary practices. Inhumation, primary burial, and the frequent use of containers left the body intact and articulated. Figure 36 illustrates the depositional state of a loculus in Tomb C (T. 32) in Palmyra, with two superimposed skeletons of a male and female accompanied by glass bottles and silver clothing attachments. Pre-burial embalmment practices and the Palmyrene examples of mummification were other means of 3 4

Sartre-Fauriat 2001, vol. I, 19–20; IGLS XV, #109, #369. 5 Cf., Cormack 2004. Gramsch 2013.

149

150

FUNERARY BELIEFS

prolonging the wholeness of the human body. Inscriptions warned against removal or disturbance of the physical remains. Disarranging bones was undesirable, and perhaps had consequences for the spirit of the deceased. When grave goods were added, they often decorated the body. Jewelry covered the corpse, which was sometimes dressed in clothes with golden appliques or wrapped in a shroud. The body entered the tomb adorned. As discussed in Chapter 3, adornment and embalming may have served to pacify the spirit of the deceased, and the artifacts also carried apotropaic or magic connotations. The dead, or the corpse, needed help and protection. Sturdy tombs with heavy doors, some with locks, and burial in coffins kept the bones safe. Syrians were concerned with keeping the body intact, protected, and perhaps also fixed inside the tomb. The individual social persona was maintained, and at the same removed from the living. This interpretation stands at odds with how archaeologists usually discover human remains in Syrian tombs. Bodies are not always found intact. Rampant looting disturbed many remains, but the disarticulation of bones was not always connected to tomb raiders. In fact, the concept of wholeness was one of the most contested or contradictory aspects of funerary practices of Syria. Skeletal remains were swept aside to make place for new burials or disarranged by the superposition of a new body (Figure 37). Population pressure and lack of control or oversight on tomb access were perhaps among the causes. Warning inscriptions indicate that sale or reuse of the tomb represented a real fear for the tomb owners and future occupants. At the same time, reuse occurred within the family group and among others specified as legal users of the tomb in the founding inscriptions. The tombs in Palmyra in particular, where we can combine the biographic information from skeletal, inscriptional, and visual information, illustrate that the burial of family members disturbed earlier skeletons (see discussion of Tomb C, p. 117). Epitaphs elsewhere occasionally refer explicitly to the use of single burial spots for multiple family interments. The disarticulation of the skeletal remains, therefore, also occurred at the hand of the family. The integrity of the individual body was frequently denied, by strangers and new generations of kin, who opened the burial spots for reburial. The most disruptive practice, in terms of wholeness of the body, was that of cremation. Burning a corpse at high temperature actively disintegrates and fractures the physical remains. Such burials remain unusual in Syria, perhaps restricted to the Roman military. As we will see later, several aspects of funerary beliefs set this group apart from the civilian population in the province.

Tomb-Side Activities Excavations of burial grounds have thus far failed to identify constructions that were clearly used for mortuary activities, other than tombs. One can think here

PROTECTION, POLLUTION, AND THE BODY

37. Hypogeum at al-Awatin (T. 45). A: Cistern. B: Oillamps on bench

151

152

FUNERARY BELIEFS

of offering places, ossuaria, places of congregation, and dining facilities. None have been discovered in Syrian cemeteries. Occasionally, a temple was on (or close to) the burial grounds, but in each case – Baalbek (Sheikh Abdallah), Dura Europos, Qanawat, and Tyre (al-Bass) – there is no evidence to suggest that it was involved in burial ritual. The tombs themselves provide better evidence for funerary activities. For instance, communal tombs often included spaces where people could gather, such as central chambers, side rooms, vestibules, porches, or sunken courts. Analysis of these spaces does not point to a consistent set of activities performed in the tomb. They varied greatly in size, with some allowing room for little more than a few people and a coffin. Occasionally benches were built in the communal tombs, structures that elsewhere are linked to reclining and banqueting practices. The tomb at al-Awatin (Figure 37) has benches running in front of the loculi along all the walls of the chambers. In one case at least (Figure 37b), the bench was used for the placement of oillamps. In other tombs, the benches also served as tables for pottery vessels, glass bottles, and lamps. Benches in an early tower-tomb in Palmyra are interpreted as burial beds, holding the human remains, rather than banqueting furniture. In Tyre, benches aligned the exterior of the walls facing the road, as well as the walls inside the funerary enclosures. The function of the benches, thus, appears to have varied across tomb type and region. Chapter 3, furthermore, demonstrated that the discovery of food remains or vessels related to food preparation, serving, and dining was limited in the artifact assemblages of Syrian tombs. There is little to suggest a common practice of dining in the tomb. Figure 38 depicts the distribution of grave goods in Tomb C in Palmyra. Directly upon entry, to the right of the steps, a well was dug. Incense burners and plastered water pots were discovered in the main chamber. In the back of the tomb, another water pot was deepened into the floor (Figure 38b), this one with remains of the pouring cups still inside. Incense burners were also placed on top of and next to the sarcophagus in the side room on the left. The discovery of these finds indicates that offerings – liquid or burnt incense – occurred in the tomb. Other Palmyrene tombs also regularly yielded incense vessels, wells, and pots plastered to retain liquids, as well as small altars and burnt remains. This ritual activity was well attested in the tombs of Palmyra. Kaizer has drawn attention to the fact that both the burning incense and libation offers were primary activities of religious ceremonies in the city.6 In Palmyra, therefore, explicit connections existed between religious rituals and funerals ritual. Lack of archaeozoological research prohibits us from identifying sacrificial meat placed inside the tomb after blood sacrifice, another common feature

6

Kaizer 2010, 26. See also Kaizer 2002, 177.

PROTECTION, POLLUTION, AND THE BODY

38. Grave goods in Tomb C (T. 32), Palmyra

153

154

FUNERARY BELIEFS

in Palmyrene religious practices. As already mentioned, neither the spatial configuration of the tombs nor the find assemblages suggest that banquets following the ritual slaughter of animals occurred in the tomb. This does not mean that banquets were not part of the funerary rites, and they may have been celebrated outside the cemetery. Offerings, however, were left in the tomb, and thus had a function in the actual burial space. Before discussing these possible functions, let us look outside Palmyra for similar evidence pertaining to liquid offers and the burning of incense. The often cursory descriptions of the contents of the Syrian tombs present real obstacles in exploring this topic, yet some evidence is available. A mausoleum in Bosra had a drilled hole in the closing slab of a loculus, interpreted as a libation hole (T. 96). Libation holes are common in Roman Italy.7 A cistern was dug at the end of the staircase leading to the hypogeum at al-Awatin (Figure 37). The abundance of basins and canals throughout the al-Bass Cemetery in Tyre could indicate the importance of water in mortuary ceremonies, although their use for nearby gardens cannot be excluded, nor their possible Byzantine date. Sachet has drawn a connection between water offers and the cisterns, pits, basins, and aqueduct that ran through the Nabataean cemetery tombs of Petra in Jordan. She notes that water was considered important as refreshment for the dead.8 Better evidence from Syria itself comes in the form of altars: small stone and plaster altars were discovered in tombs at Tyre, Hama, and Beirut, and perhaps Djel el-’Amed and Frikya (Figure 39). Sarcophagi sometimes also included an altar-shaped protrusion. Chapter 3 listed a small group of items that was associated with the burning of incense and libations, including simpula and thymiateria. The last piece of evidence may come from the small bottles found in abundance in Roman tombs. Perfumes and oils are mentioned as regular offerings in Rome and the Nabataean world (see p. 100). Their containers, small glass and ceramic bottles, are one of the most common finds in Syrian tombs. Therefore, it is possible that libations and other offerings, as well as the burning of incense, were also a component of funerary ritual in the rest of the province. It is unfortunately also possible to over read this evidence, which is limited – and that offerings were unusual, rather than common practice.

Offerings? Let us entertain for a moment the notion that the ritual offering of liquids and other items was performed at the tomb. What, then, was their purpose, and to whom were they directed? The varied location of finds related to offerings in the tombs does not point to a single or similar recipient. Some artifacts were clearly associated with individual burial spots, such as altars on sarcophagi or the libation hole cut through the closing slab of a loculus. Most (perfumed) oil and 7

For example, Lepetz & Van Andringa 2011.

8

Sachet 2010.

PROTECTION, POLLUTION, AND THE BODY

A

B

39.A l-Bass Cemetery (Tyre). A: Altar in front of M. 970 (Complex 14). B: Sarcophagus 879 (Complex 23) with altar-shaped protrusion

155

156

FUNERARY BELIEFS

other ointment bottles were placed by the corpse inside the burial spot. Incense burners and the evidence for water features generally, although not exclusively, come from the central areas of the communal tomb. Some offerings were, thus, spatially related to individual deceased, whereas others were not seemingly tied to specific persons. Epitaphs refer to two groups of recipients for the offerings. The first are the gods. Military inscriptions invoked Roman chthonic deities, the Manes. One from Baalbek explicitly asks to pour libations to the Manes, something that may be illustrated on a military stele from Apamea depicting a man holding a dish above an altar (T. 97, 98). The invocation of the spiritual world was explicit on military tombs, sometimes including veterans and family members. Debates exist in the study of Roman religion about the relationship between the dead and the Manes, and the question of whether Di Manes were in fact the collective spirits of the deceased.9 In any case, they are no longer the living, and as spiritual beings, the Manes could be asked for help, a request that could be enhanced by sacrifice and libations. Direct invocations of the divine world were rare in civilian and non-Latin epitaphs from Syria, and the few instances that exist are hard to interpret. Two Greek texts mention underworld gods, who appear to have been requested to protect the deceased. Decoration of a lead sarcophagus from Homs depicted and referred in Greek to the Olympian Zeus (T. 99). The gods were also called in to shield the (deceased) slave Phaselos against slanderers in an inscription in Bosra (T. 100). The Tomb of Abedrapsas in Frikya, finally, included a long epitaph describing the deceased’s personal relationship to the god of Arkesilaos, presumably his aid to success in life (T. 60). No clearly defined role of deities in funerary ritual emerges from these examples. The second possible recipient of offerings was the deceased him- or herself. A Hellenistic inscription from the Antioch area describes a libation and offering of a flower garland to appease the soul of the deceased.10 We have already seen that the deceased, or the corpse, needed protection and appeasement. It is possible, although at this stage difficult to prove, that offerings aimed to reach the dead, as individuals (through placement in or close to the burial spot) or as a collective group. Sacrifices to an unknown recipient were carried out at the mausoleum at Me’ez on the Limestone Plateau (193 CE), consecrating the earth and seemingly preventing disturbance or sale by inheritors (T. 95). Protection of the tomb from unwanted reuse was a clear concern, according to the epitaphs, some of which included curses (p. 143). Read in this way, the offerings were intended to protect the tomb and its dwellers. Protective deities, as well as the deceased, would be the recipients of these offerings. 9 10

Cf., Lepetz & Van Andringa 2011, 110; Rüpke 2007, 69. IGLS III/2, 501–502 (# 912); Parlasca 1982, 7 (#5.1).

PROTECTION, POLLUTION, AND THE BODY

From Body to Tomb The body was a primary focus of funerary ritual in Roman Syria. Dangerous and vulnerable, it drew together several components of funerary beliefs centered on pollution, protection, and appeasement. Coffins, spells, grave goods with magic properties, offerings, and the tomb building itself, sometimes covered in apotropaic symbols (see later), may have served to protect the physical remains of the deceased. Before the funeral, the corpse was dressed and adorned with jewelry, and these items accompanied the deceased to the grave. Sometimes, additional finds such as glass bottles or coins were placed close to the body. The find location of the items associated with the corpse indicates that they were placed there during or right after the funeral. In this phase, the body required the fullest attention. The lack of stratigraphic data does not always allow for distinguishing between various phases of usage, but overall people do not seem to have returned at a later stage to place items by the remains of the dead. In other words, the body was central during the first stage to the funerary ritual, and perhaps lost focus over time. Such reading of the ritual would fit the classical description of the funeral as a rite of passage (see also p. 14). After separation – death – rituals focused on giving the deceased a place in the world of the dead, whatever that entailed. The mourning family was polluted, and part of the funerary ritual was also aimed at them,to ensure their successful reintegration in daily life.The activities during this so-called “liminal phase” are well known from elite burials in Late Republican and Imperial Rome. Various rituals were undertaken at the house, the funeral procession, and the tomb. In particular, blood sacrifice and offerings performed during the funeral aimed to assist the transition of the recent dead into a deceased, and to fix him or her in this new realm. Failure to do so would leave spirits restless and potentially harmful to society. A banquet at the end aimed to reincorporate the family into society, to end their pollution, and to celebrate the successful integration of the deceased. Annual festivals were held to remember and honor the dead, and to ensure the pacification of those spirits that remained restless.11 The liminal phase was a particularly dangerous period in Roman customs, due to the polluting nature of the corpse and the risk of incomplete transformation of the deceased. Looking back at Syrian rituals, the focus on the body in the first stages of the funerary ritual may point to similar concerns. The transformation of the dead could be fraught with problems and needed human, magic, and perhaps divine support. In this stage, the body was kept whole, which suggests that intactness of the body was a necessary condition for the transformation of the social persona attached to it, and its reinscription as a new identity. When this 11

Cf., Scheid 1984, 2007, 168; Rüpke 2007, 69.

157

158

FUNERARY BELIEFS

stage was completed, the wholeness of the body was less important. This would explain how it was possible that family members or legal owners of the tomb could reuse burial spots and dearticulate previous burials. Did the necessity of keeping bodies whole disappear after the successful completion of the transformation? In answer of this question, it is significant that people did not come back to perform more rituals focused on the body, as suggested by the timing in placement of artifacts on or near the corpse. Ritual attention was no longer directed toward the physical remains, which could be moved or heaped in a pile to make room for a new burial. They had lost their memorial function and could be forgotten.12 This hypothesis would also help explain why cities expanded over older cemeteries despite pollution concerns. If the moment of pollution was centered on the liminal stage between death and incorporation, but ended after this stage was completed, cities could be extended over older burial grounds without having to remove the physical remains. The fears over disturbance and resale that are explicit on several epitaphs may, thus, not primarily center on disturbance of one’s bones, but disturbance too soon, before reintegration was completed, or by the wrong people, who had no legal or kinship right of entry or reuse of the tomb. We have not yet discussed the function of installations and items related to offerings placed in the central area of the communal tombs, the incense burners, lamps, water pots, wells, and basins (Figures 37, 38). The fact that these were not physically associated with a buried individual points to different purposes. We do not know whether they were used at the time of the funeral or at other moments, but it may be significant that they were accessible at all times. Examples come to mind here of continuous engagements with the deceased, which are plentiful in the ancient world, not only in the form of annual festivals in Rome or Jerusalem and royal libations in Petra, but likely also of private visits to tombs.13 These rituals could focus on deceased individuals – the integrated persona – but often also on the deceased as a collective – ancestors or underworld spirits. It is possible, if difficult to prove, that the offerings in the central areas of Syrian tombs were directed at the dead as a collective, whereas those placed on the body aimed to ensure the successful completion of the liminal phase.

Cenotaphs What if there are no bones? The heavily decontextualized material record of Roman Syria does make one wonder if commemoration was also possible without the actual grave present. Did each of the thousands of stelae mark the place that held the physical remains of the deceased? Was the fact that 12 13

Cf., Williams 2013, 1. McCane 2003, 37–39; Sachet 2010; Scheid 1984, 2007; Van Andringa et al. 2013, 925–927.

PROTECTION, POLLUTION, AND THE BODY

rock-carvings of persons found in the hills of Lebanon and the Limestone Plateau were not always associated with graves perhaps intentional rather than the result of archaeological recovery? Perhaps stelae and rock-reliefs could also serve as cenotaphs. The epigraphic assemblage yielded two cenotaphs, both from soldiers stationed at Apamea (T. 101, 102). One was erected for a soldier whose body was buried in Catabolum, possibly in Cilicia. The other concerned Aurelius Mucatralis, who died during the “successful eastern campaign” and whose colleague had a stone erected in his memory at Apamea. A third stele, this one from Baalbek, reads: “Comparing the two of you, Dionysios and you, after your deaths, I am searching the first and you, I miss you Libanos. You were both faithful, attached to your master, indispensable secretary, that was you, and he was barber, the unfortunate one” (T. 103). This unusual text could refer to the commemoration of two slaves, one whose corpse was not found or not placed in the grave. More examples of divergences between commemoration and burial are highlighted in Chapter 4. Epitaphs, reliefs, human remains, and the shape of the tomb often gave contradictory information about who was actually buried. In the case of at least two tombs (T. 60, 66), we can almost be certain that figural reliefs and epitaphs refer to people who are commemorated but not actually buried inside (p. 120). Then there is the issue of empty graves. Several loculi in Palmyrene tombs were often empty on discovery and unused according to the excavators. In a funerary enclosure in Beirut (T. 39), the stacked compartments included one that was never used, but was placed underneath a compartment that did receive a burial. This latter burial thus rendered the lower grave inaccessible, and suggests that it was left empty on purpose. It is possible that these empty graves served as symbolic graves for someone who died elsewhere. Combined, the evidence at least suggests that commemorative practices could center on deceased individuals without spatial association with their corpse. A series of commemorative inscriptions using the Semitic term “nefesh” may be seen in this light. The term “nefesh” (“nfs”) refers to self/soul and perhaps to the spirit of the deceased. Only a few instances are recorded in the sample used in this book (T. 104–107), but the term occurs on epitaphs and sometimes in literary sources across the Near East, including the Arabian peninsula, from the Iron Age to Late Roman times. Such widespread occurrence has led to an extensive range in interpretations: “nefesh” in the scholarship can denote a funerary monument, the name and patronym of the deceased on an epitaph, a rock-relief of a person, or an aniconic shape such as a cone or pyramid-shaped top.14 In Palmyra, the term on one occasion refers to the tower 14

Clauss 2002; Colledge 1976; Griesheimer 1997a, 170; Kropp 2013; MacDonald 2006, 288–290; Mouterde 1951.

159

160

FUNERARY BELIEFS

of a tomb, and, at a later stage, appears on portrait reliefs. The precise chronological development and the local interpretation of “nefesh” require extensive epigraphic and spatial research, but important for our discussion here is that the term points to a potential split between the container of the bones (the grave) and the location of commemoration. In Petra and other sites in Jordan, “nefesh” inscriptions memorializing the deceased can be found aligning the road, away from the cemeteries.15 In this context, “nefesh” inscriptions and the associated cone-shaped reliefs served as cenotaphs. The example from Petra also illustrates a wider spatial context for commemorative practices than the cemetery alone.

THE VISUAL SETTING

A different source of evidence for funerary beliefs is the decoration of the tomb and coffin. As we have seen, Syrian graves could be elaborately decorated, leading one to ask whether the adornment of the tomb provides information about rituals that took place in the tomb, or more generally beliefs about what happened after death. After all, this decoration includes banqueting deceased, gods and goddesses, and various symbols that refer to religious ceremony, such as garlands and bucrania. The issue is a thorny one in the scholarship of mortuary practices. The interpretation of the relationship between tomb-decoration and funerary beliefs is part of an ongoing debate about the function and meaning of decorative patterns. Take, for instance, the occurrence of scenes from Greek myth in various Syrian funerary contexts (p. 67; cf., Figure 40). These have been interpreted in various ways, ranging from proof of the influence of Greek beliefs concerning death and afterlife on Syrian practices, to mere displays of familiarity with Classical mythology intended to heighten prestige.16 Some scenes, such as the abduction of Persephone or the Trojan War, had clear associations with death. Depictions of gorgons and sirens were imbued with apotropaic connotations, and perhaps aimed to prevent harm from being done to the deceased. Winged Victories elevated the dead, or added a cosmic setting for the deceased.17 But, how can we be certain that this is how they were read in a funerary context? Zanker and Ewald have recently put forward an attractive interpretation. They see certain scenes on funerary coffins around Rome as pertaining to the mourning process of the living community.18 The depictions of violence, abductions, returns from the underworld, elevation of the dead, and mourning provided models of proper death and victory over death, meant to comfort the bereaved. The deceased were sometimes inserted, 15 16 17

Kühn 2005, 253–62 (see also Kropp 2013, 216–223). For the first, see Olszewksi 2001; for the second, see Kaizer 2010, 27. 18 Cf., Drijvers 1982. Zanker & Ewald 2004. See also Bielfeldt 2003 and Gessert 2004.

THE VISUAL SETTING

A

B

40. Mythological scenes. A: Scenes from the life of Achilles on Sarcophagus 954 (Complex 16) from Tyre. B: Depictions of Heracles and Alcestis (right) and the abduction of Persephone with Hermes (back wall), Tomb 1 at al-Awatin (Tyre)

in relief, into the scenes. Contextual and literary evidence that allowed Zanker and Ewald to construct their hypothesis is absent from Syria, however. It has also not been investigated what such myths and stories symbolized in a Syrian, i.e., in a non-Greek and non-Roman, setting. The sample of tombs in this book does not point to particular preferences or pattering in the use of myth and deities in tomb adornment. Furthermore, the most elaborate examples of Greek myth were carved on the sides of imported marble sarcophagi found at Tyre. These came to the Lebanese shore pre-carved in Attica, and may reflect Attic rather than Tyrian beliefs. As expensive imported coffins, they no doubt

161

162

FUNERARY BELIEFS

served as status symbols, decked with foreign or trendy imagery, testifying both to the wealth required to acquire such heavy marble and to the education of the burying community, which was familiar with the stories. Yet, how the life of Achilles and Bacchic scenes were tied to Tyrian funerary beliefs, if they were at all, remains difficult to reconstruct. In the use of decoration on Syrian tombs and coffins, nevertheless, a few aspects stand out. First, depictions of myth and deities are rare, and restricted to certain regions. They may reflect local ideas about death and ornamentation, rather than cross-provincial beliefs. The same appears true for most other forms of tomb decoration. Chapter 2 illustrated how each region developed its own symbolic language, and these were likely tied to locally specific traditions and tastes, the study of which lies outside the scope of this book. More commonplace across the province was the depiction of the deceased and their families: reliefs and round sculpture of busts and full-length persons. These images served commemorative functions, as discussed in Chapter 6. But did they also have a ritual function? The placement of lamps and incense vessels around one funerary portrait in Palmyra (T. 32) may suggest that the portrait itself was the recipient of offerings (Figure 41). At this stage, this intriguing find appears limited to Palmyra, and more work is needed on the role of figural imagery in Roman Syria in general in order to address the potential ritual significance of portraits. A second common type of decoration in Syrian tombs consists of references to religious ceremonies. Depictions of garlands, flowers, and wreaths, and less frequently animal skulls or bucrania, were added in relief and paint to the coffins, tomb walls, and funerary stelae (Figure 42; cf., Figure 17). Altars were represented in relief and some stelae had the shape of an altar. These examples indicate that symbols connoting sacrifice, offerings, and festive decoration of sanctuaries were often chosen to adorn the tomb. The deceased were placed in a venerable setting of religious ceremony. Perhaps the reclining or banqueting reliefs point in similar directions. Depictions of a reclining man holding a cup next to a seated woman and surrounded by other family members were common in Palmyra. As already mentioned, banquets formed a part of religious ritual, and it is tempting to suggest a link between the two. The deceased symbolically and eternally participated in religious rites. The depiction of some Palmyrenes with a priestly hat, the modius, indicates that it was important to stress the religious association of some of the occupants. A few examples from Palmyra appear to depict a religious ceremony. These are the sarcophagus reliefs that portray people involved in libations, offerings of fruit, and blood sacrifice, and we find a small collection of similar images (sacrifice, libation, banquets) from other contexts in Syria. The Tomb of Abedrapsas (T. 60, Figure 29) combines the two elements: Abedrapsas and Amathbabea enjoy a banquet with their daughter, as depicted on the back wall, which is also crowned by a frieze with a row of figures next to an altar. Although no images are available, a

BURIAL DIFFERENTIATION

41.Gr ave goods surrounding funerary portrait in Tomb C (T. 32), Palmyra

drawing in Butler’s publication indicates a small altar by the entrance that was covered with reliefs of musicians. The decoration and architecture of this tomb conflate various steps of religious ceremonies, including music, perhaps a procession, offering, and banqueting. A possible cross-provincial pattern, thus, is using imagery to denote a religious setting. Whether this is actually the setting of the funeral itself remains speculative.

BURIAL DIFFERENTIATION

Having established the core components of Syrian funerary ritual, revolving around the body, pollution, and perhaps offerings, I now turn to those areas where distinctions were drawn within the buried group. The previous chapters have demonstrated the variety in burial patterns that characterized the Roman period. People were buried in different tombs and containers, and accompanied by different grave goods and epitaphs. Chapter 4 correlated these differences with various identity groups, based on sex, gender, age, profession, and lineage. Here, we address burial differentiation in terms of ritual practices. Did certain people receive a different ritual?

163

164

FUNERARY BELIEFS

42. Tomb of Tiberius (T. 7), Beshindlaye (Limestone Plateau)

Who Was Buried? The funerary remains of Roman Syria include only a small portion of the provincial population. The graves of the vast majority of its inhabitants have never been recovered. Did their burial rites diverge to such an extent from the common features described so far that we fail to find physical evidence

BURIAL DIFFERENTIATION

of this part of the population? Let us look at the numbers first. A total of 517 tombs with around 4300 burial spots are included in the cat. 1 assemblage. The low estimate for the cat. 2 collection amounts to 1779 tombs and, applying the average number of burial spots of cat. 1 tombs, approximately 15 000 burial spots. Combined, the assemblage included a minimum of 19 300 burial spots in use from the 1st c. BCE to the mid 4th c. CE. Even when accounting for the frequent reuse of the graves, these tombs only held a portion of the population. In a province with an estimated population of several million people, the number of mortalities each year would have reached around a hundred thousand.19 Thousands of people would have died annually in the thirteen sites and two rural regions selected for this book. This problem of a missing buried population is not restricted to Syria or the Roman period. In most ancient societies, it is unknown how and where the majority of the people were buried. Many tombs simply did not stand the test of time, and in Syria the excavations and survey notes make very clear that many others were deemed uninteresting for the researchers. Furthermore, reuse, as we have seen, was rampant. Yet, can archaeological priorities and reuse account for the millions absent in the funerary record? Textual sources from ancient Rome speak of public crematoria and mass graves in pits.20 These may not be at locations close to the settlement or the cemeteries. In Syria, physical evidence for these practices lacks, but it is highly likely that people were buried in ways that are difficult to recognize archaeologically: outside the formal cemeteries, in trash pits or mass graves, thrown in the sea, or exposed outside the settlement. These people received distinct funerary rites than those described earlier, although it is not possible to say what these rites were, or if, indeed, any care was given to the intactness of the body, pollution rules, or facilitation of the successful integration of the deceased. We can say a bit more about who was possibly excluded from the types of burial described in this book. Graves that lacked sculpture, grave goods, inscriptions, and aboveground markers were most likely to be overlooked or not deemed interesting by archaeologists or epigraphists. In other words, the simplest and poorest graves were far less likely to be published. Chapter 4 demonstrated that children were underrepresented in the burial population. Comparatively rarely did they receive a burial that was archaeologically visible, or were they included in the group of honorees mentioned in text and image. It is likely that they form a large portion of the invisible dead in Roman Syria. Due to the problems in aging skeletons, we cannot say until what age children were excluded from formal burial. However, it appears that the young 19

20

Frier (2000) calculated an estimate of 4.3 million people in Greater Syria (including Palestine) around 14 CE and 4.8 in 164 CE. See also Kennedy 2007, 124. Bodel 2000.

165

166

FUNERARY BELIEFS

and the poor were more likely to be buried in an archaeologically less visible or invisible grave.

Special Cases Two old men in Tomb C at Palmyra (T. 32) were buried with multiple amulets. These are unusual finds in Syrian graves, and the fact that these burials contained more than one may be significant. Did these men need more amulaic protection because of a special status they held in society? In Chapter 4, we discussed how sex and age rarely explain patterns of differentiation in Syrian tombs. There is nothing to suggest that only old men received magic items. Did they perhaps die in a way that needed extra elaboration? Robb has drawn attention to the fact that burial differentiation could not (just) have depended on the social persona of the deceased and his/her family,but perhaps also related to the method of death.21 A particularly violent or untimely death could warrant different funerary rituals. Curse tablets in Beirut and perhaps Tyre, usually placed in the tombs of people who had a premature or violent death, confirm the existence of the beliefs that Robb describes. Perhaps the two men in Tomb C met their fate in a particular way that required more emphasis on appeasement of their post-mortem being. This idea of individuals requiring different ritual, not because of their status in life, but because of their mode of death – premature, violent, or self-inflicted – would help to explain some distinctions made in funerary ritual. In particular, the infrequent placement of highly charged objects in individual graves, such as amulets, gold face covers, incense burners, or cooking pots holding offerings, could be tied to the need for extra protection and appeasement. Yet, it is hard to prove this satisfactorily, and the reasons for the idiosyncrasies may just as well be entirely different. There is one area where the notion of untimely death seems more at place. Children are underrepresented, but not all were excluded from burial. It is in their graves that divergent burial treatments become apparent. Children were placed in jar-burials or small pits in hypogea, types that were rarely reused for additional burial. Higher numbers of grave goods may have accompanied the burial of children, and in particular items of jewelry and coins. Their graves did not usually include embalmment vessels. When (young) children received formal burial, their physical remains were treated differently. Their graves would not be reused and their bodies remained articulated. Relatively many of the longer inscriptions were dedicated to children, explicitly expressing the extraordinary qualities of the child and the sorrow of the parents (p. 126). However, in light of the high child mortality in this period, these 21

Robb 2007, 292.

BURIAL DIFFERENTIATION

were exceptional burials. There was something special about the status of the child or the way it died that demanded funerary elaboration. One inscription from Nisibis recorded the lament of the parents of Lucius who died at the hands of sinful murderers (Figure 33). Perhaps a violent, and thus even more premature, death of a child motivated differential treatment. Such a death would turn upside down the socially agreed conventions guiding the burial of children. Soldiers of the Roman army form another group of special cases (Chapter 4). Their epitaphs explicitly invoke the Manes. The handful of cremation burials originating from Roman Syria perhaps belonged to the military. Setting up a pyre and burning a corpse entailed a completely different set of activities than carrying a coffin or bier to the grave, and goes against the local practice of keeping bodies whole.

Collectivity of Burial Communal and single tombs arose side by side in the cemeteries of Syria. In some aspects, burials in these tombs resembled each other. Both types held inhumation burials, often in coffins and accompanied by items of personal adornment, vessels, coins, and lamps. At the same time, differences in the spatial configuration and accessibility between single and communal tombs allowed for divergent activities and uses of space. Some are directly connected to funerary ritual. For instance, rarely did the single types contain more than one burial. This may not come as a surprise, as pits, cists, and jars in pits were not as accessible for reuse as the loculi or coffins in communal tombs. However, it also means that the physical remains of the occupants remained articulated and were not subject to disturbance caused by new depositions. As with the burial of children, the intactness of the bodies in these graves was prolonged. Further distinctions in concepts of time can be identified in the two types. To the communal tombs, people returned to bury newly deceased members of the community. The construction of these buildings took longer than that of single pits, not in the least because collective tombs received more embellishment. They were also often modified over time, with new rooms, burial spots, and other features being added. Single graves were never modified after use, as far as we can tell, and they operated on a much shorter timescale. In fact, the interaction of the living community with the single tombs and their occupants may have been limited to the moment of construction and the funeral. Collective tombs included a central room or area, and some had side rooms, vestibules, benches, and porches. As discussed in the previous section, we know little about what took place in these spaces, or when. Yet, they offered the opportunity for collective activities and for the presentation of offerings

167

168

FUNERARY BELIEFS

at other times than the funeral. Distinctions in the artifact assemblages, albeit slight, point to additional differences. Single tombs of the civilian population in Syria rarely included items related to incense and liquid offers, which were more common in the communal tombs. Single tombs held fewer items of personal adornment, coins, and lamps, and more unique items and vessels, especially ones made of ceramic and other materials. The assemblages in single tombs were more often idiosyncratic compared to those of communal tombs. What do these distinctions tell us about the burying community? The fact that collective tombs were more labor- and cost-intensive than single tombs could point to economic differences,with the former belonging to a richer part of the population. It is common in mortuary archaeology to equate the energy spent on a tomb with the affluence of the occupant. Yet, numerous examples from ethnology illustrate that people of little means could also receive an exquisite tomb due to their social standing (e.g., religious importance, acts of courage, old age) or manner of death (e.g., in warfare or, as already discussed, by violent or otherwise unusual means; see p. 15). The large communal tombs covered in elaborate decoration were costly to construct, but when averaged out over the number of people buried within them, perhaps they only cost as much per person as a stele and coffin for a single pit-grave. Tombs could also be co-sponsored with several others. In Roman Syria, this was generally not done. When specified, single individuals or a spousal couple usually sponsored the communal tombs in the province. They invested resources to build a large tomb designed to hold their family members, and spent considerably more than those digging single graves or than the family members who were added later. Monumental tombs were not constructed for single individuals, illustrating that the redirection of wealth to the tomb was done in the context of a collective identity, the family. The Philippeion in Shahb¯a (T. 6) may have been an exception, but as noted earlier (p. 25), this was in any case an unusual building. The total numbers of finds in the communal tombs were slightly higher, and included more items in precious metals and stones. Wealthier members of the population seem to have been responsible for the construction of communal tombs, which were destined for members of the founder’s family. Burial in a family tomb could in itself function as a means of displaying economic wealth, or association with a wealthy patron. Pronounced distinctions thus arise between burial in a single versus a communal tomb. Single types held individuals, who were accompanied by a variety of items translating to different attitudes on how to beautify, protect, and appease the body. Communal tombs referred to collective (family) identity, as well as access to labor and resources. The grave goods in these tombs indicate more standardized ways of presenting and protecting the body. The latter could also have been the function of the congregation spaces and, occasionally, house ritual implements.

DISCUSSION

DISCUSSION: FUNERARY RITUALS OVER TIME

The Baalshamin Tomb in Palmyra provides a good example of the high level of continuity in ritual practices from the pre-Roman into the Roman period (see Appendix 1). This communal tomb dating to the 2nd and 1st c. BCE held twelve niches flanking a small corridor in which at least sixty-one individuals were buried. There was a strong emphasis on keeping the bodies intact, through inhumation and burial in wooden coffins. As in the Roman period, reuse was common, and articulated skeletons were disturbed by new burials. Sometimes, the bones were collected and pushed to the side of the grave. The finds discovered in the loculi show the importance of adorning the deceased with jewelry and clothing, and include embalmment vessels and a variety of pottery vessels. The narrow corridor of the Baalshamin Tomb leaves little space for communal activities, but the altar discovered at the entrance and possibly belonging to this tomb provided a place for offerings. The tomb was far – more than half a kilometer away – from the inhabited area.As Palmyra expanded over time,the Baalshamin Tomb came to lie within its urban boundaries. An inscription dating to 11 CE describes the possible purification ritual that followed the abandonment of the tomb (p. 285). The tomb was closed and built over, but not emptied. All core elements of Syrian funerary ritual are present in this tomb. Finds in Hellenistic and Parthian tombs from other sites confirm that keeping the body whole, protected, and beautified was a key requirement. An incense burner was found in a tomb in Beirut and a Hellenistic inscription from the Antioch area describes a libation and offering of a flower garland to appease the soul of the deceased. The few decorated pre-Roman tombs were adorned with garlands, reclining scenes, and bucrania: direct references to religious rituals. These concepts were coded in long-term funerary practices of the region. The same may be true of the separation of the dead from the living, connected to concepts of pollution. Hellenistic graves are placed outside the settled areas of the town. Further east, in Parthian Mesopotamia, intramural burial was more common. This practice may have existed in Parthian East Syria as well, but at this stage there is no conclusive evidence in support of this claim (p. 245). Both the key features of Syrian funerary rituals and the main areas of burial differentiation can be traced back to pre-Roman times. The Examples from the Hellenistic period displayed regional variation, such as a group of painted contents of communal tombs were more valuable in terms of grave goods and sarcophagi when compared to single graves. Jar-burials were mostly restricted to child burials long before the establishment of Roman rule. The shape, decoration, and inscription on a unique collection of military stelae from Hellenistic Sidon diverged from practices related to contemporary civilians (see p. 333). Keeping the body intact, protected, and away from the living, efforts to appease the spirit of the deceased or chthonic deities, and ritual distinctions in

169

170

FUNERARY BELIEFS

the burial of children, soldiers, and those in single and communal tombs were and had been essential in Syrian funerary beliefs. Even the tomb types that were introduced in the Roman period – the mausolea, tower-tombs, funerary enclosures, and sarcophagus groups on pedestals – did not alter the basic aspects of funerary rituals. Can we conclude from this that despite the changing look of funerary architecture and embellishment, and the new location of cemeteries described in Chapters 1 and 2, the way people in Roman Syria treated the dead did not alter significantly? Was the change one of fashion or convenience, facilitated by increased connectivity, which allowed for the spread of tastes, building materials, and sarcophagi, rather than being related to new concepts of burial? Even if this was the case, the mode and timing of changing taste still requires explanation. Chapter 6 focuses on this question. Here, we are concerned with funerary ritual, and despite long-term continuity of funerary beliefs, the new tombs of Roman Syria did, in fact, have an impact on the timing, location, and function of funerary rites.

Time Let us first look more generally at the different chronological phases of burial. Activities at the cemetery were carried out in three more or less distinct phases: the construction of the tomb, the funeral, and post-funeral activities. The first two phases could occur within a short time-frame, e.g., the digging of a pitgrave immediately before burial, or they could represent different time-periods, e.g., in the case of the construction of large communal mausolea. On rare occasions, the epitaphs mentioned when a tomb was constructed relative to the first interment, usually when a son built a tomb for his parents or a spouse for a husband or wife. We can assume that they constructed the memorial for their beloved before their own death, even though some tombs held space for their own burial as well. Other foundation inscriptions did not explicitly state whether the tomb was built during the founder’s life, which seems the most likely option, or after his or her death, using funds set aside for the purpose. The third phase involves activities around the grave when no funeral took place. Examples are celebrations in honor of the dead or returns to the tomb for other occasions. This phase also encompasses the “afterlife” of the funerary construction, when the associated families were no longer present or making use of the tomb, but it still remained a part of the landscape. In Roman Syria, the first and second phases were often elongated when compared to pre-Roman practices. Large, decorated tombs required a longer period of construction. Remodeling of the tomb after the first interments, such as the addition of a painted arch atop a funerary enclosure in Beirut (T. 39) or an exedra filled with sculpture covering burial niches in several Palmyrene hypogea, further extended the construction time. The prominent location of

DISCUSSION

the cemeteries made much of this building activity public. The attention of travelers visiting the towns, fields, or public buildings placed close to the cemeteries was drawn to the active laborers, chiseling, building, and decorating the graves. Transportation of heavy sarcophagi from the harbor of import to the cemetery was another time-consuming and highly visual affair.22 The addition of an aboveground marker allowed for the permanent visibility of the tomb. In earlier times, the community would mostly have interacted physically with the tomb when it was opened and thus made visible. In the Roman period, these tombs remained conspicuous whether a funeral took place or not, a feat heightened by the roadside location.Cemeteries were included in routine actions such as travel to the gardens, water supply, or other towns. Communal gatherings of a civic or religious nature, e.g., attending the theater or visiting the sanctuary, often occurred in close proximity to the tombs. The tombs were meant to be seen on a regular basis, and could actively be used to present social order. Following Barrett, visible tombs allowed people to write their origins into the past, thereby using physical grounding and visibility in the landscape to legitimize their power and position.23 In strategies of self-definition, the burying community could at all times point to the graves of deceased members of their group. Tomb buildings attained an active role in the legitimization and renegotiation of the social order, and the dead – or, some dead – continued to play a role. The amount of time dedicated to tomb-construction and post-funeral activities varied greatly. Manipulating time, or the burial pathway, arose as a prominent way of distinguishing between people in Roman Syria. If longer pathways were reserved for people of greater social significance, the monumental display tombs belonged to those who were most prominent in Syrian society, or had aspirations to such a position. In some cases, the life of a tomb was extended beyond the death of its last member. Reburial in the Byzantine period indicates that communities continued to interact with the tomb. This may not have been the intention of the original sponsor, but the permanence of these buildings meant that they kept playing a role in the local community, sometimes up to the present age. These tombs were weaved into the cultural memory of a region.

Commemoration The amount of resources spent on the tombs increased in the Roman period. This was not solely a result of the increased wealth of the provincial population, although it may have been related to it. Building activity in the towns of the 22 23

See de Jong 2010 on the imported sarcophagi in the al-Bass Cemetery in Tyre. Barrett 1990. See also Barrett 1996, 399.

171

172

FUNERARY BELIEFS

province, the greater availability of imported products, and the intensification of agricultural yields are sometimes connected to a rise of prosperity in Roman Syria. The latter two could also be the result of the extensification of trade networks and population growth, whereas the first seems to have been limited to only a small group of people in the province. In any case, an influx of resources does not explain why they were directed toward tombs, instead of, for instance, domestic architecture. The evidence, albeit slight, does not suggest great splurging on mansions and villas in Syria before the 3rd c. CE. Artifact assemblages in the graves were somewhat richer than before, but this was not commensurate with the resources poured into tomb architecture. In other words, there was something about the tomb building that was deemed important to elaborate. As illustrated in this chapter, older components of funerary beliefs concerning the treatment of the body and the use of tomb space to protect the deceased and express communal or non-communal identity remained active in the more elaborate tombs of the Roman period. Change focused on the way the memory of the deceased was presented, and thus on commemorative practices. One prominent message was the advertisement of wealth of the founder(s) of the tomb or the association with such a person. The permanence of funerary architecture discussed in the previous paragraph may in part explain why tomb-walls served as public statements about access to resources and labor. Other means of conspicuous consumption, such as donning a rich garment or sponsoring a festival,mostly had a short-term impact.They could be very affective in drawing attention to the status of an individual, but they did not endure for a long time. Tombs, on the other hand, were there to stay, and provided a long-term means of expression, and an opportunity to speak beyond the grave. To some extent, this is true for architecture in general, and particularly for the large structures erected in the towns of Syria sponsored by wealthy individuals. To investigate why funerary architecture attained these new roles, we focus on three aspects that were largely unique to the cemetery. Listed in order of popularity, these are: the expression of regional identity, the emphasis on the individual, and identification within family groups. Tombs played an active role in defining the collective identity of a region, or at least of those who received a proper tomb. The cemeteries united people on a regional level and presented the communal past of the group, through some form of regionalization of tomb shape, decoration, or furniture. Some people were remembered as individuals. Names and faces of (mostly) adult men, but also women and soldiers, often adorned the tomb. The memory of these people was preserved for longer than that of the other deceased, whose individual identities more easily merged into collective or indistinct identities over time. People could return and point to the graves of specific family members, whose memories could be resuscitated in support of claims in the present. The figural depictions did not present these individuals as they died, but in a manipulated form. The dead were beautiful,

DISCUSSION

adult but otherwise ageless, wore rich outfits, and sometimes participated in ritual activities such as banquets or offerings, or were surrounded by symbols referring to such activities.Physical decay was overruled by sculpture.The trend of individualization of burial spots occurred in the large communal examples and in the single graves. A diverse group of Syrians was thus commemorated in this form, not solely the wealthiest or the male. This was not the case with the third new aspect of commemoration: identification within the family. A smaller but not insignificant group of people in Syria directed resources to the embellishment of the family grave. Inscriptions and group sculptures commemorated this fact explicitly and highlighted which family held claims over the tomb. Tombs were not marked solely by a family name, but by the name of the founder, whose relation to the other occupants was often defined. In other words, family tombs represented a way to connect a founder to other members of the family, and vice versa. Before the Roman period, a family’s identity was probably on display during the funeral of one of its members, when the family gathered at the grave, and perhaps during other collective commemorative occasions. The visible and marked family tombs of the Roman era allowed for reference to and exploitation of family identity at any time. Changes in commemorative practices were not the same for all occupants of the burial grounds of the province. Children were usually not commemorated as individuals or family members, indicating that their memories did not function or perhaps were less affective in laying claims for the present. The wealthiest families in the province were commemorated as a collective, or this was the message that the tombs aimed to emit. Families of less affluent persons were probably of equal importance for the upkeep of the grave, but family identity was not permanently inscribed on to the tomb. The memories of men and women were manipulated and individualized, but the commemoration of collective family identify usually focused on male founders or heads of the family. Distinctions between people thus rested on whether their tomb could enter the collective memory of a community or whether the funeral was the last or only collective aspect of their burial.

Changes in Commemoration during the Roman Period Changes in commemorative practices converged in cemeteries across the province in the 1st and 2nd c. CE. Although the tombs of each region appeared different in outlook, they were the result of similar transformations in funerary practices. Commemorative practices did not remain stable throughout the Roman centuries. By the end of the 3rd c. CE, Syrian cemeteries functioned in different ways when compared to previous centuries. This process started earlier. After the first half of the 2nd c. CE, monumentality and diversity overall decreased, and became more regionally defined to certain, often rural areas,

173

174

FUNERARY BELIEFS

such as the Hauran, the Limestone Plateau, and perhaps the Middle Euphrates. Construction – although not burial – in particular slowed down in urban cemeteries.It continued in the rural areas,and some regions witnessed a new phase of elaboration of funerary architecture in the Byzantine period.Nevertheless,after 150 CE, tombs in general lost part of their previously described commemorative function. Less emphasis was placed on the elaboration of the tomb, as fewer were inscribed, lavishly decorated, or built large. The effectiveness of funerary architecture in defining a community and being used in local strategies had reduced. The memory of families of ancestors was no longer exploited in this period. The use of Greek, or civilian, inscriptions dropped after the 2nd c. CE, and after the mid 3rd c figural sculpture and inscriptions declined in popularity. By this time, the practice of individual commemoration declined sharply, and manipulated individual identities of the deceased lost their prominence. One can argue that this was the result not only of changes in commemorative practices but also of a lack of funds in the later 2nd and 3rd c. CE, which caused a poverty of funerary architecture. Economic decline is well recorded in the 3rd c. CE elsewhere in the Roman Empire, but it remains to be seen how much it impacted the availability of resources for Syrians. Furthermore, around the time of the diminished elaboration of funerary architecture, houses may have become increasingly elaborate and festivals were more prominent.24 It was thus a redirection of resources rather than, or in addition to, poverty that left the tombs unembellished. Public display of economic elite status had moved to different arenas, whereas the rest of the population had become impoverished or no longer regarded the cemetery as appropriate for expressions of identity. These patterns of diminished elaboration of the tomb partly obscure another development of the later 2nd c. CE and beyond. The regions and cities of Syria started to follow different trajectories. In Palmyra, the number of monumental tombs dropped sharply after 150 CE and stopped entirely after 236 CE. No new tombs were built in the al-Bass Cemetery of Tyre in the 4th c. CE, although the older graves continued to be used for several centuries. There is also not much evidence for elaborate construction in Baalbek and Beirut. In Bosra and the Hauran, building activity in the burial grounds, on the other hand, persevered into the Byzantine age, albeit on a reduced scale. Elaborate tombs arose in the town of Homs and in the surrounding countryside, but the practice of adding inscriptions decreased sharply. The number of visible tombs on the Limestone Plateau declined after 240 CE, but rose quickly again in the 5th c. CE. In other words, the strong patterns of convergence of the 1st and early 2nd c. CE gave way to divergent trajectories by the mid 2nd c. CE. The role of tombs and their function in commemoration was no longer the same in the villages and cities of the province. 24

Butcher 2003, 229.

CHAPTER SIX

THE GLOBAL AND THE LOCAL: ROMANIZATION, GLOBALIZATION, AND THE SYRIAN CEMETERY

S

yrian cemeteries are testament to the strong trends of convergence in funerary customs, peaking in the 2nd c. CE. Even though cemeteries in the province displayed great diversity and regional tastes, they also reveal similar attitudes to display, representation, adornment, and collectivity of burial. These burial grounds did not look the same, but the choices regarding the construction of the tombs operated on the same principles. In fact, diversity was one of these principles, as tombs were designed according to regional fashion. This chapter discusses these moments of convergence in funerary customs, as well as the return to more divergent patterns. In order to interpret the converging trends in Syria, it is necessary to step outside the provincial boundaries. Earlier scholarship on Roman Syria often fails to do this, and instead stresses the local continuities. As discussed in the Introduction, in these narratives Rome often appears as an absent ruler, whose impact remained minimal in Syria. The material record itself may point this way. If one expected to find the brick tombs of the Via Appia or epitaphs erected in Latin by freedmen, to name two “typically Roman” customs, one would be disappointed.Yet,Syrian cemeteries did not develop in isolation from the cultural and socio-political context of the time. Rather, the moments of relatively rapid change in local ritual, as demonstrated with funerary practices throughout this book, need to be understood in the circumstances of the time. The not-so-“Roman”-looking material record, in fact, holds clues about Syrians as part of the Roman Empire. Take, for example, the popularity of funerary 175

176

THE GLOBAL AND THE LOCAL

portraits in Roman Syria. Figural reliefs on stelae had pre-Roman forerunners, and one could explain the widespread occurrence of portraits on Roman tombs in Syria as emerging out of older practices. In other words, these reliefs were tied to local traditions. The Roman period witnessed an increased production of such stelae, as well as their expansion to new regions of the province. The problem with such reasoning is that the popularity of funerary portraits was not limited to Syria, but, instead, extended across the Roman world. Portraits represented new ways of displaying and advertising individual and group identity for many provincial communities, as well as people in Rome itself. The use of funerary portraits in Syria, despite having pre-Roman antecedents, cannot solely be considered a local phenomenon. Research perspectives restricted to the Syrian context obscure how Syrians aligned themselves with empirewide trends in commemoration, display, and taste. This chapter illustrates the importance of contextualizing Syrian funerary practices in a wider (Roman or global) perspective. We concentrate on three patterns of convergence that were tied explicitly to the larger Roman world: the location of urban cemeteries in connection with urban planning, the concept of individual display, and stylistic influence on architecture and decoration. The first section highlights these trends. Traditionally, scholars would associate such developments with Romanization, or the cultural influence stemming from the Roman center. This section discusses the limitation of Romanization models and turns to the field of globalization. Globalization studies here are not meant to replace Romanization, but rather are used to reflect on and interpret the entanglement of local and global developments in the empire. In the sections that follow, Syrian funerary customs are considered in light of wider developments in the Roman world. We look at the connection between civic identity and the burial ground, the emphasis on personal display, and the use of new styles to express Roman as well as local identity. The next sections discuss the timing of change and the place where we encounter some of the earliest changes: Palmyra. The conclusions address the end of the homogenizing trends in Syrian cemeteries and the emergence of regionally divergent trajectories. PATTERNS OF CONVERGENCE

In Chapter 1, we encountered the first example of how Syrian funerary customs were tied to larger trends in the Roman world: the standardization of location of cemeteries directly outside the urban boundaries and alongside the main roads. We discussed how the fixation of location of the necropoleis was correlated to a general program of urban redevelopment, during which Syrian cities acquired a new set of buildings and infrastructure. The peak of construction was in the 2nd c. CE, by the end of which cities across the province had

PATTERNS OF CONVERGENCE

a degree of standardization in planning and outlook, monumentalized public spaces, and upgraded infrastructure. The location of the cemeteries was included in this restructuring of the urban landscape. Cities across the Roman world underwent a similar building phase. Starting with Italian towns during the Late Republic, the phenomenon extended to provincial centers in the Augustan period, the public areas of the town received monumental upgrades. Not every town in the empire underwent this phase, and some witnessed little investment or were even abandoned. Yet, it was widespread enough to be considered an empire-wide phenomenon. Woolf has called the end result a (relative) homogenization of the cityscapes across the empire.1 The placement of cemeteries as part of the redevelopment of the (sub)urban landscape is rarely discussed on a (supra-)provincial level, but similarities with the Syrian case exist. This applies in particular to the roadside location and emphasis on aboveground display. Tombs flanked the main roads to the capital of Rome, as well as to other towns in Italy and to Roman colonies.2 By the 1st c. BCE, this practice found its way to urban sites in the provinces. In some regions of the empire, roadside burial grounds pre-date the Roman period, yet the standardization of this location is explicitly tied to the general phase of urban redevelopment in the Roman era. I will come back to this issue of origin in the section on globalization. By moving the burial grounds to the well-traveled public spaces outside the cities, Syrians participated in an empire-wide trend. The second way in which funerary practices illustrate the integration of Syrians into the Roman cultural world is through the rising popularity of individual display. Tombs often carried the names and faces of their owners and occupants (Figure 43). Not every grave contained such individual markers, but the phenomenon was widespread across the province. In content and language, the inscriptions varied, as did the style and mode of depiction in figural sculpture. They shared an emphasis on presenting the deceased as an individual and providing biographical notes about them. Syrians were not the only ones doing this, and we can trace the popularity of portrait reliefs and personal epitaphs throughout the Roman world. Wellknown examples come from the cemeteries around Rome, where in the 1st c. BCE relief groups with half- and full-length figural depictions accompanied by inscriptions marked the tombs of freed-persons. This trend expanded to elite graves in other areas of Italy in the late 1st c. BCE, and emerged, among other places, in Baetica and Gaul around the same time. The famous mummy portraits, some including inscriptions with (family) names, entered Egyptian tombs between the 1st and 3rd c. CE. Other types of funerary portrait gained popularity between the 2nd and 3rd c. CE in the Beth Shean area in 1

Woolf 2000, 122.

2

Von Hesberg & Zanker 1987; Zanker 2000, 30–31.

177

178

THE GLOBAL AND THE LOCAL

43. Limestone bust of Menophilia daugther of Diodoros (Hama, T. 66)

Palestine.3 The popularity of individualizing funerary space is even more apparent when tracing the expansion of inscriptions. Their use can be connected to the so-called “epigraphic habit,” or the spectacular rise of use of inscriptions in most areas under Roman control in the 1st and 2nd c. CE, peaking around the turn from the 2nd to the 3rd c. CE (see p. 141).4 Epitaphs represent the majority of this corpus. Combined, the popularity of portraits and inscriptions illustrates that the concept of personalizing space for burial became an empire-wide trend. In origin, it was not a Roman practice per se, as individual markings had adorned earlier graves in, for instance, Egypt and the Greek world. These regional traditions converged into a common trend in the 1st c. BCE and CE, and rose in popularity in, among other places, the province of Syria. The third and final trend of convergence concerns the adoption of new shapes, architectural elements, and forms of decoration that found their origin closer to Rome, or matched trends found across the empire. The portrait 3

4

Rome: George 2005; Kleiner 1987. Baetica: Jiménez 2008, 19. Gaul: Woolf 1998, 91–98. Egypt: Borg 1998, 40; Walker 1997. Palestine: Skupi´nska-Løvset 1983. Bodel 2001, 30; MacMullen 1982; Meyer 1990.

PATTERNS OF CONVERGENCE

busts, for instance, not only demonstrate the new importance of ideas about individual display current in the Roman world, but sometimes directly copied elements from busts found around Rome. Syrian men donned togas and rested their arms in slings. Women were depicted raising their hand to their face or veil, a gesture common to popular representations of elite women (Figure 43; see also Figures 30 and 31). Another example is found in tomb architecture and adornment. Chapter 2 drew attention to the fact that the shape and decoration of new tombs in Syria were sometimes modeled on trends found elsewhere in the empire. Mausolea in Palmyra and the Limestone Plateau, for example, followed Roman traditions in sacral architecture, and perhaps copied elements of theater design. Paintings of Greek myths, scenes with banquets or reclining individuals, and garlands and bucrania not only adorned Syrian graves but also were popular motifs in the Roman world. Some of these motifs came out of pre-existing Near Eastern iconography, yet their occurrence and dissemination in Roman Syria was closely connected to a process of convergence that was typical for the Roman period. In order to uncover these processes, we need to delve deeper into the issue of culture change. Such a discussion in a Roman context has to start with the concept of Romanization.

Romanization? Transformations in the provincial communities associated with Roman rule are often interpreted as a form of Romanization. The concept, literally meaning the “making Roman” of people in the provinces, involves the connection between changes in the material and textual record with the incorporation of a region into the empire. It describes an acculturation process well established across Roman territory, although its depth, purpose, and mechanisms remain highly debated.5 Its basic premise is as follows: Sometime after conquest by Rome, provincial communities incorporated new goods and ideas into their local habits. The chronological overlap of these changes with the extension of Roman power and the fact that the new elements compare to those found in Rome led scholars to connect the changes directly to Roman rule. The phase of urban remodeling already discussed, for instance, partly followed the characteristic city plan of Roman colonies, and the use of inscriptions and sculpture to display biographical information about the sponsor of a building or tomb in public is well attested in the capital city. Both phenomena, when occurring in the provincial territories, therefore, are seen as examples of Romanization. Scholars often argue that the impetus behind the process came from provincial urban elites who adopted new goods and ideas as part of their legitimization strategy. The employment of new symbols associated with Roman rule 5

For an overview, see de Jong 2007; see also references in the rest of this section.

179

180

THE GLOBAL AND THE LOCAL

helped ensure the maintenance of – or their aspiration to – positions of power. Transmission of the new cultural elements occurred through enrollment in the army, proximity to Roman colonies, and new forms of education. How nonelites and those living outside cities came to “Romanize” is a trickier topic. Some reconstruct a simple emulation process, whereby elite fashions trickled down the ladder of socio-economic status. Others argue that these populations underwent limited culture change. Soon after the introduction of the concept of Romanization into studies of the Roman world, archaeologists and historians started to point out the problems with such acculturation models. We can use the example of Syria to highlight their main drawbacks. The first concerns the fact that elements that were new in some provinces of the empire were not so new in others. As already discussed, figural sculpture and funerary epitaphs existed in pre-Roman Syria. This tradition was tied to local practices of the 6th and 5th c. BCE, as well as to influence from the Greek world that increased late in the 4th c. BCE (see examples on p. 333). Its presence in the archaeological and epigraphic record in the Roman period was not entirely novel. Moreover, cultural life in Rome in the 3rd and 2nd c. BCE was inspired by trends current in the Hellenistic world, such as in Greece, Alexandria, and Asia Minor. These regions exerted influence on Rome at the same time as on Syria. Rome, therefore, cannot have served as a direct model for the changes attested in funerary practices in Syria. To explain this phenomenon, scholars often propose a division between the Roman territories that had previously been part of the Hellenistic world, i.e., Greece, Egypt, and most of the Near East, and those that had not. The former, also called the “Roman East,” barely Romanized according to this model, and so followed a different trajectory from the “Roman West.”6 Yet, as we have seen, changes in the material record of Roman Syria were rather unambiguous. By the 2nd c. CE, its cemeteries and cities had acquired a completely new look, one that can also be discerned in the western provinces of the empire. In the western provinces, this urban development is considered the result of Romanization, but in territories such as Syria, it is tied to the opposite process – to local continuity. This creates a considerable weakness within the Romanization model. Perhaps the most difficult aspect of Romanization is its definition. What exactly is “Roman” in Romanization? Traditional scholarship takes it as a temporal (after conquest) and a geographical (coming from Italy) term. Studies, however, have pointed to the process of Romanization that took place in Italy itself between the 3rd and 1st c. BCE.7 Coming from Italy can, therefore, not be held to be similar to being “Roman.” Furthermore, the material culture 6 7

This line of reasoning is outlined and critiqued in Alcock 1989, 1993; de Jong 2007. Keay & Terrenato 2001, preface; see also Woolf 1998, 7 and Mattingly 2004, 9–10.

PATTERNS OF CONVERGENCE

identified as “Roman” in Britain and Belgium often originated not from Rome or Italy but from other provincial territories in Spain and France. Scholars have also noted that culture change did not always occur within the decades after conquest,but could pre-date conquest or,as in Syria,only become apparent more than a century later. “Roman” is thus neither an entirely temporal nor a geographical term. In more recent analyses, the concept is connected to studies of identity, whereby “Roman” is considered flexible, and not bound by ethnicity, region, or time, but linked to a group of people who shared education, social status, and aspirations. “Roman,” by the time of Augustus, designates a collective identity that was closely tied to Roman power and elite culture.8 The Romanization model also assumes the existence of a non-Roman identity, a local community to be Romanized. This group is equally hard to define. What, for instance, should we consider “Syrian” in the first centuries CE? A brief look at Millar’s book on the Roman Near East demonstrates the complexities of terminology of (self-)identifications such as “Arab,” “Phoenician,” and “Syrian” mentioned in ancient sources.9 They described the ethnic, but also the regional, pre-Roman, nomadic or non-urban, and other identities of the user.The place of the local or the non-Roman remains a stumbling block in Romanization studies. People were not made Roman, nor did they necessarily lose their Phoenician or Arab identity. Rather, they created a new, provincial identity on top of existing ones. Everyone, potentially, had some kind of input, and a new provincial mixed culture came into being.10 A final critique is that the extent of Roman influence remains debated. If urban elites were the driving force behind culture change, one would expect a degree of similarity of adoption within this group. Yet, burial grounds in Syria displayed great variation in the selection of new elements and the scale of their adoption. Funerary portraits and epigraphy were hugely popular in Palmyra and Bosra, but barely existed in the cemeteries of cities on the Lebanese coast. Most Romanization studies cannot account for this type of variation, which was dependent on the specific local context. In a place like Tyre, it was not necessary to add a name and face to a tomb to legitimize social position,perhaps indicating a different social make-up of the urban community. On the other hand, it is possible that other aspects of the tomb achieved this goal, such as the expansive sarcophagi placed on pedestals in the al-Bass Cemetery of Tyre (Figure 11). The local context, therefore, is highly relevant in explaining change, and a monolithic model such as Romanization gives little room to consider such variation. In conclusion, the acculturation model of Romanization in most of its guises does not fully map on to the situation in Syria. Why did similar trends appear 8 10

9 For discussion, see Jones 1997. Millar 1993. For discussion, see van Dommelen 1997, 1998, 2005; Jones 1997, 129; Webster 1997, 2001.

181

182

THE GLOBAL AND THE LOCAL

in the provinces of the empire while pre-existing conditions, scale, and timing were entirely different? How did culture change related to Roman expansion occur without Rome serving as the center of dissemination? In order to place Syria in the wider context of the empire, a model is required that incorporates developments on a global level, i.e., throughout the empire, as well as on the local level, in the provinces, regions, and towns of the empire. GLOBAL/LOCAL

“Globalization”refers to the process of social change as a result of the increasing connectivity of distant regions. Developed to describe the growth of connected markets in the (early) modern world in the 1970s and extending to social and cultural studies in the 1980s and ’90s, the concept has in the last decade been applied to pre-modern and ancient contexts. Connectivity, through trade and migration, was a feature of most societies in the past, but it evolved into a global phenomenon when it changed the way of life of the involved groups. In other words, both the scale and intensity of connectivity and its impact on distant communities transformed a connected world into a global one. For example, local trade networks were remodeled to adapt to global needs, but this in turn brought about changes in local consumption patterns. Events that occurred far away could impact local happenings.11 Globalization was not restricted to markets or economic activity, but could reach deep into cultural life. Perhaps the most commonly used modern example is the Americanization of world consumption, or the influence of North American trends in diet, dress, movies, music, technology, and business practices across the world. People in distant localities adopt similar goods and ideas, resulting in a global culture that reaches beyond national boundaries. In the last few years, scholars have started to look at the Roman world as an example of an ancient global context. According to Hitchner, several characteristics of the Roman Empire make it eligible for a status as a globalizing power. First of all, the Roman system of provinces, or non-sovereign territorial entities, replaced the pre-existing fragmented system of states. Second, a professional army spread across the empire, providing global security. Upgraded infrastructure of roads, ports, and water facilities, often designed to facilitate the army, also increased the movement of civilians, goods, and ideas. The new infrastructure expedited time–space compression, a hallmark of globalization. Markets of communities in distant localities were connected, and people started to employ similar cultural symbols.12 Pitts uses the example 11 12

Giddens 1990, 64. Hitchner 2008. See also Robertson (1992, 8–9) for background on the notion of compression and Jennings (2011) for the application of globalization studies in several ancient contexts. See articles in Pitts & Versluys (2014) for the globalization models applied to Roman contexts.

GLOBAL/LOCAL

of pottery assemblages in Roman Britain. Through the incorporation of British pottery trade into a global system of production and supply, events occurring far away influenced local patterns of selection and consumption.13 It was not the Roman conquest or the creation of a province that accelerated culture change, but the intensification of connectivity in the period of the establishment of permanent army camps and the phase of colonization under Caesar and Augustus. Starting in the late 1st c. BCE, people in the Roman Empire shared interests, values, and tastes.14 Many aspects of this, particularly the applicability of modern concepts of globalization to the ancient world, remain debated. Furthermore, the seeds of the process in the eastern Mediterranean and the Near East may have been sown before the coming of Rome.15 The widespread diffusion of people,goods, ideas, and language in the Hellenistic world attests to a period of increasing connectivity and should perhaps be considered a global phenomenon. On the other hand, in Syria and Mesopotamia, these key features of the Hellenistic world appear to have been limited to certain urban patches, and non-urban communities never underwent the profound change that characterizes the Roman provincial countryside.The scale and reach of global condition remains a factor, and the jury is still out on this matter. The multivariate nature of the globalization debate and the definition of the term in any case prohibit easy modeling. Instead, following Pitts, I use globalization as a phenomenon with which to explore change in the Roman world and to contextualize the patterns of convergence in Syria as they emerged in funerary practices.16

Romanization and Globalization: A Global Shared Culture Globalization brings about the creation of a global culture, which is shared across wide reaches of the globe. Here, it intersects with what scholars have interpreted as Romanization, as discussed in the previous section.17 The “Roman” in “Romanization” refers to a set of goods and ideas that appeared in communities across the Roman Empire. Roman administrators did not force their spread. Rather, the structures of the empire created the conditions for increased connectivity, such as infrastructure, intensification of trade, and peace, as well as the foundation of colonies, which became nodal points of transmission. Whereas nowadays communication technology facilitates diffusion, in the Roman world colonization and other forms of migration, as well as an accessible and multi-ethnic army, achieved this. In the modern world, the US government does not force American products on distant 13 15 17

14 Pitts 2008. Pitts 2008, 494. See also Hingley 2005; Revell 2009, 5; Sweetman 2007. 16 Nederveen Pieterse 2014. Pitts 2008, 494. For the connection between Romanization and globalization, see also Pitts & Versluys (2014, 3–31) and Nederveen Pieterse (2014).

183

184

THE GLOBAL AND THE LOCAL

localities, which instead circulate through marketing strategies on the part of global corporations, expedited by decreasing global distances. The agency lies with local communities, who adopt goods and ideas for their own purposes. At the same time, this process cannot be severed from Rome or the USA, either. Connectivity in the Roman world was largely centered on imperialism of the state, or the expansion and maintenance of the Roman state. American products are attractive because of certain meanings they carry with them, be they modern, exotic, trendy, Western, or a symbol of power. This shared global culture was connected to Roman expansion but cannot be considered “Roman” in terms of an ethnic or geographic designation, at least not by the Imperial period. The previous sections demonstrated that ideas spreading across the empire, such as a particular form of urban planning or public display through sculpture, did not have their origin in Rome, nor can they always be found in the capital city. The “Romanness” of the urban plan was not its origin, but the fact that it spread on the networks of the Roman Empire and was associated with a particular urban ideology that fit the Roman system (see p. 188). Regional trends in funerary representation of individual deceased became popular modes of display across the empire through their connection with new modes of social advertisement (p. 195). “Roman,” thus, functions as a deterritorialized term to designate culture that was popular across the empire.18 Goods and ideas that were never associated with Rome also found their way across connected networks. Examples are cults such as those of Isis and Serapis, and Christianity. While these originated in far corners of the Roman world, they became popular across the empire and supplanted local and, in the case of Christianity, state religion. Inda and Rosaldo named this process “reterritorialization,” or the reinscription of deterritorialized culture. They cite the popularity of Bollywood movies in Nigeria as an example of the same process in a modern global context. The movies did not originate from a center, but instead disseminated on the “peripheral flow of culture.”19 The non-center focus of globalization fits the flows of ideas in the Roman Empire. The period between the 1st c. BCE and the 2nd–3rd c. CE was one of profound convergence, during which people across the Roman world participated in the creation of a global, shared culture. This culture was not identical everywhere, but shared enough common characteristics to be considered part of the same process. Not all urban cemeteries in the empire stretched out along the entrance roads, but often and in far-apart regions they did, coinciding with a phase of urban remodeling. The use of epitaphs varied considerably in terms of content, but also in regional and local spread. Nevertheless, there was a remarkable rise in the use of funerary inscriptions from the European provinces to North Africa and the Near East, and these texts memorialized the 18

Cf., Jennings 2011, 125.

19

Inda & Rosaldo 2008, 14, 25–27.

GLOBAL/LOCAL

lives and deaths of individuals. Revell identifies this “paradox of similarity and variability” in the Roman world as a part of globalization.20 The local origin of aspects of the new global culture remains important, but it should not overshadow the fact that they were part of a new trend of convergence that was closely tied to Roman imperialism.

The Place of the Local A weakness of Romanization studies discussed in the previous section is the fact that variation across regions does not fit easily in a grand narrative of acculturation. Even though the patterns were similar enough to describe them as part of a trend of convergence, the occurrence of local variation remains unexplained. How to describe a process that, on the one hand, seems to erase local peculiarities, but on the other, maintains local variation? Local pre-existing social structure, as well as developments during the Roman period, had an influence on the type, means, degree, and style of adoption of new goods and ideas. Most Romanization scholars recognize this and agree that there was no blanket acceptance of new goods and ideas, but rather an adaptation of them into the existing traditions. Romanization is then presented as a mixture of two opposing, essentialist components: Roman and local. In globalization studies, the local and the global are not opposed but interconnected.21 That is, local adaptations are the global culture. There is no “true” form of global culture that is modified for local use. Rather, the amalgam of differences is an inherent part of global culture. A modern example is the so-called “McDonaldization” of the global fast-food industry.This diffusion of a global type of food and a way of running restaurants is fitted into local dietary customs and business practices. The adjusted menus in McDonald’s restaurants across the world demonstrate how shared ideas are custom-fitted or reworked into a local setting. In the previously mentioned Bollywood example, Nigerians do not adopt the cultural meanings read in India, but create their own meanings. Global cultural forms are customized through interpretation, translation, and appropriation to fit the local setting.22 This is how we should read culture change in the Roman world, as well. Across the Roman Empire, a new concept of urban planning created cities that operated along similar lines and made an urban Gaul feel at home in a city on the other side of the empire.23 But no two cities looked alike. Each had its own character, based on placement and type of buildings, street plan, decoration, sculpture, and epigraphy. The creation of a global culture, therefore, is not the same as homogenization. The cities shared enough characteristics to make 20 22

21 Revell 2009, 2. Cf., Alexandridis 2010, 253; Hodos 2014, 242. 23 Inda & Rosaldo 2008, 18. Cf., Woolf 1998, 1–4.

185

186

THE GLOBAL AND THE LOCAL

them part of global culture, but each kept its own specificities. Bauman calls this process “a matrix of possibilities, from which highly varied selections and combinations can be, and are, made; through selection and combination, from the global yarn of cultural tokens, separate and distinct identities are woven.”24 This particular interaction between globalizing forces and specific localities is by some called “glocalization,” or the local adaptation of global culture.25 Pitts connects the diverse trajectories back to connectivity. Towns that were along the main roads or close to central markets were more connected and acquired new pottery faster. Variation in the material record was guided by the proximity and accessibility of networks, and the ability of local actors to engage with them.26 A major reason for diversity was the adaptability of goods and ideas, and the extent to which they could be incorporated into local structures.27 This varied, resulting in different outcomes. Romanization was, in effect, the creation of a global culture that ran along different trajectories (levels and scale of adoption), was not necessarily center-based, and was expressed through locally specific adaptations. The following sections situate global trends that appeared in the burial grounds of Roman Syria within their specific locality.

CITY AND CEMETERY

Urban cemeteries in Roman Syria arose at fixed locations in the suburban landscape. As previously remarked, the transformation of the urban landscapes of Syria, including the burial grounds, was tied to empire-wide developments. New concepts concerning urban planning in vogue in Roman Syria extended to the location of the burial grounds. This connection was not only chronological, as demonstrated in Chapter 1, but also spatial or physical. Extramural structures of a communal and often civic nature – bathing complexes, circuses, theaters, aqueducts – arose close to or, sometimes, inside the urban necropoleis. Even though the tombs remained outside the city limits proper, their placement made them part of the urban space. The connection between cities and cemeteries can be taken further. The urban building programs emphasized visibility, durability, and adornment. Monumental structures arose in the city, made of stone and decorated with reliefs and inscriptions. These features also characterized many of the new tombs, and some even mimicked public architecture. The two tombs illustrated in Figure 44 combine several elements of sacral and public buildings: a podium with a central flight of stairs, a façade that includes pilasters with 24 25 26

Bauman 1998, 43. See also Appadurai 1996, 7, 17. See Robertson (1992, 173) for background on the term “glocalization.” 27 Pitts 2008, 504. Thomas 1991. See also Dietler 2005, 66.

CITY AND CEMETERY

A

B

44. Mausolea. A: Reconstruction of Tomb 1, ’Atam¯an (Hauran). B: Reconstruction of Tomb 36, Palmyra

Corinthian capitals, and a broken pediment reminiscent of theater architecture. Features typical in public architecture, such as columns, exedras, tetrastyles, and aediculae, found their way to funerary architecture, as did the occasional use of marble and other expensive stone types. Inscriptions marked tombs, as they did the entrances of public buildings. Figural sculpture, in the form of funerary

187

188

THE GLOBAL AND THE LOCAL

stelae, portrait busts, and full-length statues, had parallels with honorary sculpture in the cities. Both tombs in Figure 44 are reconstructed with niches for the placement of statues. Similar public display of figural sculpture occurred along the main streets of Syrian cities. The best evidence for this practice comes from Palmyra, where the consoles that held statues, perhaps in bronze, still adorn the central colonnade of the city.The concept of civic honorary sculpture was,thus, to some extent repeated in the form of funerary portraits outside the city walls. One ought not to take these similarities too far, as most tombs never completely resembled public architecture. Whereas aspects of their construction referred to the new buildings in the towns, the tombs formed their own architectural class. Nevertheless, in the construction of tombs, architects often chose similar eclectic styles current in public architecture.

Urban Planning and Elite Benefactions Shared concepts of urban planning circulated in the Roman provinces starting in the 1st c. BCE. This is perhaps the most recognizable aspect of Roman globalization, as it produced similar-looking towns from Gaul to Libya. Despite having non-Roman origins, it was the deterritorialized “Roman” version of urban planning that spread on the networks of empire. In this form, it reached Syria. In the past, scholars have emphasized the local origins of urban development in Roman Syria, as certain elements of urban planning are attested in pre-Roman cities. Beirut’s rectilinear plan, for instance, may date to the Persian-Achaemenid period, and monumental city walls with gates in Apamea and Dura Europos are usually assigned to the 2nd c. BCE.28 Nevertheless, it was in the Roman period that similar concepts of urban planning and construction emerged in towns across the province. Since the same phenomenon is attested elsewhere in the Roman world in the same period, it qualifies as a product of globalization. Scholars have noted the urban development that occurred in the provinces, and identified various reasons for it. In Britain, France, and Spain, urban construction coincided with a phase of urbanization. State-sponsored colonization, the creation of permanent army camps that grew into cities, and the growth of local centers resulted in an increase of cities in these regions. They attained the recognizable urban form through the package of buildings, infrastructure, and adornment associated with the Roman city. Urbanization in these western provinces is often tied to administrative needs. Lacking administrators, Romans depended largely on local cities to arrange taxation and security. Furthermore, in Roman thought, the city was the ideal form of settlement and connected 28

Beirut: Butcher 2003, 30; Curvers & Stuart 1998–1999, 22. Apamea: Balty 1977. Dura Europos: Leriche 2003.

CITY AND CEMETERY

to civilized life. Rome preferred cities run by several high-status families as the default form of local administration.29 As a result, the city became the central axis of power and a place of opportunity in which one could rise in the ranks of administration. Its spaces emerged as central stages for competition between local elites and the legitimization of their position. Local stakeholders arose as the most prominent financiers of urban refurbishment. Monumentalization of the cityscape, construction of public works, and upgrading of infrastructure expressed the benevolence of the local elite. This euergetism intended to establish the power relationship among the dominant families, but more importantly between them and the rest of the population in the city and its territory. The way this was done is telling: not merely in the form of tall buildings blatantly announcing the wealth and status of a sponsor, but through acts of benefaction, aimed at improving infrastructure and providing buildings that could be used by all the inhabitants. In other words, the urban redevelopment fed, or was intended to feed, a sense of community, in which everyone, or at least more than just the members of the dominant families, participated. It was not only individuals who were on display but also their connection to the wider civic community.30 The standardization of the cityscapes in the western provinces was, thus, the consequence of the growth of urbanization, which created in turn an urban elite that sponsored construction as part of its legitimization of position. Von Hesberg and Zanker, among others, have pointed to the role of tombs in this process, in Italy as well as in the western provinces. Funerary elaboration arose as a prominent location for competition in the later 1st millennium BCE, with a focus on display of social status and relations with the community. Tombs could be benefactions as well, as expressed most clearly in the columbaria built for slaves, freed-persons, and free dependents of the sponsor. Over time, the emphasis on individual achievements decreased, at least around Rome, in favor of similar tombs that stressed civic identity.31 In Syria and other parts of the Roman Near East, the link between urbanization and urban planning was different. A network of cities existed long before Roman boots were on the ground, and the region was never part of a program of colonization. Before the 3rd c. CE, army camps were rarely new foundations, but rather latched on to existing centers. The number of cities and large towns did rise in the Roman period, mostly because smaller settlements grew into urban centers, e.g., Baalbek, Homs, Palmyra, and Shahb¯a. In other words, Syria did not experience a rapid phase of urbanization like that in regions in the European and North African Roman provinces, but its settlement patterns did 29 30

31

Revell 2009, 44; Zanker 2000. For recent analyses of urban display and ideology in the western provinces, see Keay 1997; Revell 2009; Whittaker 1997; Woolf 2000; Zanker 2000; Zuiderhoek 2009. For the Near East: Butcher 2003; Segal 1997. Von Hesberg & Zanker 1987.

189

190

THE GLOBAL AND THE LOCAL

change significantly starting in the late 1st c. BCE. The structure of governance followed suit. An administrative system of cities with control over territories running local affairs replaced older systems of governance by kings and chiefs in the 1st c. BCE and CE, although independent villages and client-kingdoms never went away completely.Limited evidence suggests that high-status families in Syrian cities acted in similar ways to those in other provinces. Their members were in control through city councils, and sponsored urban constructions. Examples come from Apamea, where local elites paid for the aqueduct, basilica, baths, and part of the colonnade.32 A new importance as the seats of power and social mobility may have been behind the redevelopment of Syrian cities, as is argued for other provinces. The urbanization of Palmyra and Homs, both of which had a formerly nomadic population, may also signify a response to the Roman need for urban elites. Despite the fact that the pre-Roman conditions differed compared to areas such as Gaul and Britannia, Syrian elites appear to have behaved in similar ways vis-à-vis the urban environment. This is a sign of their cultural integration into the larger Roman world. Economic prosperity and increased connectivity, which brought ideas, building materials, architects, and artisans, facilitated this process. A new communal identity was promoted, one that had the city at its core. The question remains open as to why this happened. What was the local context in which new concepts of urban ideology were integrated? The role of cemeteries may aid us in answering this question. In fact, in order to understand the local adaptation of the urban plan in Syria, we ought to consider the fact that cemeteries were included in this process.

The Deceased Community Those buried in the urban cemeteries of Syria were spatially and visually represented as the extension of the urban community. They lay close by, connected to the city by the main thoroughfares, and were surrounded by civic architecture. The architectural development of tombs ran, in part, parallel with transformations of the urban landscape, thereby actively drawing the deceased into a communal, urban or civic community. Before the Roman period, funerals may have intersected with rituals of civic expression, but in different ways. The expression of the civic identity of the deceased did not always occur in urban burial grounds before the 1st c.CE.Furthermore,the visibility of Roman graves made the connection between city and cemetery visually prominent and longlasting when compared to the underground tombs of the pre-Roman period. The proximity of a circus or sanctuary meant not only that funeral processions weaved through a civic landscape but also that visitors of games or religious ceremonies crossed the burial grounds. The memory of the dead remained 32

Millar 1993, 261–262; see also Butcher 2003, 89.

CITY AND CEMETERY

linked to the city; a memory that was both inscribed in the permanence of the tomb structures and performed through frequent movement through the grave fields on the way to fields, gardens, buildings of entertainment, and religious sites. In Syria, the dead played a role in the expression of communal, civic identity. The cemeteries did not contain the entire urban community, nor was everyone memorialized in the same way. Male elites were most prominently on display, as were wealthy families. Women were remembered as key partners within the family, and their role as sponsors and owners became more prominent in later centuries. Underage children were largely absent. The poorest may have been included in the urban cemeteries, but their graves remain invisible. Between the wealthiest and the poorest, a sizeable group of urbanites was on display through simpler funerary markers. The cemetery was part of the grand view of what a city looked like, with elite families most prominent, but with other members of the community also represented. The global idea of citizen communities had seeped through the burial spaces of Syrian cities, which presented the different positions of their members. These positions, however, were not set in stone. The urban cemeteries may have functioned as an area of constant negotiation of status position, allowing space for newcomers and others with aspirations off-setting the status quo. Later, I suggest that cemeteries offered a stage for self-definition to a much larger group than just high-status families. That this form of competition and display had moved to visible tombs, or buildings symbolizing the long-term memory of a community, indicates that older avenues of expression were no longer available or had lost their potency.

The Countryside If urban cemeteries presented their occupants as part of the civic community, what happened in the burial grounds of smaller settlements in Syria? Urban and non-urban were not always strictly separate in the Roman world. Cities had a territory over which they exercised control, and this territory included smaller settlements. Roads aligned with tombs connected villages to the main centers. This is not to say that uninterrupted rows of tombs stretched from the main center to the smaller settlements, but rather that village burial grounds flanked the same roads as those of the main center. As such, villagers were drawn into the communal identity of a center. In several areas of Syria, an administrative structure centered on a city with a territory is not immediately apparent in the archaeological and historical record. In the steppe of East Syria, roughly the area along and between the Euphrates, Balikh, and Khabur rivers, urbanization lacked. That is, existing settlements did not obviously grow into cities and new foundations are

191

192

THE GLOBAL AND THE LOCAL

absent in the first two centuries CE. Its existing cities do not appear to have undergone a phase of urban remodeling, and elite benefactions are not attested.33 Cemeteries in East Syria, e.g., those of Selenkahiye, Hammam et-Turkman, and Tell Sheikh Hamad, did not extend alongside the road; nor did they undergo changes focused on monumentality and visibility, with the possible exception of those related to military sites.34 This region, added relatively late to the Roman Empire, clearly developed along a different trajectory, one that requires further investigation beyond this book. Two other areas of Syria consisted of towns and villages without an obvious or extensive network of cities: the Limestone Plateau and the Hauran. In the Hauran, villages like Jmarr¯ın and Si’ were situated in the territory of Bosra and Qanawat, but many others cannot be connected to a city. By the 2nd c. CE, it appears that the villages were self-administered, with a degree of independence not witnessed in other parts of Syria (p. 253). The exact form of administration control in the Limestone Plateau remains unclear, but there is no direct evidence for an urban structure. The construction of baths or temples and the presence of local sponsors indicate that some towns gained urban aspirations, but none has yielded evidence for minting capabilities or the bestowment of city titles. Burial grounds of both areas lay at various distances from the settlement, and some were seemingly not at all connected to a site. Cemetery location, therefore, was not consistently used to emphasize the communal identity of nearby residents in these rural areas. Yet, unlike in the East Syrian steppe, these tombs were not hidden from view, but often were very prominent in the landscape. In fact, some of the most visible constructions in the province arose in these rural areas, such as tall distyle tombs and monumental mausolea (Figure 45). Elements from urban architecture frequently appeared on these constructions, from pediments and cornices to figural sculpture and inscriptions. Mausolea on the Limestone Plateau consisting of four pillars carrying a pyramid-shaped roof are reminiscent of a tetrapylon, which adorned the main streets of cities such as Palmyra and Bosra. The villagers were aware of current trends in tomb construction, and they used the tomb to draw attention to individuals and families. Later, we explore possible links between monumentality and the arrival of newcomers in these rural areas.

“WE ARE ALL INDIVIDUALS”

Throughout the Roman world, funerary portraits and epigraphy increased in popularity in the 1st c. BCE and subsequent centuries. Epitaphs added a name 33 34

De Jong 2011, 2012; de Jong & Palermo in press. For instance, Nisibis: Gatier 1988, 227–229. Selenkahiye: see Appendix 1. Hammam et-Turkman: Thissen 1988. Tell Sheikh Hamad: Novák 2000.

“WE ARE ALL INDIVIDUALS”

45. Aboveground tombs on the Limestone Plateau.A:Tomb 1,Sitt er-Rum.B:Tomb of Aemilius Reginus, Qatura. C: Aboveground portions of Tomb 1, Bamuqqa

and select biographical information to the tomb, and figural sculpture depicted the deceased and members of his or her family or household. Both inform us about the deceased as individuals and their position within a larger whole, most often centered on kinship or dependency relationships such as master and (former) slave. Sometimes, membership of a non-household group is stressed, such as the military or a professional association. Great variation existed in the choices for funerary text or style of portrait across the Roman Empire. Nevertheless, the shared result of adding names and faces to the tomb was the individualization of the actual grave-site and the linking of the deceased to a larger group. Both interior and exterior portions of the tombs were used, indicating that the individualization worked on two levels: as a means of public display and as a way to highlight specific members of a group for a selected audience.

193

194

THE GLOBAL AND THE LOCAL

Scholars of the Roman world are not accustomed to investigate the individualization of funerary space as a cross-provincial phenomenon building on shared meanings. Instead, this process is usually related to specific local contexts, and there is good reason for this. Stelae in the vicinity of Rome, for instance, depicted freed persons and freed families. Their prominence is tied to assertions of family identity, one that had been denied to slaves.35 In Egypt, communities incorporated new portraits into existing traditions of depicting the deceased. Riggs suggests that the concern with survival of individual social roles, common in urban Greek and Roman settings, influenced mummy portraits and other figural imagery in Egypt. Exposure to these concepts happened through migration and the building programs of Augustus.36 McCane sees a similar cultural influence, in this case Hellenization, at play in the preservation of the individual identity of the deceased in ossuaries of Jerusalem, which represents a departure from the traditional practice of assembling the remains of multiple burials through secondary burial.37 Recent studies in Roman art, in turn, move away from models of Romanization and Hellenization, toward tracing thoroughly “un-Roman” elements in figural representations seemingly borrowed from a Roman iconography.38 The funerary inscriptions have received more attention, and the “epigraphic habit” is often connected to a growing need for public display of social persona. Woolf, for instance, argues that the increasing complexity of social life in Roman society, coupled with wealth disparity, new prospects of social mobility, and possible loosening of family ties, produced a new urgency for the display and the preservation of one’s reputation and deeds. This spread from the funerary epigraphy of the elite in Rome,through colonists and soldiers,to local highstatus groups, who emulated this mode of public presentation.39 Bodel points out that the new form of public display dates to Augustan times and was particularly concentrated in cities and militarized zones of the provinces. It was here, he argues, “where distinctions of social rank, and attraction of publicly asserting status were acute.”40 In other words, changes in the existing social, economic, and political life caused by Roman rule turned people to new avenues of competition. The need for public presentation of a social persona perhaps explains the fact that the majority of inscriptions were funerary in nature. The permanence of the tomb provides the opportunity for long-term self-advertisement. George argues that freedmen around Rome had few other options than the cemetery in which to display their newly acquired status, particularly their legal 35 37

38 39

36 George 2005; Kleiner 1987. Riggs 2005, 95, 140, 246. McCane 2003, 10–14, 44. See Masséglia (2013) for the adaption of typical Hellenistic motives, such as the arm-sling and representations of children, to the local specificities of Roman Phrygia. See, for instance, articles in Scott & Webster (2003). See also Webster 1997, 2001. 40 Woolf 1996; see also Woolf 1998, 101–104. Bodel 2001, 6–8.

“WE ARE ALL INDIVIDUALS”

family status.41 In Egypt, urban elites dominated the public sphere of the settlements by publicizing their social position. The cemetery became a way for a larger portion of the population to engage in similar advertisement, according to Riggs and Borg. Funerary imagery was used on coffins placed in the tombs, out of public view, on a much wider scale than display in the public sphere.42 These interpretations are in part relevant for the Syrian context. An increasing need to assert individual identity in the public arena was related to a changing social environment. The previous section discussed how this happened in the public space of the city, where competing elites actively reshaped the cityscapes to draw attention to themselves and the civic community. Tombs were particularly potent media because of their association with deep memory. Whereas only a small group of high-status individuals employed both the urban and the mortuary landscape for such announcements, the rest of the provincial community had access only to the latter, the cemetery. In Syria, much of the display occurred on the exterior of the tomb,rather than hidden inside,as in the Egyptian examples. The fact that funerary architecture sometimes mimicked urban contexts, as illustrated in the previous section, also points to a democratization of public display. The stele functioned as the poorer man’s or woman’s version of a public statue. The emergence of individual display in the cemetery would, thus, signal the emulation of elite practices. It may also be read as a challenge to the existing ranking of social status. The cityscape advertised only its benefactors, but cemeteries included people who never paid for buildings. Social positions were renegotiated, by freed families in Rome or by non-elites in Syria. Those who were barred from participating in political life, such as women, the unfree, and the poor, could be commemorated in the cemetery. This line of reasoning does have its limitations, such as its reliance on emulation strategies. Why would non-elites copy what occurred in the city center? Furthermore, we do not know who exactly used the portraits and epigraphy. There is no indication that they were, for instance, mostly freedmen or other groups marginal in society. In Palmyra, members of the important families buried in one of the four monumental cemeteries and those of presumed lesser status laid to rest in the Northeast Cemetery all participated in the trend. Perhaps the most important objection to the interpretation of civic display is the fact that epitaphs and portraits on Syrian tombs cannot be fully equated with status display in the city. With the exception of military examples, the inscriptions did not refer to a career or deeds done during life, other than the provision of a tomb for the family. The way social persona was ameliorated was not dependent on benefactions to the community. Public display in the city, often tied to acts of euergetism, had a function of legitimization, competition, resolving or at least addressing social tensions. Tombs formed a 41

George 2005.

42

Borg 1998, 88; Riggs 2005, 28, 95–97, 140, 155.

195

196

THE GLOBAL AND THE LOCAL

different context. In order to address the local function of the individualization of the deceased in Syria, we need to look more closely at what was actually advertised. How were individuals commemorated, and for what audience?

Individualizing the Deceased in Roman Syria Both the interior and the exterior space of a tomb carried messages about the individual deceased. Inside a communal tomb, single graves could be marked with texts or portraits, thereby identifying members of the burying group, and sometimes adding a hierarchy within this group. In these cases, the texts and images pointed to the most prominent members of the family. Messages on the outer walls of the tomb or on stelae marking graves illustrate the wish to advertise in public. The visibility of the tombs themselves further underscores the fact that a larger public than those with access to the communal tombs was meant to notice references to the deceased individuals. Public display was, thus, a function of the individualization of a grave. The sculpture often depicted the deceased in a luxurious setting, by adding rich garments, jewelry, and banqueting implements (Figures 29–31). Since the tomb building also promoted messages about the wealth of the buried, it seems likely that images of opulence aimed to produce the same effect. Some epitaphs explicitly referred to the owner or sponsor of the tomb, who publicized the capability to build a tomb and offered burial space to others. The portraits and epitaphs displayed a kind of worldliness, such as knowledge of hairdos and facial hair trends in Rome, or access to Greek literature. These Syrians presented themselves as global elites. In other cases, portraits made minimal reference to global styles, and epitaphs could be as short as a single name. The individualization of the deceased, thus, went beyond a mere display of status and global identity. The inclusion of names and faces on the tomb ensured two things. First, the living could return to the grave and point to particular deceased members of their community. The dead could perform a role in the lives of the survivors. Second, text and image usually placed the deceased within the context of a larger group: the nuclear family, the household, or the military. These groups were actively drawn to or reminded of the care and commemoration of the dead. A key function of the epitaphs centered on the expression of legal ownership of the tomb. The Palmyrene foundation and cession inscriptions appear to have been copies of contracts, which perhaps was also the case for the examples outside Palmyra. Changes in law, e.g, associated with Roman citizenship, could have resulted in the creation of such contracts, forcing a closer definition of rules of ownership. However, this tells us little about their placement. Why copy the record of ownership on the walls of the tomb? Furthermore, the epitaphs not only specified ownership and rights of access, but also actively

“WE ARE ALL INDIVIDUALS”

engaged the community in policing these rights, e.g., by referring to fines. Members of the family, in particular, were reminded of their role in protecting the physical remains of the deceased. Remains of offerings and references to such ceremonies in funerary decoration should perhaps be read in similar ways: they signaled that the proper rites had been carried out and reminded the visitor of these rites. There is no evidence that funerary beliefs had changed, such as the arrival of new views about the wholeness of the body or the negative impact of unquiet spirits. In fact, the way the tomb and coffin ensured the physical integrity of the body resembles pre-Roman traditions. The new importance of individualizing the dead more likely points to a need for a closer definition of the relationship between the (deceased) individual and the living. Since this happened on the wall of the tomb, the funerary space itself became a new arena of self-definition for a community. The personal relationship between spouses, adults and parents, and occasionally parents and underage children was highlighted on the tomb. Ageless and wealthy, funerary portraits depicted the perpetuity of life. In some sense, these portraits and, sometimes, inscriptions denied death, or at least physical decay, by maintaining an ideal version of the deceased. Perhaps such a presentation of the dead aided in alleviating grief, which was also the function of the consolation terms and positive biographical notes that now adorned the tombs. Bereavement had moved to the physical space of the tomb and could be focused on a deceased individual. This ensured the possibility of continuous interaction with the deceased. The need for a closer definition of the relationship between the living, the dead, and the space of the tomb in Roman times suggests two developments. First, the structures with which ownership and the integrity of the body had been ensured in the past were under pressure or failed to be effective. Second, the living used the deceased, and in particular members of their kinship groups, in more active ways. Concern over the survival of certain social roles of the deceased, expressed through individual markings on the grave-site, was a trend that swept the Roman world, disseminating on the waves of globalization. In Syria, these individual dead were used to mitigate anxieties about disruptions and to identify responsibilities. Social positions were more closely defined, centering on the display of wealth and worldliness, kinship and professional (i.e., military) groups, and personal connections with the deceased. In its “glocalized” form, the individualization of the deceased addressed specific local issues in Syria. The idea of individual display was customized to fit local Syrian contexts and tied to existing funerary beliefs. Maintaining the integrity of the body had always been a key component of funerary rites. Wealthy and unaging, the new depictions of the beautiful bodies of the deceased fit this tradition. Naming the deceased may have had similar purposes, ensuring the survival of a person after his or her body had disintegrated.

197

198

THE GLOBAL AND THE LOCAL

A GLOBAL STYLE, A LOCAL WAY OF DEATH

The limestone bust in Figure 43 not only illustrates the importance of individualizing the deceased but also highlights the influence of imperial sculptural styles on Syrian portraits, including hairdo. Dress and gestures on this and other figural imagery reflect Greek concepts of proper body posture, denoting modesty, virtue, and restraint, current in the Roman world.43 We have already listed other examples of empire-wide trends appearing in the architecture and decoration of Syrian tombs, such as coffin shapes and materials, architectural shapes derived from sacral and theater architecture, and the inclusion of scenes from Greek myth. Elements of funerary practice in Syria built on a stylistic and symbolic language that was in vogue throughout the empire. They illustrate the knowledge among Syrians of trends in art and architecture current across the Roman world, and their participation in a shared global cultural language. Aspects of this global culture, however, were not entirely new in Syria. It is here that not Romanization but globalization studies help to uncover the process of culture change. Let’s take the example of sarcophagi. Marble examples from Greece and Turkey had been exported to the Levantine shore long before the Roman period. Famous are the lavishly decorated coffins found in the – possibly royal – tomb in the Ayaa area of Sidon, dating to the late 5th and 4th c. BCE (p. 333). The rise in popularity of imported coffins in the 1st c. CE in Syria was, thus, in part a local phenomenon. However, in the same period, imported coffins also were also popular elsewhere in the empire. With the emergence of inhumation as the fashionable mode of burial for elites in Rome, marble coffins produced in the Eastern Mediterranean flowed to Italy. Cormack has suggested that the rising popularity and wider use of coffins across Turkey post-dated the increase in production for foreign markets.44 In other words, demands from abroad altered local consumption patterns, a phenomenon characteristic of globalizing markets.The use of sarcophagi in Turkey and in Syria was, therefore, both tied to older traditions and a new means of burial. In Roman Syria, these types of burial transformed from a restricted royal and high-elite practice to one that was available to a larger group of people. The popularity of imported coffins also stimulated the production of imitations made in cheaper, locally quarried stone. This phenomenon was not restricted to sarcophagi, either. Alexandridis describes a similar development with regards to female portrait sculpture,which emerged in the Greek world of the 4th c. BCE. By the 1st c. BCE, they adorned the public spaces of Rome and other Italian cities. Reinvigorated by the attraction of Roman elites, female portraits now became more popular than ever in 43

Cf., Fejfer 2006.

44

Cormack 1997a, 146–147. See also Ng 2015, 547–548.

A GLOBAL STYLE, A LOCAL WAY OF DEATH

Greece and Turkey, where they followed both local conventions and stylistic forms created in Rome.45

Hellenization Understanding changes in Syrian funerary practices as building simultaneously on existing traditions and on a new global network shines a different light on an old debate, that of Hellenization, or post-Alexandrian cultural influence from the Greek world. Chapter 2 pointed out that the new tomb shapes sometimes harked back to elite tombs of Hellenistic date, erected in nearby Turkey and the southern Levant. Examples are monumental, rectangular, and heavily decorated mausolea, hypogea with aboveground portions and pyramid-shaped roofs, and façade tombs. In Syria, they appear in the 1st c. CE, around the same time as the construction of tombs with similar elements in other parts of the Roman world. One can ask, therefore, whether the new aspects of funerary architecture entered Syria through the global network of the Roman Empire, or whether they should be traced back to a period of Hellenization pre-dating the Roman period. The same is true with regards to figural sculpture and other iconographic elements with strong links to the Greek world. Was the use of Hellenistic styles in Syrian funerary architecture and decoration in the 1st c. CE a local development that occurred irrespective of Roman networks, or, instead, did this Hellenization occur because of Roman globalization? To reframe the question, the references to Greek learnedness, expressed, for instance, in decoration of Greek myth or certain body postures in figural sculpture, can be read as a reaction against cultural appropriation by Rome or as the result of this fact. The lack of material evidence from Hellenistic Syria complicates this matter, and this is not restricted to funerary architecture. Other areas in which Hellenization frequently occurred, such as inscriptions, urban planning, public architecture, and the sculptural arts, remain scarcely represented in the archaeological and epigraphic record of Syria and Lebanon between the 4th and the 1st c. BCE. Many scholars interpret this shortage of Hellenization in pre-Roman Syria as a matter of missing evidence, instead of an absence of the process.46 The changes in the material record of Roman Syria, however, are unequivocal in one respect: in the 1st c. CE, hallmarks of Hellenized culture appear on Syrian tombs. These include the previously mentioned architectural elements, the sculpted and painted decoration, the mythological content of the decorative scenes, and the use and content of Greek inscriptions. In other words, widespread “Hellenization” of Syrians occurred only when the region was incorporated in the Roman sphere. 45 46

Alexandridis 2010, 253–255. See de Jong 2007 for a full discussion of the Hellenization debate in Syria.

199

200

THE GLOBAL AND THE LOCAL

New styles and ideas flowed across the networks of empire, and did not only reach Syria. Between the 4th and the 1st c. BCE, a regional stylistic language spread across the Mediterranean. Architectural styles influenced by models from Turkey, Greece, and Egypt (Alexandria) appeared on elite tombs in this period, from Rome to Petra. The connectivity of the Roman world transformed these elite modes of burial into a global style in the 1st c. BCE. By this time, the global shared language arrived in Syria in its full force, impacting funerary architecture and decoration, among other things. Hellenization in the Roman period was a deterritorialized phenomenon, as it was not explicitly linked to direct contact between Syria and the Greek world. The new cultural language was more diverse than its original Greek form, and borrowed from Italian,Egyptian,and Near Eastern cultural traditions.Features that sprung up in distant contexts across the Mediterranean were combined in an eclectic form. Hellenization in the Roman period may have been the most classic form of globalization.

Shared Meaning/Local Interpretation What, then, was the attraction of this new language to Syrians? The answer probably varies for the various components of the new style. Chapter 5, for instance, discussed the possible meanings of depictions of Greek mythology in Syrian tombs. Interpretations ranged from the insertion of Greek myth into local beliefs systems to the use of such themes to support the bereaved, or to advertise culture and education. The selection and placement of these symbols on Syrian tombs did not follow clear patterns, and this issue eagerly awaits further attention from art historians. The question also hinges on debates concerning shared meanings and local interpretations. The spread of certain trends across the empire suggests a degree of shared meaning among the consumers, while at the same time adhering to local interpretation. We have already seen the example of the global importance of using sculpture and epigraphy for individual display. Locally, in Roman Syria, such display helped to relieve tensions concerning ownership, fear for and of the dead, and the maintenance of family structures. One particular attraction of the new global style for local communities across the empire could be its association with the high-status culture of display. Knowledge of Greek myth and literature, for instance, was part of empire-wide elite education and was considered part of being “cultured.” Some of the choice in mode of burial was aimed at impressing the living community with the – perceived – high status of the buried, and it seems likely that learnedness could function as part of the message. The popularity of elaborately decorated sarcophagi was also tied to elite culture across the empire. The use of marble, the depiction of banquets scenes, and the adornment with rich garments and

A GLOBAL STYLE, A LOCAL WAY OF DEATH

46. Elevated sarcophagus (Complex XVI), al-Bass Cemetery (Tyre)

jewelry in the portraits of the deceased indicate the importance of portraying luxury and wealth. In other words, some Syrians displayed through funerary architecture their worldliness and knowledge of what was current in Rome and other places of cultural significance.47 Yet, the association with elite culture may have gone deeper than that. The Hellenistic models that influenced a new eclecticism in funerary architecture – the mausolea in Turkey and Jerusalem, the façade tombs of Petra, and even the older Achaemenid tombs – often belonged to royalty or other members of the ruling class. Perhaps this particular aspect, the reference to royal power heightening the status of those buried in the tombs, explains their popularity. These non-global aspects were equally relevant. Chapter 2 proposed that aspects of the new tomb types in Roman Syria originated from local prototypes dating to the Achaemenid period, the Iron Age, and perhaps even the Bronze Age. The elevated coffins of Tyre’s al-Bass Cemetery (Figure 46), for instance, may have copied the nearby Tomb of Hiram, whereas the Amrit tombs, also built between the 6th and the mid 4th c. BCE, may have influenced monumental grave markers in northern Syria. These tombs linked Syrians to the deep past, pre-dating Roman and Hellenistic influence. Such older tombs carried with them powerful associations that may even have counteracted the idea of a shared global culture. Borrowing from specifically non-global models could anchor the new tombs in their landscape settings. This process is further explored in the section below. 47

See Purcell 1987.

201

202

THE GLOBAL AND THE LOCAL

Syrians rarely adopted an entire foreign “package,” but selectively incorporated aspects of the global style in their tombs. The tombs with temple façades in Palmyra and the tetrastyle monuments of the Limestone Plateau, while recognizably inspired by global styles in architecture, each form their own, distinctively local, architectural category. Cormack noted this trend in Asia Minor, where Roman tombs became a “pastiche” of older styles and new architectural trends.48 They turned into local trends, and it is to this phenomenon that we turn next.

The Localization of Funerary Practices The previous chapters outlined the immense variation within the funerary practices of the Syrian province. This diversity was most explicit in tomb shapes and decoration, furniture, and inscriptions. Differences were strongest between regions, i.e., between geographically defined territories encompassing multiple settlements. Funerary enclosures, for instance, were only discovered on the Lebanese coast, whereas distyle markers remained limited to sites on the Limestone Plateau. In some cases, it is possible to identify local styles, particular to a city and its territory. Palmyra produced funerary architecture, epigraphy, and sculpture that remained unique in the region. Other examples are the epitaphs of Bosra and Homs (cat. 2). The development of unique aspects of funerary customs in Syria appears to have operated on a regional and local level. Diversity in funerary customs existed before the Roman period, but accelerated between the late 1st c. BCE and the 2nd c. CE. By this century, each region, and sometimes each city, gave its own form to the tomb buildings. The increasing heterogeneity concentrated on architecture and decoration. Despite regional preferences for glass objects, silver coins, and perhaps facemasks, the selection of grave goods did not seemingly follow regional patterns, and instead varied from tomb to tomb, cemetery to cemetery, and site to site. The same is true for the treatment of the body and the traditions in communal and co-burial, which were more or less similar across the province. It was in the shape and adornment of the tomb that regionalization reached its greatest expression. These were often the visible parts, putting the regional distinctions on display in the Syrian cemeteries. The entire burying community, or those whose graves we have, participated in this trend. In Palmyra, we find the unique Palmyrene script and relief-decoration embellishing simple pit-graves and the grand tower-tombs and mausolea. The taste in decorated coffins on the Lebanese coast included expansive marble sarcophagi, but also versions produced in various local stone types. 48

Cormack 2004, 17.

A GLOBAL STYLE, A LOCAL WAY OF DEATH

The regionalization of funerary architecture, I argue, may have been closely tied to globalization, whereby the increased connectivity of the Roman world made it necessary to define local identity. Jennings calls this process the re-embedding of local culture, as a response to increasing inter-regional interaction. People felt the need to protect local uniqueness.49 In globalization studies, this phenomenon is sometimes branded “localization”: a heightened sense of local or regional identity vis-à-vis shared global culture. If globalization had the potential to erode differences between people, who now consumed similar products and ideas, and to erase earlier boundaries, localization should be considered a counter-reaction, producing a stronger awareness of local identity fixated on boundaries. In a Roman context, the rise of the Second Sophistic is sometimes mentioned as an example of localization.50 Scholars interpret this literary movement as a reassertion of Greek identity in answer to Roman imperialism, and in particular to the appropriation of Greek literature by Roman elites. Another example is the popularity of mummification in Egypt. This old and signature elite burial rite reached its peak in Roman times. Mummification was not only traditionally Egyptian but also explicitly non-Roman. As within the Second Sophistic, it created a tradition in which Roman power had no role. Instead, both phenomena referred back to a preRoman world. The past became a resource by which to ground a community in a specific locale, and to create regional identities. Yet, the Second Sophistic in the Greek world and Egyptian mummification would not have developed in this form were it not for the new global conditions. In other words, the regional counter-reaction was predicated on the existence of a global culture, in whose context it arose. The regionalization of funerary architecture in Syria should be seen in this light, as being intended to express an identity tied to a specific region and sharing a tradition or history. In some cases, the tombs represented entirely new architectural forms, such as mausolea in Palmyra or on the Limestone Plateau. More often, new elements were added to existing forms, e.g., sarcophagus lids or storage jars covering pit-graves or painted walls in hypogea. Existing shapes were “localized.” New elements sometimes referred to pre-Roman traditions: sarcophagus burials and their placement on pedestals, hypogea with decorated façades, and the aboveground monolithic markers loosely signaled forms from the Achaemenid period. In the Hauran, Roman-period tumuli invoked earlier traditions of circular and aboveground tombs. Aspects of funerary buildings thus referred to a pre-Roman past, although the whole package – the tomb – was an entirely new creation. Jiménez has called this the creation of a “new traditional,” whereby goods and ideas appear to repeat the old ways, while 49 50

Jennings 2011, 136; see also Bauman 1998, 43–45; Hodos 2010, 17; 2014, 242–243. Hitchner 2008, 7. For the Second Sophistic, see Borg 2004, Goldhill 2001, Gruen 2011.

203

204

THE GLOBAL AND THE LOCAL

in fact representing something new.51 The new traditional does not represent a continuation of older practices, but an active referral back to a period when tombs could be strong visual statements of power. How important this aspect was is hard to gauge. Our understanding of Achaemenid burial practices in Syria remains limited, as is our understanding of the meaning or association of these tombs in the Roman period. Furthermore, none of the Roman examples faithfully copied pre-Roman local traditions. They represent eclectic compositions of different styles, with various origins. The elevated sarcophagi of the al-Bass Cemetery were reminiscent of Hiram’s Tomb of the Achaemenid period, but the coffins were imported from the Greek world and decorated according to the latest style of global culture. Perhaps we should interpret the references to the past as a particular means of grounding the funerary style to a locality: to embed the imported marble coffins into local Tyrian traditions. Syrians took elements from the new shared culture, mixed these with existing traditions and references to a much deeper past, and created regionally distinct styles. The outcome – the tomb – was a form of glocalization, by adapting global styles to local needs, and of localization, by strengthening local identity. The cemeteries displayed communal identities that had meaning in Syria, concentrated on regional identities of the Limestone Plateau, Syrian Upper Euphrates region or non-urban steppe, Lebanese coast, southern Lebanon, Central–West Syria, Hauran, Beqa’ Valley, and Palmyra. Here, distinct features of funerary architecture arose. Borders were likely fuzzy, and further distinctions were created on the local level, between cities. It is difficult to answer the question of why this process was so pronounced in the funerary context. Two reasons come to mind. First, tombs tie a community to a place and a past. Territoriality and memory both served to strengthen non-global identities. Second, most tombs stood close to the urban landscape, arguably the landscape most affected by globalization. Situated on the edges of cities, tombs could offer a strong counter-narrative to the cityscape. If the expression of citizenship of the Roman Empire was on display in the public space of the city, the cemetery could be used to pronounce membership of a much smaller, local community. THE TIMING OF CHANGE

These changing attitudes to funerary customs in Roman Syria became visible in the archaeological record of the 1st c. CE. This section investigates its timing. Butcher and Millar both note that evidence for the political and military 51

Jiménez 2008, 25; see also van Nijf 2010, 177.

THE TIMING OF CHANGE

205

incorporation of Syrians into the Roman system was apparent around the same decades. By the late 1st c. CE, Syrians seemed to enter the imperial army, and in the next century they appear as Roman senators. This ran parallel with a process of cultural integration that was exemplified, according to Butcher, by local elites embracing the Second Sophistic literary movement, the standardization of architecture, and images on coinage.52 The transformation of funerary practices, therefore, accompanied the integration of Syrians into the Roman world in political, military, and cultural terms. The timing of these developments is surprising for two reasons. First, it did not immediately follow conquest and the creation of the province of Syria, which occurred roughly a century earlier in 64 BCE. Second, the developments in Syria happened late when compared to other provincial territories, where changes started in the Augustan period (late 1st c. BCE–early 1st c. CE). Scholars tie this to the tremendous social change instituted under the first emperor, connected to infrastructural development, colonization, restructuring of provincial territories, and reorganization of the army. Peace under Augustus brought enough safety to intensify trade networks, which now became global. Syrian communities, however, were late to join this global world. To some extent, the situation in Syria was different. It received only a single colony, Beirut, and the absence of a large program of migration may in part explain the tardy integration of the province. Immigration started perhaps only in the 2nd c. CE, when inscriptions of veterans appeared in the rural areas of Syria. Augustus himself does not appear to have been closely involved in matters in Syria when compared to, for instance, Asia Minor or Egypt. Prominent external agents of change – the colonists, veterans, and the emperor himself – were therefore not involved in Syrian matters in the Augustan period. Soldiers, on the other hand, were present in great numbers. Millar illustrates that the province was considered a military region and housed three legions under Augustus, extending to four under Tiberius.53 Furthermore, Syrian trade networks had been highly developed long before Roman conquest, with wellplaced harbors on the route of cross-Mediterranean trade. Roman involvement stimulated caravan trade in places such as Palmyra. Yet, items of global trade did not reach the cemeteries in great numbers before the 2nd c. CE. Globalization studies teach us that external factors are crucial, but that the impetus for change ought to be sought locally. It was not the presence of a global infrastructure per se, but the adoption of empire-wide goods and ideas by local communities, that started the processes of integration. The main question, then, is not why Syrians were late to integrate, but why integration happened when it happened. 52

Butcher 2003, 44; Millar 1993, 88.

53

Millar 1993, 31–32.

206

THE GLOBAL AND THE LOCAL

Political Structure and Competition Despite the creation of the Roman province in 64 BCE, only part of the provincial territory fell under direct Roman rule. Old political families, chiefs, and client-kings governed the rest of Syria (see also p. 12). Over time, this amalgam of local systems of rule gave way to a more direct form of Roman rule, and by the late 1st c. CE, the chiefdoms, city-states, and (client-)kingdoms had disappeared, except on the edges of the province, such as around Edessa. The core territory of Syria was now the direct responsibility of the governor, and local administration was largely in the hands of cities. It is possible that the fragmented system of political power of the 1st c. BCE and early 1st c. CE stood in the way of integration. Hitchner argues that the breakdown of the fragmented system of states, followed by the installation of provinces, signaled the start of globalization of the Roman Empire.54 In Syria, this fragmented system remained in place for another 100 years after the installation of the province. The full integration of its inhabitants into the empire may have started only after its disappearance. A chronological link, thus, existed, but how did changes in the political structure of the province correlate to altered funerary customs? We have in fact already seen an example. As cities attained new administrative importance in the province, notions of civic identity and urban community changed. These were expressed in the burial grounds surrounding the cities. Syrians adopted the global concept of civic ideology and translated this into a new idea of urban funerary space. Scholars of the Northwestern European provinces of the empire put forward a second example. They argued that the imposition of a new superstructure of governance under Rome removed some of the older avenues for elite competition and legitimization, such as warfare, military protection, and patronage. They sought new arenas for the promotion of elite position, such as an official career within the Roman administration. Roymans outlines the demilitarization resulting from Roman peace, the disappearance of martial ideology, and the emergence of a new “civilized” ideology centered on benefactions, inclusion in Roman administration, and conspicuous consumption of luxury items associated with Rome.55 Did such a scenario also apply to the Syrian context? Unfortunately, we know little about the central tenets of elite ideology in Syria in the decades before the coming of Rome, particularly as they related to the petty kings and semi-nomadic chiefs. Kropp has demonstrated that they were in any case already closely entangled with Roman power and the material culture of the so-called Greco-Roman world.56 Perhaps we can look at the question the other way around. By the 2nd c. CE, 54 55 56

Hitchner 2008, 3. Roymans 1996, 9–126. See also Derks 1998, 33–35; Millett 1990, 98; Woolf 2002, 13. Kropp 2013.

THE TIMING OF CHANGE

old leaders and ruling families in Syria had to refocus their methods of legitimization and possibly compete with newcomers. Novel forms of social mobility presented themselves through a career in civic administration and in the army. In this light, it is perhaps not a surprise that two early monumental tombs in Syria can be connected to old ruling families. These constructions of the 1st c. CE, in Homs and Baalbek, belonged, respectively, to the son of a tetrarch (T. 68) and a descendant of the Samsigeramos family (T. 108), known from sources as a local ruling family. Both monuments presented claims to power in a world where such claims perhaps were no longer obvious. The reason why funerary architecture was singled out for this purpose may have been related to its function as a commemorative monument to the family. As further explored later, lineage became a new factor in competition and a powerful source of legitimization.

Demography and Migration The 1st c. CE marks the start of a series of demographic developments that could also have brought about the social change visible in the cemeteries of the province. The rural Hauran and Limestone Plateau were largely repopulated in the Roman era. New settlers pouring into the countryside were responsible for extensive building programs and agricultural development. Inscriptions indicate that this group included veteran-soldiers, who could have been foreigners or, if they were locally recruited, people with a different status than the rest of the provincial population. In both areas, tombs and temples represented the first monumentalized structures. In the Hauran, they took the form of circular constructions, which perhaps referred to old tombs that were still visible in the landscape, such as Bronze Age tumuli. The new circular tombs provided an architectural link to the past for the new population of the Hauran. On the Limestone Plateau, the shapes of the earliest tombs varied, but they usually included an underground chamber (hypogeum) with an aboveground portion such as a façade, stele, or double column. These tombs may be connected with those of the Achaemenid age, constructed in the vicinity of the Limestone Plateau. In both areas, the monumentalization of funerary architecture coincided with demographic change, agricultural development, and the entry of new settlers. Large and visible tombs, some reminiscent in shape of much older types, grounded the newcomers in the rural landscape under development (Figure 45). They could serve to reterritorialize the community. Dentzer records an interesting observation regarding unpublished tumuli in the countryside between Si’ and Qanawat in the Hauran. Rather than randomly scattered in the landscape, they appeared inserted into field systems,

207

208

THE GLOBAL AND THE LOCAL

through placement in the center or along low walls demarcating the farmland.57 There are no maps to illustrate this pattern, but if Dentzer is correct, such tombs could serve as field markers by pointing to ownership of new land under cultivation. We lack the spatial data to investigate this hypothesis more closely, but the suggestion is attractive and would explain the variation in location of rural tombs and the lack of fixed placement when compared to roadside urban cemeteries. An inscription on a tomb at Me’ez explicitly refers to the connection between land, family, and grave-site, when it states that the tomb was made on land owned by the sponsor and consecrated by his father (T. 95). The issue is an old one. In the 1970s, Renfrew developed the concept of funerary field markers to explain the rise of monumental tombs in prehistoric Europe. New farmer communities came into contact with non-farming groups and staked their claims by binding ancestral spirits to the land through funerary monuments. Renfrew’s theory does not necessarily fit the Syrian context and has received much critique, yet the connection between changes in demography, land use, and the construction of monumental tombs remains a valid one to be explored.58 Not all tombs in the Hauran and the Limestone Plateau marked fields. Some stood next to farms or on top of an arid hill. They did visibly signal the presence of new inhabitants farming the land. As family tombs, they advertised a long-term connection to the fields. The audience of these statements would consist of other immigrant groups and perhaps people with other or older claims to the land, such as nomadic pastoralists. It would be a mistake, however, to consider the tombs solely as token land claims. The act of arranging a lasting memorial to one’s kinship group was particularly potent in the face of migration and changes in modes of subsistence. These earliest monumental shapes may have purposely referred to examples from a distant past, such as Bronze Age tumuli in the Hauran or Achaemenid aboveground funerary markers. If so, the tombs created a past not only by housing deceased family members but also by providing an architectural link with earlier tombs. Changing patterns in demography perhaps also initiated monumentalization of funerary architecture elsewhere in the province. Palmyra seems a case in point. Late in the 1st c. BCE, the settlement grew rapidly, presumably as a result of the sedentarization of nomadic populations. Communal, tower-tombs belonging to extended families were among the first types of architecture to be aggrandized. I discuss the example of Palmyra in more detail later. Palmyra and the other instances suggest that the combination of demographic change with other rapid developments, such as sedentarization, agricultural extensification, and the influx of wealth, created a situation in which local communities turned to the cemeteries for new means of expression. 57 58

Dentzer 1985, 82. Renfrew 1976; see also Parker Pearson 2008 [1999], 133–137.

THE TIMING OF CHANGE

209

Families Outside the rural areas and Palmyra, the evidence is less clear-cut. The growth of cities and the infilling of the civic territories with villages point to population growth throughout the province, but it is not certain whether this was rapid or fueled by migration. The same is true for the influx of wealth, changes in subsistence, migration, and urbanization. All occurred in Roman Syria, but the scale and speed are not clear.On the other hand,at the risk of circular reasoning, patterns in Syrian cemeteries could be understood in relation to demographic and related social developments. Zuiderhoek has argued that high mortality and high degrees of social mobility among elite groups in Roman Turkey created an unstable situation, which resulted in an increasing emphasis on the legitimacy of elite positions. He connects the rise of civic euergetism with this instability.59 Syrian cemeteries perhaps offered a similar stage for a society in flux. Both high elites, who dominated the city centers, and lower-status social groups visually occupied the burial grounds. Status uncertainty as a result of population growth, prosperity, migration, and new forms of social mobility generated a need for closer definition of one’s place in the community. The reason such activity moved to the cemetery seems connected to a new purpose of family groups. This book has demonstrated that, often, and particularly in the wealthy tombs, kinship is symbolically elaborated. Inscriptions highlight the importance of families and burial within the context of the kinship group and, albeit to a lesser extent, the household. Such mention of family relations in funerary epigraphy is comparable to that elsewhere in Roman world, as noted, e.g., by Saller and Shaw.60 One could argue that the new medium of the epitaph came with an emphasis on lineage, and that families in Roman Syria were not more important when compared to earlier periods. Syrians, however, also drew attention to these kinship links in other ways, most notably by building aboveground and visible communal tombs, some adorned with group portraits. These monumental constructions provided a new visibility for family identity in the burial grounds of the Roman province. Earlier, we highlighted to the competition over power and resources in these groups, and it is possible that lineage attained greater relevance in this process. The ancestral tomb became a place for marking distinction and legitimization. Visible commemorative structures created the opportunity to define long-term memory and to project origins into the past.61 Extending a lineage into the past could create legitimate claims in the present. Again, we should not restrict the interpretation of these tombs solely as tokens of elite competition. They are the end result of a series of commemorative and memorializing actions, through which the deceased was repositioned among the living and in the 59

Zuiderhoek 2009, 1–5, 151–152.

60

Saller & Shaw 1984.

61

Cf., Barrett 1990.

210

THE GLOBAL AND THE LOCAL

supernatural world. In these acts of separation, repositioning, and longer-term commemoration, family identity became important for the deceased and the surviving community. Perhaps, following Jiménez, the emphasis on lineage was a way to tie oneself to the past, in response to changing population structures.62 The increasing emphasis on families in Syria could signal instability within this group. Earlier, I suggested that the heightened importance of marking the burial space of individuals and groups, the close definition of ownership, and the efforts to draw families in to the upkeep of the tomb and the associated ritual, point to a loss of older ways in an effort to ensure continuity. If social cohesion within a community broke apart, or if there was a fear of this happening, the structures for taking care of graves may have been perceived as being in danger. Whereas in earlier times the collective memory of resident kin preserved ownership and care for the tomb, migration, displacement, and/or high mortality would disrupt these continuities. It became necessary to tie a family explicitly to a tomb.In large cities with a poor and displaced community,the practice of illegal reuse of graves may have increased at the same time as the weakening of (earlier) social or family ties by migration. The fact that none of these developments occurred in the eastern steppe of Syria may point in the same direction. The absence of urbanization, population growth, and the possible relative stability of settlement patterns left communities here more stable.63 Tombs in this area did not acquire a new role in an effort to mitigate pressure on social cohesion. PALMYRA

The particular timing of change was different – in fact, earlier – in Palmyra. Throughout the chapters of this book, Palmyrene funerary customs have emerged as strong outliers. Forms of mummification, the use of Palmyrene script, tower-tombs, and the funerary portraits are unique for this site. Burial and commemoration in Palmyrene tombs occurred in the context of larger kinship groups than elsewhere in Syria.Axiality and clustering of tombs suggest different attitudes to funerary space. Perhaps even the grave good assemblages were somewhat distinct, with high numbers of water-pots and incense burners, and smaller amounts of items of personal adornment, vessels, and coins. The strong insistence on Aramaic as the language of epitaphs diverges from other parts of Syria, where Aramaic was no doubt spoken, but rarely ended up as the written language of a funerary inscription. It is difficult to assess how unique Palmyra was. The dry conditions of the desert and the fact that the site was well preserved and extensively investigated may well explain why Palmyra yielded a different and richer material record 62

Jiménez 2008, 26.

63

De Jong 2011.

PALMYRA

for funerary customs. Perhaps with a full publication of the funerary sculpture of places such as Hama, Homs, and the Hauran, the uniqueness of Palmyra will fade. We have also already noted the strong tendencies toward regionalization in the Roman period. Palmyra is a good example of what such a regional style in funerary architecture and decoration looked like. Moreover, the ritual practices customary in Roman Syria are also attested at Palmyra, albeit in an elaborated form. The underlying principles of keeping the body whole, adorned, and protected, as well as presenting offerings in the tomb, seem to align with province-wide funerary beliefs. The uniqueness of Palmyra is very much in the eye of the beholder. Was Palmyra, then, an outlier? My answer would come down on the side of yes, and for two reasons. First, by the 2nd c. CE, funerary customs in Palmyra were more or less comparable with those of other places, in terms of monumentality, visibility, degree of adornment, eclectic use of style, incorporation of portraits and epitaphs, and grave good assemblages. However, in Palmyra, the steps toward this funerary fashion started earlier than elsewhere in the province. The earliest tall tower-tombs arose in the 1st c. BCE, whereas in most places in Syria monumental tombs started to appear by the mid to late 1st c. CE or later. Funerary busts were introduced in Palmyra in the second half of the 1st c. CE, and elsewhere in the 2nd c. CE. This chronological difference is not likely related to archaeological biases or chance discoveries, but rather is a real sign that Palmyra developed different funerary customs. The second reason is Palmyrene family structure. The tribal background of the Palmyrenes is well known, and the inscriptions make clear that this aspect is also stressed in funerary customs. Perhaps we should also read the clustering of tombs, the forms of dress and adornment, and the axiality of the tomb in this light. They put the spotlight on the larger kinship groups of the town. We have already seen that tombs often emitted messages about family relations, and a distinct familiar structure could set Palmyrene funerary customs apart. The settlement history remains obscure before the 1st c. BCE, and Palmyra likely was a small settlement or group of settlements in the well-watered oasis, consisting of scattered dwellings, orchards, tombs, and perhaps sanctuaries. The last quarter of the 1st c. BCE witnessed construction on a large scale in several locations,including the temples of Allat,Bel,and Nabu.Palmyra appears to have grown into an urban center in the 1st c.CE,which,perhaps not incidentally,was also the period in which Rome’s hand was visible. The construction of a road from Palmyra to the Euphrates by Romans, as well as their sending Palmyrene traders or diplomats to the Persian Gulf, illustrates direct meddling of Roman administrators in local affairs in the early 1st c. CE. The Roman governor fixed the western boundaries of the territory of Palmyra around 11–17 CE, and a commander of the Legio X Fretensis erected statues of Germanicus, Tiberius, and Drusus in the Bel Temple in 17/19 CE. The desert around Palmyra shows

211

212

THE GLOBAL AND THE LOCAL

47. Tower-tombs in the West Cemetery of Palmyra

signs of activity starting in the 1st c. CE, evidenced by small sanctuaries and settlements. Schlumberger has argued that these belonged to semi-nomadic groups involved in camel breeding for Palmyrene trade.64 The 1st c. CE marked the establishment or intensification of the Palmyrene trade network heading to the Euphrates and into Mesopotamia toward the Persian Gulf. In this context,the first,monumental tower-tombs arose in the desert around the settlement, and it is tempting to connect them to the changes in power structures, mercantile economy, and demography (Figure 47). In other words, the pronounced changes in Palmyrene society led some – perhaps the elites who were gaining from the trade – to highlight their position by building monumental tombs for their kinship groups. In this period of flux, monumental funerary architecture became a way of strengthening, or an attempt to strengthen, positions.65

DISCUSSION: THE END OF CONVERGENCE

This chapter started by outlining patterns of convergence that emerged in Syria during the early empire. From coast to desert, from city to hamlet, Syrians 64

Schlumberger 1951.

65

See also de Jong in press.

DISCUSSION

adopted elements from an empire-wide or global culture into their own mortuary traditions. On a general level, the social function of the tombs was similar across the province, despite their highly varied and regionally specific character. This final section discusses the end of this phenomenon, or rather, the slow disintegration of homogenizing pulls and the emergence of different trajectories within the burial grounds of Roman Syria. The emphasis on civic community through the placement of burial grounds vis-à-vis the cityscape, at first glance, did not alter significantly through the Roman ages. Cemeteries continued to occupy the well-trafficked spaces surrounding the settlements after the 2nd c. CE and into the Byzantine era. Yet, the period of monumental construction, both in the cemetery and in the public spaces of the city, was over by the 3rd c. CE. Instead, older tombs were reused, and sometimes remodeled. Civic identities were no longer emphasized with new construction, although the physical connection between city and cemetery remained in place. A real break with existing patterns came in the Byzantine era, when Christian identities entered the cemeteries. In the 5th and 6th c. CE, tombs in the al-Bass Cemetery at Tyre were remodeled to fit Christian architecture, and many tombs carried Christian symbols. At other locations, the creation of churchyards now moved burial grounds into the inhabited space of the city. The production of funerary portraits diminished after the mid 3rd c. CE, and the number of inscriptions decreased sharply in the 4th c. CE. The two elements by which the deceased were commemorated as individuals and as part of family groups were no longer common, highlighting a changed expression of the relationship between the living and the dead on the tomb walls. Ideas about human remains had not, seemingly, altered. Although beautiful bodies no longer adorned the tomb walls, the importance of intact and protected bodies remained in the Byzantine period. The example from Tyre suggests that embalmment vessels continued to accompany the deceased, and that items of personal adornment, such as jewelry, clothing, and shrouds, decorated the body. Amulaic items may have been more popular in the 5th–6th c.CE graves of Tyre. After a lull, epitaphs made a comeback in Byzantine Tyre, as well as in rural areas of the Beqa’ and Orontes valleys, the Limestone Plateau, and the Hauran. The Tyrian examples now emphasized professional identity, whereas the other collections continued to mention various aspects of the deceased persona. New was the mention of the relationship with the Christian god. With the reduction of lavishly adorned tombs, and of epitaphs and figural sculpture, the use of a shared global style became less pronounced. That is, the influence of sculptural styles developed around Rome,the use of Greek myth in decoration, and integration of elements of public architecture in tombs, among other things, are not as apparent by the 3rd and 4th c. CE. On the other hand, new trends flourished, visible in decorated lead coffins and new mausoleum shapes in the rural areas of Syria. Their alignment with empire-wide tastes

213

214

THE GLOBAL AND THE LOCAL

will have to be investigated separately. Globalizing processes had not ended, but the impact of these networks on Syrian cemeteries is more difficult to trace. Different trajectories became apparent in the funerary practices: display tombs arose mostly in the rural areas of the province, epitaphs were limited to a few sites and regions, and lead coffins entered graves in western Syria. Perhaps the most distinctive funerary traditions of the region, those of Palmyra, had ceased entirely. The city has not yielded any funerary material dating after the reconquest of Palmyra in 275 CE. Monumental construction in the cemetery of Tyre also ceased, but, as already noted, was revived in a new form in the Byzantine era. The town of Bosra, by contrast, witnessed few breaks. Epitaphs and large tombs continued to fill the burial grounds of this city. The latest example in the sample of a military grave stems from this city and dates to 320 CE. Elsewhere, the soldiers’ epitaphs had disappeared from the epigraphic record around the end of the 2nd and start of the 3rd c. CE, e.g., at Apamea and Baalbek.Yet,soldiers did not leave Syria;far from it.The military reorganization under Septimius Severus and the subsequent Sasanian invasion of the mid 3rd c. CE created a dense network of camps between Palmyra and the Euphrates, downstream on the Euphrates, and in the Khabur plain (p. 12). These yielded few cemeteries, and when they did, soldier identity was not visibly expressed in them. The processes of convergence characteristic of Syrian cemeteries of the 2nd c. CE were no longer dominant across the province a century later. From the perspective of the cemeteries, it is difficult to pinpoint the end of globalization trends in Syria. Rather, trajectories started to diverge at different sites and in different regions of the province. A new, global idea had taken foothold in the province in the 2nd and 3rd c. CE: Christian religion. Christian identities started to enter the cemeteries of the province around the 4th c. CE, but not seemingly in similar ways. A proper or agreed-upon Christian method of burial and commemoration did not yet exist. An area that emerges from the investigation of the later cemeteries of the province as key for the understanding of continuity and change in funerary practices is the countryside. Surveys around Homs have identified multiple mausolea prominently placed on high points and associated with villages, dated by the surveyors to the Roman and Byzantine period. In the Hauran and the Limestone Plateau, the construction of rectangular mausolea, sarcophagi, and other tombs continued into the Byzantine era, some following distinct regional styles. A prominent tomb with a pyramid roof and covered in relief that housed two sarcophagi came from Blouné (Mešta Blouni), around 15 km southwest of Apamea. At Tabus, in the desert between Palmyra and Deir ez-Zor, a necropolis of the 3rd–6th c. CE included sarcophagi and painted tombs. If the dates are correct, we can also add to this collection the tumuli of the 3rd and later centuries around Selenkahiye and the Middle Euphrates

DISCUSSION

region.66 The importance of monumental funerary architecture was thus maintained in the rural communities of the province, or, in other cases, newly introduced in the later Roman years. Epitaphs appeared and reappeared in the Byzantine period in the Beqa’ and Orontes valleys, the Limestone Plateau, and the Hauran. New research needs to be directed at these communities, whose rituals of burial and commemoration attained new importance in the altered landscapes of the Byzantine era. 66

Blouné: Mouterde 1949, 16–21. Tabus: Lönnqvist et al. 2005.

215

POSTSCRIPT

The last chapter finished by pointing to new areas of investigation, and in particular to the Byzantine countryside, where funerary customs remained central to the expression of local identity in imperial settings. Throughout this book, I have identified other topics that would greatly benefit from further analysis. This postscript reflects on these and intends to take stock: where we are now and what needs to be done. Let us start with the here and now. As I write this, in the winter of 2015, the civil war is raging in Syria. As this book took shape, I experienced the distinct displeasure of writing about cemeteries and tombs while they were being destroyed by shelling, looting, and, in some cases, the dynamite of fanatics. Palmyra has presented us with particularly heartbreaking examples of the vulnerability of cultural heritage and the people who strive to protect this heritage. The losses to Syrian cultural heritage are immense, and the damage of historical and archaeological sites, museums, archives, and storehouses, as well as the murder and displacement of local heritage workers, is still ongoing. If something good has come out of the devastation, it is the flourishing of initiatives for the digital preservation of cultural heritage. People in Syria and across the globe have seized on the opportunity to employ new digital and online techniques to record and disseminate the extensive archival resources concerning Syria. Similar initiatives are underway for the heritage of Iraq. Virtually restoring lost and inaccessible heritage is a potent answer to the cultural crimes being committed in the region. I hope that these trends continue, 217

218

POSTSCRIPT

and that the online database that accompanies this book, originally meant as a resource for further investigation,will also serve as a form of digital preservation of tombs from Roman Syria. This brings me to the next point, regarding so-called “legacy data.” Much of the information used for this book was collected a hundred years ago or so, and found in archives. This material can be difficult to access and analyze; yet it is there. The methodological framework employed centers on dealing with fragmented and decontextualized funerary assemblages. I hope to have demonstrated that it is possible to work with such data. Now that Syrian sites and museums are not accessible, we still have excavation reports, survey notes, and epigraphic corpora to consider. Once we do that, I believe that we will discover that this rich material not only provides a unique insight into the lives (and deaths) of Roman Syrians but also presents valuable examples for the discussion of culture change, Romanization, globalization, ancient imperialism, and local identity. It is hoped that Syria and the rest of the Near East feature more prominently in theoretical discussions of culture change under Rome. WHERE ARE WE NOW?

In this book, I set out to analyze the ways Syrians buried and commemorated the dead in the Roman era.Particular attention was paid to understanding these customs in the wider cultural and socio-political context of the time, and the belief system that guided the funerary rituals. Let us start by summarizing the main findings. Chapter 1 investigates the location of the cemeteries vis-à-vis the settlement, natural features, and non-funerary constructions. The placement of the necropoleis demonstrates a concern for keeping separate funerary and domestic space. Yet the dead were never far removed from the town, and were buried directly on the other side of the city walls, surrounded by civic buildings such as theaters and temples. Several trends emerge from the discussion of the spatial arrangement of cemeteries, which are further developed in the subsequent chapters. The dead needed to be kept at a distance from the living community because of their polluting nature. These concepts of pollution come out of pre-Roman, Near Eastern traditions and provide evidence for continuity in funerary beliefs. Local variation was apparent in the planning of the cemeteries. The spacing and facing of the individual tombs, as well as clustering within the burial ground, reveal great variation across Roman Syria. At the same time,choices for the selection of location illustrate similar concerns about visibility. Tombs were placed along the main roads to the settlement or on top of hills. This emphasis on visibility represents a second trend and was largely a new phenomenon in the Roman period. The proximity of the burial grounds to the town and to civic buildings suggests an intimate connection between the urban citizens and those interred in the surrounding graveyards.

WHERE ARE WE NOW?

This spatial link between the living and the dead was most pronounced in the 1st and 2nd c. CE, when the burial grounds were incorporated and sometimes moved as part of a program of urban development. This phenomenon was largely restricted to the larger towns of the province. In the rural areas, the location of the burial grounds was less standardized. Chapter 2 discusses the architecture of the tomb and includes decoration and tomb furniture. Communal hypogea and single pit-graves were most popular, but they co-existed with several other tomb types. In fact, Roman cemeteries are characterized by a high degree of variation in type of tomb, as well as construction method, decoration, and mode of burial. This accounts for the heterogeneous necropoleis mentioned in the Introduction. The common tomb types existed in earlier centuries, and no older forms seemed to have been abandoned after Roman rule. At the same time, a series of new architectural types were added to the existing assemblage. Tombs, on average, received greater adornment when compared to earlier periods. Aboveground markers added a new form of visibility to the cemeteries, and some tombs were erected on a truly monumental scale. A new role of funerary architecture in Roman Syria centered on the advertisement of the wealth and social position of its owners. Visibility and greater elaboration of the tomb occurred in the urban cemeteries, but also outside rural villages and farmsteads. Although this cross-provincial pattern dates to the Roman centuries, the models for the new architectural elements originated not only from the contemporary Roman world but also from earlier periods such as Achaemenid Syria and Hellenistic Asia Minor and the southern Levant. Roman tombs, therefore, represent an eclectic mix of new and old, foreign and indigenous traditions. Different cities and regions in Syria developed their own types of tomb and decoration,highlighting a growing tendency toward regionalization of funerary architecture. The three interconnected trends of visibility, greater elaboration, and regionalization commenced on a small scale in the late 1st c. BCE, and by the 2nd c. CE had become a widespread phenomenon. After the 3rd c. CE, they were no longer pronounced in the funerary record, but restricted to certain regions. Chapter 3 analyzes the objects that were chosen to accompany the dead. Vessels, jewelry, clothing, oil lamps, and coins formed the common categories of grave goods. Over time, more tombs included artifacts, but fewer of them, and more similar in range. The number of items placed in the graves remained small, and many deceased never received (non-perishable) goods. People were not buried with similar sets of artifacts, and differentiation existed in particular between those occupying single and those in communal tombs. The placement of artifacts in the central areas of communal tombs signifies a different concept of space and likely a more elaborate funerary ritual when compared to single graves.

219

220

POSTSCRIPT

The grave good assemblages were likely meant to adorn the body during the funeral, and to safeguard the deceased through magic and offerings. Their usage demonstrates a high degree of continuity with older traditions. Rituals to take care of the dead, therefore, did not differ significantly from earlier practices, nor did they change much throughout the Roman period. This stands in sharp contrast to the developments in funerary architecture discussed in the previous chapters. As the edifice of the tomb fulfilled a new function in provincial Roman society, many of the rituals that took place around and inside the tomb remained closely tied to long-lasting traditions. Chapter 4 demonstrates similar patterns of continuity and change, while focusing on the occupants and sponsors of the tombs. Using three different sources – skeletal remains, inscriptions, and figural sculpture – this chapter investigates the treatment of the body and modes of commemoration. Inhumation remained the prevailing mode of burial, and practices of communal and co-burial co-existed alongside single graves. A new phenomenon of the Roman era was the popularity of inscriptions and figural sculpture. Together with skeletal remains, these allow for an investigation of the distribution of sex, gender, and age groups. Biological sex did not seemingly result in differentiation with regards to burial, but strong distinctions existed in the epigraphic and sculptural sources concerning gender identities. Text and image more often commemorated men, who appeared inside the tomb as well as on the exterior façade, often as founders of the communal tomb. Commemoration of women most often occurred inside and in the context of the family group. Children remained underrepresented in each category of evidence. Text and image drew attention to the communal identity of the burying group, usually that of the family or the army. Soldier burials stand out as a separate group. Their rites of commemoration and perhaps body treatment diverge from local practices, and align with soldier burials across the Roman world. Portraits and epitaphs often referred to individuals. Personal memories adorned the tombs, perhaps to support the surviving and grieving community, and indicate that the physical space of the tomb attained a new function in the interaction between the living and the dead. Restriction of access to the tomb was also explicitly announced on its walls. Yet, the osteological, epigraphic, and sculptural evidence indicates that such restrictions were hardly ever followed. Common occurrences of reuse, sale, removal, and appropriation of graves highlight the continuous interaction with the tomb, long after its original construction. Chapter 5 examines the belief systems governing mortuary ritual in Roman Syria. Funerary practices are a series of rituals through which relationships between the living, the dead, and the cosmological order are produced. Concepts of pollution ensured the separation of the dead from the living, and much of the ritual was aimed at the body, which was simultaneously beautiful,

WHERE ARE WE NOW?

vulnerable, and harmful. When reconstructing the funerary ritual, it is possible to distinguish between phases. Initially, the wholeness of the body was important, and this was perhaps linked to transformative rituals. Afterward, intactness appears less relevant, leaving room for reuse of burial spots and disturbance of human remains. The evidence also suggests that memorials could be erected without the presence of physical remains. Offerings of incense and liquids were occasionally placed in the tomb. Evidence for other tomb-side activities is more idiosyncratic, leaving little room to reconstruct dominant patterns. The same holds true for the involvement of deities, some of whom were explicitly invoked in support of the deceased, but who more often remained absent in text and image. Many aspects of the funerary ritual came out of pre-Roman practices. Yet, changes in the location of cemeteries, as well as funerary architecture and decoration, signal new forms of commemoration. The tomb building attained a more permanent or long-lasting role in the display of the wealth, individual identity, and family connections of the burying community. The living could use the memory of their ancestors in a more active way. In the early 2nd c. CE, the changes were most profound, and attested across the province. After this period, the regions and towns of Syria started to follow divergent trajectories. Chapter 6 sets developments in Syrian funerary practices in a wider Roman perspective. Too often, analyses of Roman Syria have been restricted to the local context, thereby ignoring the profound impact of the Roman Empire on the provincial communities. The burial record draws attention to three ways in which Syrian practices followed empire-wide trends: the simultaneous redevelopment of urban and funerary space, the popularity of individual display, and the insertion of new styles into local practices. Traditionally, scholars would associate such developments with Romanization, or cultural influence stemming from the Roman center. In the Syrian case, changes in the funerary record are difficult to connect directly to Roman forms of burial. Cultural influence reached Syria from many areas, and some of the popular elements of burials had local origins. The concept of Romanization is too limited in scope to explain the cemeteries of the province. Instead, this chapter draws on globalization studies as providing models that incorporate developments on a global level, i.e., throughout the empire, as well as on the local level, in the provinces, towns, and villages of the empire. Global cultural forms are customized through interpretation, translation, and appropriation to fit the local setting. Across the Roman world, the public spaces of cities underwent redevelopment. This process sprung out of a new importance of urban communities and the need to display visually the (perceived) important members of these groups. The added value of visible tombs in capturing the long-term memory of a community explains why, in Syria, cemeteries were closely tied to the expression of urban, communal identities. A similar local customization of

221

222

POSTSCRIPT

an empire-wide trend can be detected in the popularity of adding names and faces of the deceased to the tomb. This practice built on pre-existing funerary beliefs about the integrity and protection of the body. At the same time, individualization of funerary practices heralded a new interaction between the living, the tomb, and the memory of the deceased. Social positions were more closely defined, centering on the display of wealth and worldliness, kinship and professional groups, and personal connections with the deceased. The influence of styles current in the Roman Empire and originating from the Hellenistic world was particularly visible in the funerary architecture, decoration, and sarcophagi of Syria. The attraction of these styles rested on their reference to high culture and luxurious elite lifestyles. Although not originating from Rome, such fashions spread via the global networks of the empire. The same process of globalization also stimulated a stronger awareness and definition of local identity, expressed in Syria through the pronounced regionalization of funerary architecture. The regionally specific styles helped to stress local or regional identity vis-à-vis shared global culture. The integration of Syrians into the Roman world occurred late in the 1st c. CE, coinciding with the end of a politically fragmented system and the rising importance of cities as administrative centers. Such changes in political structure may have turned cemeteries into new areas for competition and legitimization. The tombs also played a role channeling rapid social change as a result of population growth, migration, urbanization, the influx of wealth, and agricultural development. By situating newcomers in the landscape, announcing continuity of lineage, and engaging family members to ensure the provision of burial rites, funerary practices addressed status uncertainty and fears about reduced social cohesion. The fact that these trends appeared earlier in Palmyra than elsewhere in the province may be related to the rapid change that its inhabitants experienced, in the form of sedentarization, the stimulation of long-distance trade, and direct meddling of Rome in local affairs. The iconic monumental tombs of this site announced lineage, local identity, and knowledge of imperial artistic styles. After the 2nd c. CE, patterns of convergence become harder to trace, as sites and regions started to follow distinct trajectories. This was not the end of globalization per se, as the flourishing of Christianity makes clear. Yet the responses to intense connectivity and integration no longer found a way to the burial grounds of Roman Syria. By the Byzantine period, a new phase of funerary elaboration had become obvious in the countryside of Syria. WHAT NEEDS TO BE DONE?

In this book, I have mentioned several topics that require further study. The epigraphic corpus ought to be mined extensively for data about the Syrian

WHAT NEEDS TO BE DONE?

dead and those who buried them. Of particular importance would be to combine the Greek and Latin sources with Semitic epigraphy. Overcoming this strong disciplinary boundary opens up new areas of investigation and a deeper knowledge of the interplay between language and represented identity. We can take the example of Palmyra, whose epigraphists and historians have crossed the divide between Aramaic and Greek scholarship. Another disciplinary barrier that looms large in this book is between Ancient Near Eastern and Classical (post-Alexandrian) periods. In particular, funerary customs developed in the Achaemenid period may be important. Did Achaemenid royal styles have local significance that endured or was rediscovered in the Roman era? The visual world of Roman tombs in Syria also begs for closer inspection. I suspect that the figural representations of the dead will provide powerful cues to local identity formation. Think here of hairstyles, dress, jewelry, and facial hair. How do we read the so-called Parthian, and thoroughly un-Roman, dress of the reliefs of some Palmyrene men? Perhaps they illustrate the emergence of Bhabha’s hybrid identities, with a particular subversive, anti-Roman character.1 What about the long and loose hair of women in Hama and the Hauran, or their elaborate headgear? Syrian figural sculpture also represents great opportunities for studying gendered identities, a topic that is in great need of investigation. This book has paid special attention to Palmyra, which is both an example of Syrian funerary practices and an outlier. There is another region that has emerged as distinct: East Syria, or the land between the Euphrates and Tigris rivers in North Mesopotamia. Graves here never attained that “2nd c look,” with aboveground markers advertising social position, individuality, family identity, and worldliness. Unlike Palmyra, where the evidence is abundant, East Syria has given us only scant material and epigraphic remains. Yet, this is not a region that has been overlooked by archaeologists. In fact, before 2011, the steppe lands surrounding the Euphrates and Khabur rivers were extensively surveyed and excavated. Even though the center of attention of these projects was on earlier periods, it is highly unlikely that monumental Roman tombs, inscribed stelae, and figural sculpture were not recognized. It is more likely they never existed in any great numbers. Communities in East Syria did not use the cemetery for the conspicuous display of social persona. In fact, they also did not seem to use the urban landscape for such messages, at least not before the 3rd or 4th c. CE.2 Aside from the military stationed in the region, the other inhabitants were not or were only slightly involved in the globalized Roman world of the first centuries CE. As remarked upon in the Introduction, the provincial boundaries of Syria were not cultural boundaries. East Syria provides an example of a different cultural unit. 1

Bhabha 1994.

2

De Jong & Palermo in press.

223

224

POSTSCRIPT

My final remark centers on the difference between burial and commemoration. Inscriptions, funerary portraits, and physical remains tell different stories about who is buried in a grave and how his or her body is treated. What is mentioned or depicted aboveground is not what archaeologists encounter belowground. These discrepancies have great implications for the study of funerary ritual in general, as well as any questions about demography or the function of epitaphs. To investigate these questions, we may need to leave Roman Syria and concentrate on better-excavated contexts with human remains. We also ought to look outside the cemeteries. Commemoration of deceased individuals could occur at various points in the landscape and was not always associated with actual places of burial. But let us stick to the cemeteries of Roman Syria for now, as they present us with unique insights into Syrian communities and the ways they renegotiated their place in the local community and the larger structures of empire.

APPENDIX 1

SITES

APAMEA1

The ancient town of Apamea (modern Afamia) is located in the Syrian Orontes Valley. The settlement consisted of a citadel, now occupied by the modern village of Qal’at el-Mudiq, and a lower town (Figure 48). The earliest archaeological phases on the citadel are prehistoric, but the site is best known as a (re)foundation by Seleucus Nicator in 300/299 BCE, when it was named after his wife. Little is left from this period, and the development of the site remains largely uncertain until the 2nd c. CE. A funerary stele found in Beirut (T. 75) mentions a census conducted around 6 CE recording 117 000 people in the territory of Apamea. The rebuilding phase after the destructive earthquake of 115/116 CE appears prominently in the archaeological record. In the first half of the 3rd c. CE, the winter camp of the Legio II Parthica and two auxiliary cavalry units were garrisoned at Apamea. At its largest extent, the city covered around 250 ha. The Sasanian king Shapur briefly conquered Apamea in 256 CE, after which the fortification walls of the city were reinforced using, among other things, funerary stelae as building material (see later). The construction of several churches, as well as large houses with lavish mosaic floors, indicates that Apamea’s wealth was not diminished in the Byzantine period. At least two ecclesiastic buildings contained tombs.2

The Cemeteries Investigations at Apamea have yielded a total of sixty-three graves dating between the 1st and 4th c. CE, with a concentration in the 3rd c. CE (Online Appendix Apamea 1). The cat. 2 assemblage includes over 128 tombs, many of which stem from excavations in the cemetery area in the 1930s (Online 1

2

J. Balty 1969, 1984; J.C. Balty 1988, 1994; Balty & Balty 1977; Millar 1993, 250–251, 262–263; Viviers & Vokaer 2007. Cemeteries: IGLS IV; J.C. Balty 1977, 1981, 1988; Balty & Balty 1972; Balty & van Rengen 1993; Mayence 1938; Prentice 1908b; Vandenabeele 1972; van Rengen 1969, 1972. The East Cathedral held at least thirteen tombs, and the Atrium Church thirty-three cistgraves, perhaps dating to the 5th and 6th c. CE (J.C. Balty 1969, 77; 1972, 197; 1981, 109, 115).

225

226

APPENDIX 1

48. Plan of Apamea

Appendix Apamea 2). Fires during and shortly after the Second World War destroyed the records of these excavations. The North Cemetery stretched out on both sides of the main road leaving the North Gate. Excavations yielded thirty-two tombs, thirteen of which were included in the cat.1 database, dating between the 1st and 4th c. CE and perhaps later. No dimensions or maps exist for the entire burial ground, but a section published by Vandenabeele extended over at least 53 m along the road and 5 m from the road. In the excavation trenches, the tombs were close to one another, spaced within 1–2 m apart, but the overall spacing and density in the cemetery remain unknown. Over time,

APAMEA

227

120 100 80 60 40 20

cat. 1 cat. 2

0

chart 8. Distribution of tomb types, Apamea

the space of the burial ground filled in with more tombs. The North Cemetery yielded jar-burials (15), pit-graves (9; Figure 13), cist-graves (3), hypogea (2), a mausoleum (Figure 10), a stele, and a sarcophagus in the open air (Figure 14). The East Cemetery yielded three stelae, a pit-grave, a funerary sculpture of an eagle, and three sarcophagi on a pedestal. They date to the 1st–3rd c. CE. No plans have been published, and the exact location of the cemetery and its internal layout are unknown. The remainder of the tombs from Apamea were spoliated or unprovenanced (Figures 16, 34). The rebuilding of the city wall after the attack by Shapur in 256 CE included 130 funerary stelae commemorating soldiers, of which thirty-seven are published. A second group of stelae with Greek inscriptions of the 2nd c. CE originated from several locations. Chart 8 provides the tomb types of the cat. 1 and 2 assemblage. Stelae were common in all periods, while small and single tomb types, such as jars and pits, characterized the 1st–2nd c. CE. The 3rd c. CE witnessed the introduction of larger and communal tombs. Cist-graves were perhaps typical of later centuries (3rd–4th c. CE). The information about grave goods and skeletal remains is limited, as most assemblages remain unpublished and/or were robbed. Seven single tombs contained thirty grave goods (Chart 9, Figure 25). At least nineteen cremation burials are reported from Apamea, fifteen in jar-burials and four in pit-graves. They appear to date before the 2nd c. CE. The excavation reports mention numerous burials of children and adults in the pit- and cist-graves. The age seems to have been determined based on size, the problems of which are discussed on p. 103. The inscriptions provide better insights into age and gender distributions. In total, forty-nine graves carried

228

APPENDIX 1

7 6 5 4 3 2 1 0

chart 9. Distribution of grave goods, Apamea

fifty-five inscriptions. The majority (37) were carved on the military stelae of the 3rd c. CE. These military inscriptions were written in Latin, with one bilingual (Greek and Latin) exception. The text was fairly standardized, mentioning double name, rank, age, an invocation of Roman protective deities, the Manes, and a specification of who had erected or paid for the stele. Usually, this role was filled by fellow soldiers, but in two cases the stele was set up by the wife, and in three by freedmen. One hypogeum held two sarcophagi dedicated to centurion and their wives (Figure 17). The remaining inscriptions in Greek varied in composition, and could include year, month, single name, and an expression or exclamation. These mostly 2nd c. CE inscriptions were erected to an equal number of men and women. The cat. 2 inscriptions include more examples: at least ninety-two Latin military inscriptions, four Greek dedications, and one bilingual text.3 BAALBEK4

The modern town of Baalbek lies in the central Beqa’ Valley between the Litani (Leontes) and Orontes rivers.Scholars sometimes mention Baalbek in conjunction with the Ituraeans, who lived in the Lebanese mountains and caused trouble for the Roman administration in the 1st c.CE (T.75).5 Baalbek is considered 3

4

5

Multiple funerary inscriptions from the vicinity of Apamea are published. These include Greek and Latin inscriptions from the 2nd and 3rd c. CE, and one in a painted hypogeum (IGLS IV, 98–113). Butcher 2003, 116, 365–368; Jidejian 1998 [1975]; van Ess & Rheidt 2014. Cemeteries: IGLS VI; de Jong 2011 [2015], 2014; Freyberger & Ragette 1999; Hakimian 1987; Krencker et al. 1923; Petersen 2003; Ruprechtsberger 1999; Salamé-Sarkis 1987; van Ess 1998; van Ess & Petersen 2003; van Ess & Weber 1999; van Ess et al. 1999; Wiegand 1921. Butcher 2003, 365; Jidejian 1998 [1975], 20.

BAALBEK

229

the religious center of the Ituraean homeland, although there is little evidence in support of this.6 The pre-Roman phases of the town remain obscure thus far. Most scholars assume that the main Roman sanctuary on the citadel stands on an earlier, perhaps Hellenistic podium. Baalbek may have been part of the territory of the Roman colony at Beirut from the late 1st c. BCE, and became known as Heliopolis. The high numbers of Latin inscriptions provide evidence for the presence of colonists in the vicinity of the town. Coins announce the independence of the colony (Colonia Iulia Augusta Felix Heliopolis) under Septimius Severus in the late 2nd c. CE. Enlargement of the main sanctuary on the citadel, dedicated to Jupiter, may have begun shortly after the colonization of Baalbek in the late 1st c. BCE. Additional temples, other public buildings such as a bath and theater, and a city wall arose between the 1st and 3rd c. CE.

The Cemeteries Previous research in and around Baalbek has identified several cemeteries of possible Roman and Byzantine date. In total, fifteen tombs were sufficiently published to be included in the sample (Online Appendix Baalbek 1). These dated between the 1st and the 4th c. CE. A minimum of ninety-nine tombs make up the cat. 2 assemblage, and date between the 1st and the 5th c. CE (Online Appendix Baalbek 2). Most were discovered during a survey by the author in 2010. Three cemeteries surrounded the urban center, extending on the west, east, and north slopes of the Sheikh Abdallah Hill, the el-Sharauni area, and the el-Solh neighborhood. The tombs by Ras el-Ain mentioned by Wiegand may have been the eastern extension of the Sheikh Abdallah burial ground. Two additional cemeteries perhaps belonged to villages in the territory of Baalbek: Douris and the region of the el-Kayyal quarry (Figure 49). The cemeteries surrounded the ancient settlement on all sides, except on the west side between el-Kayyal and the citadel. This area is less rocky, and is nowadays mainly covered by agricultural fields. It is possible that this land had a similar purpose in Roman times and was never used for burial. Three distinct tomb types could be identified in the Baalbek burial grounds, and the cat. 2 assemblage included more shapes (Chart 10, Figure 17). Only one tomb, a sarcophagus in Douris, yielded finds and skeletal remains (Chart 11, Figure 24). Fick mentions two golden face-masks from the Baalbek cemeteries, one from a grave on the north slope of the Sheikh Abdallah Hill and one from an unknown location. The latter perhaps dated to the 1st c. CE.7 The cat. 1 tombs yielded four inscriptions: Greek (2), Latin (1), and Greek and Latin (1); another eighteen are listed in the cat. 2 database (eight Latin, ten 6

Myers 2010, 173.

7

Fick 1999, 77–81; see also Jidejian 1998 [1975], 157.

230

APPENDIX 1

49. Plan of Baalbek 50 45 40 35 30 25 20 15 10 5 0

chart 10. Distribution of tomb types, Baalbek

cat. 1 cat. 2

BAALBEK

9 8 7 6 5 4 3 2 1 0

chart 11. Distribution of grave goods, Baalbek

Greek). Two of the Latin texts specify the military background of the deceased or sponsor. In general, the text of the Greek and Latin inscriptions is comparable, although the former sometimes includes consolation terms. The published funerary stelae from the larger Beqa’ area include numerous Latin funerary stelae, likely belonging to veterans or colonists.8 The sculptural evidence from Baalbek (Figure 30) includes a libation scene (cat. 1), busts, seated women, couples, and funerary sculpture of eagles (all cat. 2). Three stelae were carved with raised lower arms.

Sheikh Abdallah Cemetery The largest burial ground of Baalbek covered the slopes of the Sheikh Abdallah Hill, south of the ancient city (Figure 49). This area yielded four cat. 1 and fiftynine cat. 2 tombs. The earliest dated evidence came in the form of a funerary inscription of the 1st c. CE, probably originally belonging to a mausoleum, dedicated to Zenoros and sons by his wife. The tombs were found in four areas of the Sheikh Abdallah Hill: the north and west slopes, the south quarry, and the Hotel Palmyra area. Tombs extended from the northern tip of the hill, just west of the staircase of the Mercury Temple, to the quarry area that lies 900 m to the southwest. The width of the burial ground is unclear, but reached at least 200 m between the mid-section of the hill slope and the theater. The burial ground thus covered 22 ha or more. The tombs did not maintain strict patterns in orientation, but seem largely to have followed the direction of the bedrock outcrops, and often opened toward the road. The tombs on the west slope were concentrated on the middle portion of the slope, close to the modern road, and the upper stretches of the slope appear devoid of tombs. More tombs 8

IGLS VI.

231

232

APPENDIX 1

can be expected in this area, but they now lie underneath modern residences or are obscured by debris. The northern and eastern slopes of the Sheikh Abdallah Hill, toward Ras el-Ain, also could not be investigated. Older reports mention the discovery of multiple tombs in this region, as well as on top of the hill, in an area currently occupied by an army base. The cemetery aligned the main road leading southwest out of the city and lay close to the Mercury Temple and theater, both of the 2nd c. CE. Five stelae, a hypogeum, a funerary statue, and sculptures of eagles originated from the area of the Palmyra Hotel, which was also the location of the Roman theater. A large hypogeum flanked the lower section of the staircase to the Mercury Temple. It is possible that these graves pre-dated the non-funerary construction in this area, and were vacated at the time of redevelopment.

Other Cemeteries The second urban cemetery, in the el-Sharauni quarter, arose along the road to Homs. Other than Wiegand’s remark about the late date of the hypogea, there is no indication about the chronology of this burial ground. The brief survey in 2010 investigated the quarry at Moghr al-Thin, north of the ancient city along the main the road to Homs. This area was likely part of the larger elSharauni Cemetery reported here. If so, the area were tombs are reported was at least 900 × 400 m. Southeast of this quarry extends the el-Solh neighborhood, where according to current residents construction work in the area east of the old city wall yielded the remains of Roman tombs. Dense habitation hindered further investigation of this area. This could also be part of the el-Sharauni burial ground. One tomb in cat. 2 was found on the road to Nahle, located more or less between the two areas. If the entire area was part of a single cemetery, it covered at least 39 ha. A group of eleven tombs originated from the village of Douris on the western outskirts of modern Baalbek. This location was ca. 1.5 km from the Sheikh Abdallah quarry, and it is not clear whether these tombs were the western extension of the urban burial ground, or belonged to a separate settlement. They flanked the same road as the Sheikh Abdallah Cemetery. The tombs from Douris were found in two nearby trenches excavated in 1996 and 1997. The exact location of the 1996 trench (ca. 14 × 5.5 m) is not published. The 1997 trench measured ca. 2.5 × 2.1 m. All tombs dated between the late 2nd and the 4th c. CE. Approximately 2 km northwest of Baalbek was the quarry of el-Kayyal, currently used as garbage dump and a residential quarter (Figure 49). This area yielded four sarcophagi from two locations. The el-Kayyal location extends ca. 2.2 km northwest of the Sheikh Abdallah Cemetery and over 2 km from the sanctuaries. It is possible that the tombs here belonged to a village in the territory of Baalbek, rather than to the urban population, similarly to

BEIRUT

Douris. The remainder of the cat. 2 tombs from the Baalbek area were reused in later architecture or published without context.9

BEIRUT10

The Lebanese capital on the Mediterranean coast covers habitation layers as old as the Palaeolithic period (Figure 50). The ancient coastline was more inland than at present, and the original settlement stretched out over the ancient tell (“acropolis”), around the harbor area, and southwest into the modern souq area. A city wall with round bastions surrounded the Hellenistic city, running from the southeast corner of the tell toward the south and turning to the west. The exact construction date is not clear. Coins indicate that the site, renamed Colonia Julia Augusta Felix Berytus, became a colony for veterans from two Roman legions (V Macedonia and VIII Gallica) in the late 1st c.BCE. The territory of this colony, the only one in Syria, may have extended to Baalbek. Septimius Severus diminished the city’s territory after it supported Niger’s claim to the throne in 193 CE by awarding Baalbek its independence. Probably starting in the 3rd c. CE, Berytus became an important center for the study of Roman law. The Roman city expanded beyond the pre-Roman settlement, although the extent is not known. The archaeological evidence points to a new building phase in the 1st c. CE, with more construction in the 2nd and 4th c. CE.

The Cemeteries Large-scale excavations in Beirut have uncovered large numbers of Roman and Hellenistic tombs, the majority of which are not yet published or currently under investigation. The discussion that follows is based on the published materials. Pre-Roman burial grounds extended west, southwest, and east of town. An Iron Age cemetery followed the ancient coastline west of the city and consisted of hypogea and shaft-graves (Figure 19). At least seven hypogea and one shaft grave here were still used in the Hellenistic period, and perhaps into the 9

10

Excluded are the stone sarcophagi that currently stand in or around the Qalaa, which were studied by Petersen (2003). The original location of these coffins is unclear and may have been elsewhere in the Beqa’ Valley. Butcher 2003; Butcher & Thorpe 1997; Curvers & Stuart 1997, 2004; Lauffray 1944–1945; Millar 1990; Perring 2001; Saghieh-Beydoun 2005; Saghieh-Beydoun et al. 1998–1999; Seeden & Thorpe 1997–1998. Cemeteries: Cagnat 1925; Carington Smith 1984; Chéhab 1934a, 1935, 1944–1945; Cumont 1929; Curvers & Stuart 1998–1999, 2005; de Jong 2001; du Mesnil du Buisson 1924–1925; Jidejian 1993; Jones Hall 2004; Koch 1977; Koch & Sichtermann 1982; Lauffray 1977; Mouterde 1929; Renan 1864–1874; Skupi´nska-Løvset 1999; Stuart 2001; Ward-Perkins 1958, 1969. The post-civil war excavations that started in 1993 are numbered Bey 001, 002, etc.

233

234

APPENDIX 1

A

B

50. Plan of Beirut. A: Location of sites and roads. B: Area of Beirut

early Roman decades (see later). East of the city wall was a field with twenty pit-graves (Bey 152) of the 2nd c. BCE. The third location with tombs (Bey 045; Figure 19) was perhaps inside the walled area of the Hellenistic lower town, making it the only intramural burial site of the region. On the other hand, these four 4th–3rd c. BCE pit-graves could have pre-dated the construction of the fortification wall.11 Slightly more than half of all tombs contained grave goods, which were relatively simple and few. The grave goods consisted of items of personal adornment, vessels, lamps, bronze discs or mirrors, figurines, an iron nail, a spindle whorl, and a bone die (Table 4). Most tombs contained a single individual; multiple individuals (up to seven) were found in the hypogea. Of these, ten were adults and two children. 11

Curvers & Stuart 1997, 184–189; 2005; Stuart 2001; Thorpe 1998–1999, 63.

BEIRUT

235

60

50

40

30 cat. 1 cat. 2

20

10

0 pit-grave (2)

hypogeum funerary (29) enclosure (5)

stele (2)

coffin (60) column (3) unknown shape (5)

chart 12. Distribution of tomb types, Beirut

The Roman cemeteries of Beirut extended west, south, and east of the city. The majority of the tombs have received only cursory attention; fifteen could be included in the analysis (Online Appendix Beirut 1). Their dates range from the early 1st c. CE to the 4th c. CE, with a concentration between 100 and 250 CE. The cat. 2 sample included a minimum of ninety-one tombs (Online Appendix Beirut 2). Current excavations at Beirut are uncovering numerous tombs, which await publication. The distribution of types is shown in Chart 12. The locations of the graves allows for the reconstruction of burial grounds on the slopes of the Ras Beirut and Ashrafieh hills. Ras Beirut rose west and south of the Roman city. Archaeologists discovered tombs in several trenches in the Kanthari area (Bey 022,104) that were part of a single cemetery.Three more find locations (Bey 095, 163, and 076) likely were also part of this cemetery. If the course of the Decumanus Maximus has been reconstructed correctly, it probably passed the cemetery.12 The burial ground measured at least 150–230 × 270 m and contained tombs dating to the 1st–4th c. CE. Three funerary enclosures with loculus burials, sarcophagi, and pit-graves were discovered at Bey 022. It is likely that the other locations (Bey 104 and Bey 095) also contained enclosures, but damage and partial excavation make this original placement of the rock-cut pits and built loculi uncertain. Excavations have uncovered tombs on the west slope of the Ras Beirut hill as well, a location almost 3 km from the Roman city. These tombs date to the 3rd and 4th c. CE (cat. 2). Scholars have speculated about the location of a circus, which appears in 4th c. CE sources. Remains of possible steps cut in bedrock and foundations of the eastern turn of 12

Curvers & Stuart 2005.

236

APPENDIX 1

the hippodrome were found in Bey 072 in the Wadi Abu Jimal area, supporting the hypothesis that the modern street plan in this area follows the outline of the ancient circus.13 On the other side of the city, one or a series of burial grounds covered the northern and western slopes of Ashrafieh. Tombs are reported from the lower section of this hill, along the modern and presumably ancient road leading east of the city. The tombs stretched out over a wide area, and some may have been as far as 1 km away from the Roman town. There is no information about the chronological development and internal layout of the burial ground(s) at Ashrafieh. The published tombs from this area consist of four decorated lead coffins and one terracotta example of the mid 2nd–early 3rd c. CE (cat. 1).14 It is not clear in what type of setting these coffins stood, but it was likely a hypogeum or funerary enclosure. The cat. 2 material includes hypogea and numerous coffins. A third location was at Bey 018/040/063. There is some indication that the Iron Age and Hellenistic cemetery at this location was reused in the early Roman period (cat. 2). An unknown location in Beirut yielded a funerary stele the Latin text of which indicates that it belonged to a Roman official who built the tomb for himself, his freedwoman (wife?) and son (cat.1). This is the only text from the region that lists the cursus honorum of a Roman official. One Proconnesian and two sarcophagi of unspecified material came from unknown locations in the city (cat. 2). A Latin inscription of a soldier adorned the Proconessian coffin, and one of the other sarcophagi had a Greek epitaph. A total of 119 grave goods were recovered from nine tombs (Chart 13).Compared to the other sites, the Beirut tombs included more glass vessels, more coins, and fewer lamps. Within the category of items of personal adornment (thirty-six in total), golden face-covers and headbands take up a relatively large proportion. The hypogea included more finds from most categories, except lamps, when compared to the other types of tomb in Beirut. The assemblages from the two different cemetery areas did not differ considerably. Reliable osteological data come from Bey 022, which allowed for the identification of seventy-five individuals in twenty-eight loculi (2.7 individual on average). Twelve individuals were buried alone. The other burial spots 13 14

Curvers & Stuart 2005, 199; Davie 1987, 155; Mouterde 1964, 166. There is some discussion about the date of the lead sarcophagi in Ashrafieh. Mouterde (1929) gives one in the late 2nd and early 3rd c. CE, based on chronological development of the lead sarcophagi. Carington Smith (1984) argues that the burial could be older (25–100 CE), based on the finds in one of the coffins: an amphoriskos (perhaps 1st c. CE), a gold ring (possibly before 3rd c. CE), and an engraved bracelet (Greek lettering could be 1st c. CE). However, while Smith’s analysis shows that the finds could be dated earlier, it does not prove an earlier date for the burial. Furthermore, lead sarcophagi do not seem to have appeared in the Syrian region before the middle of the 2nd c. CE (Koch & Sichtermann 1982; Rahmani 1999). Therefore, the date is mostly likely between the mid 2nd and early 3rd c. CE.

BOSRA

237

25 20 15 10 5 iron object

bronze object

nail

terracotta figurine

spindle whorl

lead envelopes

lamp

gold coin

coin unknown material

lead vessel

bronze coin

glass amphoriskos

glass unknown shape

golden headband

glass unguentarium

golden mask

golden wreath

necklace

gold leaves (set)

pin

ring

earring

bracelet

0

chart 13. Distribution of grave goods, Beirut

contained between two and nine individuals, placed on top of one another or pushed to the side. The assemblage included seventeen women and twenty men, fifty-five adults, and nine children below twelve, including one infant. Females, males, and children were buried together; however, the (sexed) single burials belonged to a male, and in one case an infant. Few of the tombs are inscribed. The published inscriptions, including cat. 2, included a higher number of Latin (10) than Greek (2) inscriptions, a testament to the area’s colonial status. BOSRA15

The modern town of Bosra (ancient Bostra) is located south of the Jebel el’Arab in the Nuqra plain. It was already mentioned in Bronze Age texts, and remains from this period come from multiple locations. The area was part of the Nabataean kingdom from around the late 2nd c. BCE to 106 CE, when it was incorporated into the province of Arabia. The evidence for urban redevelopment comes from the 1st c. CE, and the town covered a surface area of ca. 90 ha by the end of the Nabataean period (late 1st c. CE). Bosra became the principal center of the province of Arabia under Roman rule (106–330 CE). This new role and the presence of a legion and auxiliaries stimulated more building programs, such as baths, colonnaded streets, a circus, and a theater. The town expanded beyond its earlier limits and the army camp lay north of the urban center. Inscriptions and evidence for parceling of fields (centuration) indicate that Roman colonists and veterans of the army settled in and around 15

Bopp 2008; Dentzer 2003, 181–184; Millar 1990, 52. Cemeteries: CIS ii; IGLS XIII/1, XIII/2; RES IV; al-Maqdissi 1993, 474; Dentzer 2007; Dentzer-Feydy et al. 2007; Oenbrink 2003, 89; Sartre-Fauriat 2001, vol. I, 19–52, vol. II, 3–7; 2007a; 2007b, 329–331.

238

APPENDIX 1

51. Plan of Bosra

Bosra. During the reign of Severus Alexander (222–235 CE), Bosra gained the title of colonia.

The Cemeteries Five or more burial grounds surrounded the city on all except the northwest side (Figure 51). Only eight tombs have been published, dating between the 2nd and 4th c. CE, and seven stelae with dated inscriptions also make up part of the sample (Online Appendix Bosra 1). The cat. 2 assemblage includes 306 (fragments of) tombs (Online Appendix Bosra 2). Thus far, no tombs have been securely dated to the Nabataean period (late 2nd c. BCE–106 CE), but the tumuli in the West Cemetery may date from this period, and perhaps some of the stelae. Reports also mention pre-2nd c. CE tombs in the Tell Aswad Cemetery. Due to the lack of plans, little can be said about the internal layout and chronological development of the burial grounds.

BOSRA

239

180 160 140 120 100 80 60 cat. 1 40

cat. 2

20 0

chart 14. Distribution of tomb types, Bosra

The distribution of types is visible in Chart 14 (Figure 10). The construction of large rectangular mausolea was confined to the period after the creation of the province of Arabia in 106 CE. Evidence from the inscribed lintels in Latin that perhaps once adorned standing funerary architecture also suggests a later date of mausoleum construction. The greatest diversity of architectural types also dates to the Roman period (106–330 CE), and lasted until the 5th or 6th c. CE. The tombs have not yielded any grave goods, but the cat. 2 collection included a golden necklace, lamps, and glass vessels. Southwest of town extended the Tell Aswad Cemetery. Although many tombs have been reported from this location dating between the 1st c.BCE/CE and the 5th/6th c. CE, only four (mausoleum, hypogeum, sarcophagus group, and a stele) have been published in detail (cat. 1). The tombs lay just outside the southern enclosure wall and close to the amphitheater and circus, both built after 106 CE. The cat. 2 material includes more sarcophagi in the open air, a hypogeum or shaft-tomb, stelae with Greek inscriptions, a Latin inscription, and a possible mausoleum. According to Sartre-Fauriat, the area also included lintels with inscriptions in Greek and Latin, indicating the presence of more mausolea. Another cemetery is marked on maps southeast of town, between the theater and the western reservoir (Southeast Cemetery). Reports mention Greek, Nabataean/Aramaic, and Latin funerary stelae and cippi from this area, as well as sarcophagi, aniconic stelae, and a tomb for a child of a Roman soldier (cat. 2). No tombs were added to the cat. 1 database.

240

APPENDIX 1

The West Cemetery extended along the ancient road outside Bab el-Hawa. From this cemetery originated a tall circular mausoleum and inscribed altars and cippi (2nd or 3rd c. CE, cat. 1). Sartre-Fauriat and al-Maqdissi also mention pit-graves (fossae), tumuli, and sarcophagi in the open air, lintel inscriptions, and a small lead sarcophagus or ossuary (cat. 2). Some of the basalt coffins were decorated with lion heads. Greek inscriptions were placed on stelae, cippi, and unidentified stone blocks. The presence of tumuli mentioned in the reports indicates that the cemetery had an earlier phase of the 1st c. BCE–1st c. CE or even earlier, based on comparison of such tombs from other sites in the Hauran. On the left side of the road leading north out of town,excavations uncovered a mausoleum (cat. 1) and a hypogeum (cat. 2) of Late Roman and Byzantine date (North Cemetery), with reused elements of older tombs. This location was 3 km north of the theater of Bosra and ca. 900 m beyond the north wall of the military camp. The date of the published tombs could indicate that the cemetery was in use later than the other burial grounds of Bosra. The area north and east of Bosra was littered with the remains of funerary architecture, according to Sartre-Fauriat, who postulates that the burial ground(s) extended along the road for at least 1 km east of the theater. The area is subdivided into the North and Northeast Cemeteries, but it is not clear whether these were distinct burial grounds. Three large hypogea with built interiors, dating to the 2nd or 3rd c. CE, rose perpendicular to the road (East Cemetery). Stelae, sarcophagi, and lintels with Nabataean/Aramaic, Greek, and Latin epitaphs were also reported close to this area. Similar materials come from the Northeast Cemetery, according to Sartre-Fauriat. A separate section may have been dedicated to the burial of soldiers of the Legio III Cyrenaica. The Northeast Cemetery was used in the Byzantine period, as well. In total, 248 inscriptions are published from Bosra: Greek (192), Latin (37), Latin and Greek (1), and Aramaic/Nabataean (18)16 texts on stelae, lintels, sarcophagi, and sometimes sculptural busts. Eight were part of cat. 1 tombs. The majority of the Latin inscriptions date between the early 2nd and the 4th c. CE, and they were mostly published without provenance. They are found on altars, stelae, cippi, and pieces of architecture. Most or all of the stelae belonged to soldiers of the Roman army, most prominently from the Legio III Cyrenaica. The texts begin with a dedication to the Manes, then add the name, rank and unit, and age of the deceased, and sometimes the name of the dedicator. Male adults form the largest category of recipients of the stelae. However, thirteen stelae were erected for family members – such as wives, mothers, and children – of the soldiers. The most common 16

This number was likely much higher; see Nehmé 2007, 16; 2010.

DEB’AAL

sponsors of the graves were wives, fathers, and husbands. One centurion dedicated two separate inscriptions to two adopted sons. The inscriptions provide clear evidence that the soldiers were living in Bosra with their wives and children, and sometimes with parents. Two freedmen dedicated a stele to their former master and soldier, and one master dedicated to three freed slaves. Greek or bilingual inscriptions could also be associated with the military, children (mostly male), and one with a soldier who died in Mesopotamia. Since there was a high level of reuse and the town is still partly inhabited, the distribution of the (portable) blocs with inscriptions provides little evidence for the presence of cemeteries. The area close to the military camp yielded many stelae and altars of soldiers of the Legio III Cyrenaica, which, according to Sartre-Fauriat, belonged to a military cemetery. Stelae belonging to soldiers of the same legion, however, were also uncovered in other cemeteries.17 Nabataean/Aramaic inscriptions were mostly carved on stelae, but could also adorn a sarcophagus or (in two cases) stone block. Most were short, adding just the double name of the deceased, equally distributed over the genders. The remaining 192 inscriptions were in Greek, and were carved on stelae, stone blocks, cippi, altars, and coffins. The cat. 1 examples (2) date to the 2nd and the 4th c. CE. The cat. 2 sample is mostly undated, but includes some 3rd– 4th c. CE examples. The area also yielded multiple Byzantine-period examples. The texts usually included a single or double name, and often an age. Sometimes, “courage” was added. The texts were erected to men (45), women (31), male children (16), and female children (7). The dedicators were usually men (husbands, sons, fathers), but also included a wife and mother (2) and one slave (who dedicated the tomb to themselves). In seven cases, the deceased is marked as “councilmember” (bouletes), and the sample included one mosaicist. The cat. 1 sample does not include sculpture, but a small number of the cat. 2 stelae included figural and other reliefs: a sarcophagus lid, made of Attic marble, with a sculpture of reclining man and woman on decorated mattress; a headless bust; a bust of a veiled woman; a bust; an altar with floral decoration; a basalt stele with a large decapitated, toga-clad bust on top; a small female bust; and a basalt stele depicting a headless woman and a bust of a little boy.

DEB’AAL18

In 1961, the Lebanese Antiquities Service excavated a large hypogeum at Deb’aal, 14 km east of Tyre. The tomb was not associated with a settlement and it is unclear whether it was part of a cemetery. The construction date was between the late 1st and early 2nd c. CE, and it was used until the early 17 18

For instance, IGLS XIII/1, 9176, 9178, 9172, 9198. Hajjar 1965; Koch & Sichtermann 1982, 572; Rahmani 1999.

241

242

APPENDIX 1

52. Deb’aal 1: plan of tomb

3rd c. CE (Figure 52, Online Appendix Deb’aal). Elaborate painted decoration of garlands of fruits and flowers, pediments, columns, and rosettes adorned the walls around the loculi. The tomb contained thirty-four coffin burials, including twenty-nine lead, four stone, and one terracotta example. Iron nails in three loculi demonstrate the use of wooden coffins. The coffins were placed on the floor of the central chamber, in a pit-grave, but mostly in loculi. Here, they stood side by side,on top of one another,or separated by a short wall built inside the loculus. In three cases, a lead coffin occupied a stone sarcophagus. One sarcophagus carried a Greek inscription dated to 136 CE. The tomb yielded 123 artifacts in thirty-three loculi (Chart 15). The numbers of glass vessels, lamps, and coins were high when compared to provincial averages. Two lamps stood in a loculus, but outside a sarcophagus. Six coffins are listed as belonging to child burials, presumably based on their size. DURA EUROPOS19

The ancient site of Dura Europos is located on the south bank of the Euphrates, near the modern border between Syria and Iraq. The city was founded as a 19

Allara 1991; Downey 2000; James 2007; Leriche & Gelin 1997; Leriche et al. 1992.

DURA EUROPOS

45 40 35 30 25 20 15 10 5 0

chart 15. Distribution of grave goods, Deb’aal

military fortress around 300 BCE, possibly by one of Seleucus Nicator’s generals. The initial settlement consisted of a small military garrison on and around the citadel, on a cliff overlooking the river. Expansion took place in the 2nd c. BCE, and the city reached its full size in the Parthian period (113 BCE– 165 CE). The Roman conquest in 165 CE, and especially after 210 CE when the Roman garrison at the site increased in size, modified the cityscape of Dura Europos. A Roman camp with barracks and a palace or headquarters dominated the northern sector of the settlement. Dura Europos received the status of colonia in the 3rd c. CE. At some point, Durenes controlled a territory extending to the confluence of the Euphrates with the Khabur in the north and down to Abu Kemal on the modern border with Iraq.20 The city fell to the Sasanian army in 256 CE, after which its population was reportedly deported. Excavations in the southeast quarter have revealed one or two burials in ovens.21

Pre-Roman Cemeteries22 The excavators identified over 1000 tombs outside the city wall, and probably 100 more were located under the spoil heaps of the 1920s and 1930s excavations (Figure 53). Of the 107 published tombs, thirteen dated Hellenistic (303–113 BCE), fourteen Hellenistic–Parthian (303 BCE–165 CE), thirty-nine Parthian (113 BCE–165 CE), and eleven Roman (165–256 CE). The remainder 20 21

22

Geyer & Monchambert 2003, 273–274. Saliou & Dandrau 1997, 104–105. Toll (1946, 6, 12) mentions four pit-graves covered with tiles in the cement floor of the Zeus Dolichenus temple, presumably constructed after the temple was abandoned. Buchet 2012; Downey 1977; Leriche 2001, 2012; Matheson 1992, 126–127; Pfister & Bellinger 1945; Rostovzteff 1937, 198; Toll 1946.

243

244

APPENDIX 1

A

B

53. Plan of Dura Europos. A: City and cemetery. B: Detail of cemetery

DURA EUROPOS

were undated. Most tombs were located outside the city wall, in an area stretching from the northern to the southern wadi and beyond, and between the city wall and the main northwest–southeast road. The Hellenistic tombs consisted of small hypogea with a few loculi (1–3) in the mound opposite the main gate, as well as a group of cist-graves found about 800 m east of the city wall. Four Hellenistic tombs contained grave goods, a low proportion compared to the general percentage of tombs at Dura Europos containing such items. The finds consisted of jewelry (rings, earrings, beads), pottery and bronze vessels, coins, and a piece of a woolen tunic (see Table 4). Five cist-graves contained the remains of five adult individuals, with two possible males and two females. One female was aged around 60 years. One terracotta slippershaped sarcophagus found by the city wall was either Hellenistic or Parthian in date. The Parthian-period tombs were found in several places. At least nineteen of the thirty-nine Parthian tombs came from the mound, and seven were located elsewhere in the cemetery. Several tombs were reported inside the settled area: one was located in tower 10 of the city wall and was probably dug after the tower was built. A group of twelve pit-graves formed a small necropolis on the citadel, which in this period had been abandoned after part had fallen into the river.23 The Parthian assemblage included a greater variety in tomb forms than before: hypogea (19), pit-graves (12), stone tower-tombs (7), and a reversed tub-sarcophagus.24 The hypogea were larger than the Hellenistic types, especially between 100 BCE–100 CE, with an average number of loculi in the hypogea of twenty-three. An earthen mound covered many hypogea, one of which carried an inscription in Greek, mentioning a date (36 CE). Seven tower-tombs were erected on the edge of the necropolis of Dura Europos. The towers perhaps dated between 50 BCE and 50 CE. A total of twenty Parthian tombs, mostly hypogea, contained grave goods, on average forty-five goods per tomb. Items of personal adornment were most common; other finds consisted of vessels, coins, tools, weapons, a lamp, and figurines (Table 4). In addition to the loculi, finds were placed in the chamber on benches in front of the loculi or near the entrance, such as pottery vessels and, in one case, a basket filled with glass bottles. In four pit-graves, the body was found wrapped in a decorated shroud.

23

24

A sarcophagus of Hellenistic or Parthian date was found inside the city at the foot of the city wall between towers 4 and 5. A recent catalogue listed a statue of a Palmyrene girl found in fragments in several locations inside the city. It is dated to 100–150 CE. The reason for the identification as a funerary statue is not certain (Brody & Hoffman 2011, 374). A shoe-shaped terracotta sarcophagus decorated with masks and a garland was found in Dura Europos. This coffin dated to the Hellenistic or Parthian period (Kurtz & Boardman 1971, 211). Henning (2013, 102) mentions an eight-tower tomb.

245

246

APPENDIX 1

12 10 8 6

cat. 1 cat. 2

4 2 0 pit-grave

hypogeum

jar-burial

stele

chart 16. Distribution of tomb types, Dura Europos

The Roman Cemetery25 The extramural cemetery of Dura Europos remained in use after Roman armies conquered the site in 165 CE. Eleven tombs stemmed from the Roman period. The excavator dated three tombs from the mound area to the final century of Dura’s existence, and thus the Roman period (Online Appendix Dura Europos 1). Five tombs from this area stylistically dated to Roman or slightly earlier phases, and are added to cat. 2 (Online Appendix Dura Europos 2). Three cremation graves in coarse-ware jars came from the mound area. Because this practice was usually connected to Roman soldiers, it is likely that these are Roman in date (see p. 110). Toll notes that interspersed among the hypogea were numerous simple ditches with single burials, presumably pitgraves. Based on stratigraphy, some of these were perhaps Roman. Four tombs were at a distance from the city and mound with graves, between 273 and 360 m west of the city wall and between 186 and 450 m from the mound. Based on typology and location, Toll dates these to shortly before the fall of the city. Their plaster walls were relatively fresh, and in one case the prepared coating was left on the floor. One stele with a Latin inscription was reused in the floor of the Artemis temple, but it is unclear at what date (Figure 34). It probably belonged to a Roman soldier depicted in reclining position on the relief of the stele. The temple itself was refurbished several times in the Roman period.26 The Roman tombs lay scattered between the older tombs and the periphery of the burial ground (Chart 16, Figure 9). The majority were hypogea (7) covered by an earthen mound and consisting of a square or trapezoid chamber with benches along the walls. They held on average fourteen loculi and 25 26

Baur & Rostovtzeff 1929; Leriche 2001; Matheson 1982; Toll 1946; Toynbee 1971. Downey 1988, 89–91.

DURA EUROPOS

6 5 4 3 2 1 0

chart 17. Distribution of grave goods, Dura Europos

were smaller than the Parthian examples. The loculi were wider than in earlier periods, or alternated between being narrow and broad. Plaster and moldings decorated the interiors. The loculi often had arched ceilings or alternating arched and gabled ceilings. Several hypogea remained unfinished, and others were cut into older tombs. The tombs yielded the remains of several terracotta and wooden sarcophagi, including a green glazed sarcophagus decorated with portraits/heads separated by columns.27 Hypogeum 4, perhaps constructed right before the fall of the town, had two Greek graffiti above the loculi. The Roman-period hypogea, although slightly smaller, represents a direct continuation of older practices. New types are represented by cremations in jars (3) and the funerary stele. Another carved stone is interpreted as a stele depicting a bust of a veiled woman (cat. 2). At least ten pit-graves and a jar-burial of a child are recorded close to the hypogea in the mound (cat. 2). Many more hypogea covered with tumuli are reported from the area ca. 3 km west of Dura (cat. 2). Five tombs contained finds (18), all of which were discovered outside the burial loculi (Chart 17). It is not indicated whether this was the original placement or the result of later removal or robbing activities. The loculi in the tombs in cat. 2, however, contained numerous finds, as well as items deliberately placed on benches outside the loculi (a basket turned upside down filled with glass bottles, two green glazed vases, two coarse-ware bowls and a glazed vase standing at the front of entrances to the loculi on the bench, and two iron arrowheads on the bench in front of a loculus). The cat. 2 tombs yielded far greater numbers of finds. Five contained 272 finds, of which eighty-nine were found outside the loculi, including a large proportion of rings, bronze bells, glass bottles, arrowheads, and bronze mirrors. 27

The type was unusual and had parallels with Parthian examples at Seleucia-on-the-Tigris, but Toll (1946, 97–99) dates it to the late 2nd c. CE.

247

248

APPENDIX 1

54. Plan of Hama

The reports mention the discovery of nine individuals in the central chamber of Tomb 3, but do not offer further comments on their location and placement. The cat. 2 tombs yielded eighty individuals, of whom thirty-three were discovered in a single pit-grave. The children were buried in jars, and the interments in the hypogea may have taken place in wooden coffins. HAMA28

The large modern city of Hama, ancient Hamath, is located on the west bank of the Orontes River (Figure 1). The site consists of a high mound (336 × 215 × 46 m) and a lower town extending south of it (Figure 54). It may have been 28

Christensen & Johansen 1971; Christensen et al. 1986; Foss 1997, 230–237; Ingholt 1934, 1940, 1942; Lund 2003; Ploug 1985; Riis & Buhl 2007, 15–25.

HAMA

refounded as Epiphania in the Hellenistic period. A considerable rebuilding phase on the mound perhaps dated to the 2nd and 3rd centuries, and was followed by another in the Byzantine period.Archaeologists date the expansion beyond the tell to the 3rd–4th c. CE. The Great Mosque of Hama may stand on a Roman temple. Remains of villas from this date were discovered in the vicinity of the town. Material evidence from the Roman period is scarce, and it is doubtful whether Hama should be considered an urban center. Habitation continued in the Byzantine period, and the countryside around Hama was densely populated in late antiquity.

The Cemeteries29 A total of sixteen tombs have been published, dating between the 1st and 6th c. CE, with a concentration after 200 CE (Online Appendix Hama 1). Three find spots southwest of the citadel yielded tombs, which lay at some distance from one another and from the tell. Tomb G XXIX (101 CE) was located ca. 500 m from the southern tip of the mound and 200 m from the Iron Age cemetery. No contemporary tombs were found in the vicinity. The south slope of a hill in the Karm el-Haurani area, approximately 1.1 km to the southwest and 1.65 km from the southern tip of the tell, yielded fourteen graves (2nd–6th c. CE) flanking the modern road. They covered an area of ca. 47 × 37 m, but the extent of the cemetery is unknown. The tombs were irregularly and relatively closely spaced (1–10 m); one tomb cut another. The other end of town in the Madina quarter, 2 km west of the Great Mosque and 2.5 km west of the tell, yielded a 4th c. CE tomb (Tomb So. 18). This location was close to the ancient road running west out of town. The cat. 2 sample includes a lead coffin found at sounding 14 (Online Appendix Hama 2). Perhaps this location corresponded with So. 14 on Ingholt’s map, which would place it midway between the Karm el-Haurani area and So. 18. A stele also originated from Hama and was added to the cat. 2 sample, although its funerary nature remains uncertain. The burial grounds of Hama yielded two types of funerary architecture (Chart 18, Figures 9, 31). The original placement of the cat. 2 stele and lead sarcophagus is not known. The pit-graves (after 1st c. CE) came from the Karm el-Haurani area and followed a similar north–south orientation. The hypogea appeared in all three locations and contained burials in stone sarcophagi, in loculi, and on the floor. Four different shapes have been identified, perhaps indicative of chronological development.Two tombs (G XXI and XXVIII) had depositions on the floor of the single chamber.The finds dated to the 2nd c.CE, but the shape, with rounded edges and burial on the floor, was unusual, and 29

Christensen et al. 1986; Ingholt 1940; Will 1965.

249

250

APPENDIX 1

12 10 8 cat. 1

6

cat. 2 4 2 0 pit-grave

hypogeum

stele

coffin

chart 18. Distribution of tomb types, Hama

reminiscent of Persian-period tombs.30 One hypogeum (G XV) followed an irregular plan consisting of four side chambers with loculi (3rd c. CE). Another hypogeum (perhaps 4th c. CE) had a single chamber with three parallel loculi (G XXVII). The final type consisted of two rectangular rooms with a vaulted ceiling and loculi perpendicular to the room (Tomb G XXIX/Habbasi Tomb). Shallow niches were carved above the loculi in this tomb, of which eight still held funerary sculpture dated to the 2nd c. CE: three busts of women, a statue of a woman in Isiac garb, a statue of a boy, two male busts, and a plaster cast of a death mask of a man (Figures 31, 43). One of the niches with a female portrait carried a Greek inscription dated to 101 CE. Crosses, curtains, and other inscriptions were added in paint in the Byzantine period. Parlasca argues that the plaster cast, along with another one of a woman found elsewhere in the tomb, originally decorated wooden panels closing the loculi, in the style of the Palmyrene closing slabs.31 The grave good assemblage from Hama consists of 294 artifacts from twelve tombs (Chart 19, Figure 25). These are high numbers, 4.4 on average for single tombs and 7.7 per burial spot inside the communal tombs. In comparison to the general patterns in Syria, the Hama tombs contained fewer vessels and coins, and more rings, bracelets, pendants, and bronze bells, as well as items of clothing such as belt buckles and shoes. Assemblages of pit-graves and hypogea differed significantly. The pits, with an average of four finds, contained jewelry, clothing, glass bottles, and bronze mirrors. Coins and lamps were absent. The hypogea included many more finds (7.7 per burial spot), mostly consisting of 30 31

Persian-period tombs with a similar layout are listed in Stern 1982, 82–83. Parlasca 2006; the excavator also commented on the parallels with Palmyrene sculpture (Ingholt 1940). One limestone statuette of woman is identified by Ploug (1985) as a goddess and by Parlasca as a private person portrayed as an Isis priestess.

THE HAURAN

140 120 100 80 60 40 20 0

chart 19. Distribution of grave goods, Hama

jewelry, clothing, and high numbers of lamps. Vessels in hypogea were fewer. The entrance of Tomb G XV was flanked by four amphorae (two on either side).Within Tomb G XXIX,the stone sarcophagi placed inside the loculi contained more finds, and more jewelry and shoes, when compared to the loculi lacking a stone sarcophagus. The communal tombs had fewer finds on average per individual (2.3 compared to 3.7 for the single tombs). The distributions differed too: the communal graves contained many lamps, some coins, some vessels, and items of personal adornment. The single graves lacked lamps and coins but contained more jewelry. The tombs yielded the remains of 114 individuals. Three pit-graves contained a single individual; the remainder stemmed from the hypogea. Eleven individuals were deposited, perhaps in wooden coffins, on the floor of G XXI. The skeletal material from G XV and G XXIX (Habbasi Tomb) were aged: eight infants, twenty-six children, and sixty-six adults. Adults and children were co-buried in loculi, and adults were also buried alone. Few of the remains were sexed: eight males and six females from eight loculi in two tombs. THE HAURAN32

The region of the Hauran is not well defined in modern scholarship or ancient sources. In this book, I concentrate on the (modern) Syrian section of the Hauran that includes the hill country of Jebel el-’Arab/Jebel Druze, the lava 32

Dentzer 2003; Graf 1992a; Kennedy & Freeman 1995; Millar 1993; Sartre 2005, 2007; SartreFauriat 2001.

251

252

APPENDIX 1

55. Map of the Hauran

plateau of el-Leja, and the Nuqra plain, which extends west and south of these areas (Figure 55). Bosra and Der’a (Adraa) form the southern border of the region and Shabh¯a (Philippopolis), Qanawat (Kanatha), Si’, and Suweida form the western edges of the Jebel el-’Arab. The area consisted of pasturelands and water reserves on the basalt plateau and on a fertile plain (Nuqra). It bordered the desert and probably housed (semi-)nomadic pastoralists in addition to farmers. Political control over the region changed hands frequently. Part was incorporated into the Roman province of Syria in 64 BCE, and part stayed in

THE HAURAN

the Nabataean kingdom, the client-kingdom centered around Petra in southern Jordan, which had expanded northward in the mid or late 2nd c. BCE. Throughout the 1st c. CE, control of the region alternated between clientkings of the Herodian dynasty, Nabataeans, and Roman governors. When the Romans annexed the Nabataean kingdom in 106 CE, the southern Hauran became part of the new province of Arabia, with Bosra as its primary center. Small-scale architectural development commenced in the late 1st c. BCE or slightly earlier. The later Nabataean and the Roman centuries witnessed exponential settlement growth, large-scale construction in the towns, and agricultural development. By the 4th c. CE, dense habitation covered the Hauran, especially the mountainous area to the north of Jebel el-’Arab. Bosra, Qanawat, Si’, and Suweida represented larger centers, but most inhabitants lived in villages. The political and administrative status of these villages is not clear. Some probably fell within the territory of the larger towns, whereas others appear to have been independent, and to have been able, for instance, to construct public buildings with their own funds. Inscriptions indicate that veterans of the Roman army set up residence in the northern part of the Hauran.

The Cemeteries Early explorers identified hundreds of tombs scattered over the landscape of the Hauran.33 In 2001, Sartre-Fauriat published a comprehensive study of the funerary architecture of the region, recording over 100 monumental tombs, 2000 stelae, and seventy-five sarcophagi in the open air.34 The vast majority of these remain unexcavated and undated. The cat. 1 assemblage includes sixtythree tombs, and the cat. 2 a minimum of 486 (Online Appendix Hauran 1, Hauran 2).35 The tombs came from twelve sites, including small villages and farmsteads, as well as the larger centers of Qanawat, Shahb¯a, Si’, and Suweida. The sites are described individually later in this section. A preference for locating the cemetery north and west of town may have existed,at least for the larger settlements. The assemblage from the Hauran material reveals two chronological stages. The earliest burial grounds grew between 100 and 1 BCE, around the same time as the resettlement of the region. A second phase occurred in the 2nd and 3rd c. CE, when the region was part of the Roman province of Syria and Arabia. In this period, the cemeteries became more crowded and displayed a greater variation in tomb type and decoration. 33

34 35

Most notably, Waddington in 1862 and Butler in the first years of the 20th century (SartreFauriat 2001, 1). Sartre-Fauriat 2001, 4, 217. This number excludes the fifteen tombs from Bosra, which are discussed in a separate section of this appendix.

253

254

APPENDIX 1

300

250

200

150

100

cat. 1 cat. 2

50

0

chart 20. Distribution of tomb types, Hauran

Seven tomb types can be distinguished in the Hauran (Chart 20). The cat. 2 sample added to the assemblage numerous stelae and sarcophagi, and perhaps a funerary enclosure. Local basalt was the standard material for the masonry of tombs, stelae, and sarcophagi. Characteristic for the Hauran were hypogea with a built masonry interior. Round buildings such as tumuli and circular mausolea were characteristic of the period between the 1st c. BCE and 1st c. CE. By the 2nd c. CE, rectangular mausolea had replaced the circular forms. These later mausolea were more decorated on the interior and exterior, and often inscribed. From the 2nd c. CE onward, the cemeteries included a greater diversity of funerary architecture. The cat.1 sample included few decorated coffins and stelae,but the cat.2 collection, as well as additional examples published by Sartre-Fauriat and Parlasca, gives a sense of the rich, and locally distinct, visual world. The sarcophagi were made of basalt,and adorned with garlands,wreaths,geometric patterns,and cartouches. One depicted Victories holding panels with portraits, and another the battle of Centaurs and Lapiths. These date to the 2nd and 3rd c. CE. The stelae portrayed busts of men, women, and sometimes children, as well as standing or seated women. Communal scenes included three or four people, usually an adult male, female, and children.36 The sculpture in general displayed influence from both the Greek and the Roman world, as well as local elements. Dentzer 36

Sartre-Fauriat 2001; Parlasca 1982.

THE HAURAN

sees parallels with the Syro-Hittite style of the early 1st millennium BCE in North Syria.37 Grave goods (213) came from twenty-four tombs found at five sites, ranging between one and thirty-eight per tomb (Chart 21). Vessels formed the largest group (77), with higher numbers of ceramic and bronze vessels than the provincial averages. Over time (1st–3rd c. CE), the number of pottery vessels declined and the range in type of finds expanded. The items of personal adornment (65) included more rings and fewer earrings when compared to the rest of the province. Only a single tomb yielded a coin, and the number of lamps was small as well (10). Items in the “other” category, on the other hand, were frequent (62), and mostly came from pit-graves in Nawa-tell Umm al-Hauran. Most tombs were robbed, and no information about skeletal remains is available. Only five tombs yielded skeletal remains: one circular mausoleum in Qanawat contained at least three individuals, whose remains were pushed to the side to make place for new burial; two pit-graves at Nawa-tell Umm al-Hauran contained a single individual. More detailed results come from two tombs in Suweida. A hypogeum here held at least fifty-three individuals (fortyone adults and twelve children), and a mausoleum held the remains of twentyfive (sixteen adult and nine subadult). A discussion of the epigraphic evidence is incomplete, because a portion of the Greek and Latin inscriptions is not yet published, although it is scheduled for publication as part of the IGLS series. The publication of hundreds of Nabataean/Aramaic inscriptions, the majority of which were funerary in nature, is dispersed over various journals and collections, some of which are fairly inaccessible.38 The following general patterns can be detected in funerary epigraphy. The Nabataean/Aramaic stelae are the oldest, and were popular between the late 1st c. BCE and 1st c. CE. They were replaced by or supplemented with Greek epitaphs in the late 1st c. CE. Figural images accompanying the inscriptions were restricted to the Greek inscribed examples and started to appear in the 2nd c. CE. The cat. 1 tombs yielded only four published epitaphs: two bilingual Nabataean/Aramaic and Greek, one Nabataean/Aramaic, and one Greek. The cat. 2 material included 109 epitaphs: three bilingual Nabataean/Aramaic and Greek, fourteen Nabataean/Aramaic, ninety-one Greek, and one in an Arabian dialect. The inscriptions were placed on stelae and lintel, and dedicated by women, men, or groups of men. Others solely commemorated the deceased: 37 38

Dentzer 2003, 202–204. Nehmé 2010. Sartre-Fauriat mentions 400 funerary inscriptions on blocs and lintels and 2000 on stelae: 1651 Greek, 170 Nabataean, four bilingual, and three Latin (Sartre-Fauriat 2001, vol. II, 21, 102–104). The Latin epitaphs originated from Bosra (Sartre-Fauriat 2001, vol. II, 21, 102–104).

255

20 18 16 14 12 10 8 6 4 2 0

chart 21. Distribution of grave goods, Hauran

THE HAURAN

three men and two women. Children were not explicitly mentioned, except in one tomb that was built for father and son. Spouses and parents could share the tomb with the dedicator.39 The inscriptions always mentioned a double name but rarely a date. Funerary epigraphy continued or perhaps even increased in popularity in the 4th and subsequent centuries.

’Atam¯an40 South of the Nuqra plain, 5 km north of Der’a, lies the village of ’Atam¯an. Nothing is known about the ancient village, but one square temple-shaped mausoleum of the late 2nd or first half of the 3rd c. CE was located inside the modern one (cat. 1, Figure 44). Three unpublished basalt stelae were found without context (cat. 2).

Bez¯ayiz41 This site was located south of the plain of Hauran, 7 km southwest of Bosra. It yielded one circular mausoleum (ca. 100 BCE–100 CE) and two unpublished stelae with inscriptions in Greek and Nabataean/Aramaic (cat. 2).

Dhak¯ır42 On the northeastern edge of the Lej¯a lies Dhak¯ır. Two adjoining rectangular mausolea of the 2nd–3rd c. CE were erected southeast of the village (cat. 1). Three unprovenanced basalt stelae are listed as cat. 2.

Jmarr¯ın43 This site lies 2–3 km north of Bosra along the Wadi al-Zedi. It yielded several buildings, including a monumental house or farm of the 2nd c. CE and a possible subsidiary building. Ancient mills, presses, and a cistern lay on the other side of the wadi. A bridge led south to the residential quarter. North of House 1 stood a large rectangular mausoleum (Tomb 1), known only from a photograph and notes by Butler. The lower part dated to the 2nd c. CE, whereas inscriptions placed the addition of a higher level in the 4th–5th c. CE. The cat. 2 material includes stelae with Greek (4) and Nabataean/Aramaic (1) inscriptions, and a lintel inscription in both languages. 39

40 41 43

The distribution, according to Sartre-Fauriat (2001, vol. II, 112–113), was around two-thirds male and one-third female. By the 3rd–4th c. CE, this had changed to two-thirds female. Sartre-Fauriat 2001, vol. I, 12–17, vol. II, 17–18. 42 Sartre-Fauriat 2001, vol. I, 18–19. Sartre-Fauriat 2001, vol. I, 61–65. IGLS XIII/1, 341–342; IGLS XIII/2, 139–148; Bopp 2008; Butler 1915; Dentzer-Feydy et al. 2007, 310, 317–318; Milik 1958; Sartre-Fauriat 2001, vol. I, 77–79, vol. II, 18.

257

258

APPENDIX 1

Naw¯a-Tell Umm el-Hauran44 A grave field was discovered on Tell Umm al-Hauran, approximately 1.5 km north of the site of Nawa. Excavators uncovered ca. 250 pit-graves and numerous other types of burial, of which nineteen pit- and cist-graves were entered into the database (cat. 1, Figure 8). The find assemblages of five tombs dated between the late 1st/early 2nd and the 3rd c. CE. Based on similarities in tomb construction and grave goods, at least ten other tombs dated to the same period. The published map suggests that the tombs covered an area of 90–120 × 90 m, with a heavy concentration on the highest parts of the tell. The accompanying text is too limited to identify clusters based on tomb type or other characteristics. The pit-graves consisted either of a narrow shaft with a width of ca. 0.45 m, covered by a slab, and opening to a larger rectangular pit (ca. 1 × 0.86 × 0.5 m), or a large shaft (ca. 4.02 × 3.68 × 2.44 m) leading to a smaller pit (ca. 2.08 × 0.93 × 2.34 m) covered with six basalt slabs. In the latter case, grave goods were placed in the larger shaft. The cist-graves (2) followed a similar layout, with the lower pit walls aligned with basalt blocks. In total, eighteen pit- and cist-graves from Nawa-tell Umm al-Hauran yielded ninety-one finds, mostly dating to the 2nd c. CE. The graves contained relatively high amounts of grave goods and often included unusual types, such as bronze tools, strigiles, weapons, and armor. It is possible that these graves belonged to members of the Roman military (see p. 134).45 Two graves were occupied by a single person, possibly in a wooden coffin decorated with bronze. No other information is available about the human remains. The cat. 2 sample includes a hypogeum, a possible funerary enclosure consisting of three graves surrounded by walls, and two tombs of unknown shape (cat. 2).

Nimreh46 This site is located on the northern part of Jebel el-’Arab, 9 km south of Shaqqa. Nothing is known about the ancient habitation, but a tall rectangular mausoleum of the 3rd or 4th c. CE stood alongside the wadi. An inscription mentions sarcophagi close to the gates of the village, and Nimreh also yielded many epitaphs engraved on lintels, including two in meter, as well as at least eight basalt stelae (cat. 2). 44 45

46

Abdul-Hak 1954–1955; Bounni & Saliby 1956; Sartre-Fauriat 2001, vol. I, 97–115, vol. II, 9. A similar collection of weapons originated from two graves in Chisphin in the Golan, dated to the late 3rd c. CE. The site also yielded a dedicatory inscription of veteran of Legio III Cyrenaica from the 3rd c. CE (Gogräfe & Chehadé 1999). Sartre-Fauriat 2001, vol. I, 116–117. Sartre-Fauriat also lists an interesting Nabataean-Arabic inscription of a king (328 CE) published by Dussaud (1902, 409–421), but this was found at El-Nemara, which is further east than Nimreh.

THE HAURAN

Qanawat47 Qanawat (ancient Kanatha), on the northwest slope the Jebel el-Arab, first appeared in textual sources in the 1st c. BCE and is sometimes listed as one of the Decapoleis. Sporadic issues of coins indicate its status as a city in 1st c. CE. The sanctuary site of Si’, and perhaps also Suweida, was included in the territory of Qanawat. Outside the walled town, on the opposite side of the Wadi al-Ghar, stood a small theater and a nymphaeum or temple building. West of the town arose a temple dedicated to Helios. Four cemeteries surrounded the town in the west, east, north, and southeast, yielding a minimum of 151 tombs, of which twenty-two are in the database (cat. 1). The assemblage dated between the 1st c. BCE and the 4th c. CE and consisted of mausolea, tumuli, a cist grave, and a hypogeum marked by a stepped podium. The circular shapes (mausolea and tumuli) represent the earliest tombs (1st c. BCE–1st c. CE); the other types are 2nd c. CE and later. At least one circular mausoleum was in use until the 3rd c. CE. The cat. 2 sample included tumuli (71+), inscriptions (26), mausolea (18), stelae (3), inscribed lintels (3), a sarcophagus burial, a hypogeum, a bust, and several tombs of unknown shape (20). The West Cemetery extended on both sides of the main east–west road outside the West Gate. Here, thirteen tombs lined the road. The easternmost tomb (Q1) stood ca. 44 m outside the gate, and the burial ground extended at least 750 m to the west. If the cemetery also included Tomb 6a (4th c. CE or later), it covered over 1 km. Most tombs remain unpublished, except for four mausolea: three rectangular (2nd–4th c. CE) and one circular (1st c. BCE or CE). Sartre-Fauriat mentions earlier reports of fifteen tower-tombs, presumably tall rectangular mausolea, along the road between Qanawat and Suweida. Their exact location remains unclear, but they may have been part of the West Cemetery.48 The East Cemetery was located on the plateau east of the Wadi al-Ghar and yielded four tombs situated close to the theater and nymphaeum/sanctuary. A round mausoleum of the 1st c. CE, a cist-grave of unclear date, and two tumuli or circular mausolea from the 1st c. BCE or CE filled this burial ground (Figure 15). The North Cemetery was a large space directly outside the city walls, with at least twenty-four tombs scattered over an area of at least 719 × 438 m. Of the twenty-one mausolea listed on maps, ten were published: one hypogeum with an aboveground marker (2nd–3rd c. CE) and nine circular mausolea dating between the 1st c. BCE and the 1st c. CE. Q9 is the only tomb that yielded grave goods, which came from the disturbed upper layers and included jewelry, vessels in glass, and pottery. The Southeast Cemetery flanked the road to Si’ and had two mausolea (cat. 1), 47

48

RES II, 1919; Burns 1999 [1992], 195–198; Millar 1993, 418–419; Oenbrink 2000, 2003, 2006, 2010; Prentice 1908a; Renel 2010; Sartre-Fauriat 2001, vol. I, 118–136. Sartre-Fauriat 2001, vol. II, 13.

259

260

APPENDIX 1

a sarcophagus, and a tumulus or a circular tomb (cat. 2). Tomb 6b, a rectangular mausoleum of 4th c. CE or later, was perhaps part of the same necropolis. The valley between Qanawat and Si’ reportedly was filled with circular tombs or tumuli, of which Renel counted at least seventy-one (included in cat. 2 list of Qanawat). Dentzer mentions that they were inserted into the system of fields, either placed in the center or along a low wall.49 The final set of tombs at Qanawat came from unknown locations and included a female bust, a lintel with busts of two men, a lintel with three individuals, three stelae with Greek inscriptions, two lintels with inscriptions in the same language, and at least nineteen Greek and six Nabataean/Aramaic inscriptions (cat. 2).

Rdeimeh Ash-Sharqiyyeh50 Nothing is known about this settlement on the north section of Jebel el-’Arab, 8 km east of Shaqqa. It yielded a rectangular mausoleum of the 4th c. CE, whose exact location is unknown. One inscribed Greek stele and two Greek inscriptions, including one in meter, are not yet published (cat. 2).

R¯ımet al-Lohf 51 The site of R¯ımet al-Lohf occupied the southern edge of Lej¯a, 9 km southwest of Shahb¯a. One rectangular mausoleum of the mid 3rd c. CE stood on the west side of the village (Figure 10). This tomb was enlarged with a dovecot in the 4th c. CE, as evidenced by three unpublished Greek lintel inscriptions, of which one was in meter. The mausoleum was close to a building, but the chronological relation between the two structures remains unclear. A tumulus and a Greek inscription are also reported from this site (cat. 2).

Shahb¯a52 Shahb¯a or Philippopolis in the north section of the Jebel el-’Arab was one of the last towns in the Near East to receive colonia status (240s CE), because it had produced the emperor Philip the Arab (244–249 CE). The site was extensively explored by Butler and others, but little survives today and the accuracy of Butler’s map is uncertain. Survey and test-trenches revealed a large village pre-dating the expansion in the mid 3rd c. CE, perhaps covering 10 ha. The town grew in size and received an urban form as a result of its new status and connection with the emperor. The new, almost square plan followed a grid 49 50 52

Dentzer 1985, 80–82; Renel 2010, 383 n. 6. 51 Sartre-Fauriat 2001, vol. I 139. Sartre-Fauriat 2001, vol. I 140–145, vol. II, 18. RES IV, 2048–2050; al-Maqdissi 1993, 463–469; Burns 1992 [1999], 220–223; Darrous & Rohmer 2004; Freyberger 1992; Prentice 1908a; Sartre-Fauriat 2001, vol. I 151–164, vol. II, 14–15.

THE HAURAN

surrounded by a city wall, encompassing ca. 60 ha (Figure 7). Two wide main streets with porticoes defined the north–south and east–west axes, with a passageway (tetrakoinon) at the crossing. West of the main crossing extended a plaza (23.3 × 35.0 m), surrounded by an exedra building of unclear function in the west and a public building/tomb in the south. The building activity was short-lived and unfinished, likely related to the abrupt end of Philip’s reign, although a residence with elaborate mosaic floors dated around 325–350 CE. The area yielded at least twenty-three tombs, of which two fell into cat. 1. The majority of the tombs were discovered within the walled city. Early maps identified a cemetery in the northwest corner of town, ca. 300 m from the early village. A circular mausoleum with columns and sculpture, as well as a rectangular mausoleum, stood at this location (cat. 2). The date of these tombs is not certain, and it cannot be established whether they were built before the construction of the fortification wall. Elsewhere in the Hauran, circular tombs were an early phenomenon, pre-dating the construction of the walls at Shahb¯a. The vicinity of the plaza yielded two pit-graves (cat. 1). This location is also within the pre-Philippan town, but again their early date (1st c. BCE) suggests a construction before the urban expansion. One of the pits yielded eighteen pottery vessels,including cooking pots,table-wares,and storage vessels (Figure 25). Centrally located alongside the main square stood a large building connected to the commemoration of the emperor’s father, Julius Marinus. This so-called “Philippeion” of the mid-3rd c. CE is often considered a tomb, which would be unusual because of the intramural location (cat. 2). Although parallels with the exterior decoration of mausolea in the region are evident,the funerary function of this monument remains uncertain. Recently, Darrous and Rohmer have argued that it functioned as a council house, based on the presence of a low, oval-shaped wall in the middle of the room, which might have provided seating.53 The only necropolis outside the walled city extended directly outside the East Gate. One hypogeum and several stelae (4+) came from this location (cat. 2). According to Butler, this was a Christian (i.e., Byzantine) cemetery. Two stelae with Nabataean/Aramaic inscriptions, one found in a tomb, come from northeast of the town (cat. 2). The unprovenanced assemblage includes Greek inscriptions in stelae (4), lintels (3), and stones of unclear shape (3), as well as a Nabataean/Aramaic stele (1), a stele depicting a seated woman, and a mausoleum (cat. 2).

Si’54 This site on the slopes of Jebel el-’Arab is located 3 km southeast of Qanawat, in whose territory it was situated. The site seems to have served primarily as a 53 54

Darrous and Rohmer 2004, 12–15. Dentzer 1985; 2003, 184–185; Dentzer-Feydy et al. 2007; Graf 1992a, 451. Tombs: RES II, 1090; Littmann 1914; Renel 2010; Sartre-Fauriat 2001, vol. I 174–191, vol. II, 17.

261

262

APPENDIX 1

religious center, perched high on a cliff overlooking the valley. Two large courts with monumental gates, a theatron, and a temple for Baalshamin in the back of the west court made up the sanctuary. Two other temples (Temples 2 and 3) and other buildings stood on the sides of the courts. Construction here started between 33/32 and 2/1 BCE and continued until the late 2nd–early 3rd c. CE. A residential area extended east of the sanctuary, including a fortification wall with a gate and a canal or aqueduct system. Another sanctuary, northeast of the cemetery at the foot of the hill, was built between 75 and 100 CE. By the 4th c. CE, the site had been abandoned. Perhaps over 100 tombs occupied the plain north of Si’, but very few remain visible today (see description of Qanawat in this appendix, Figure 14). Maps locate cemeteries northeast and northwest of town; no plan exists for the cemeteries themselves. The Northwest Cemetery followed the ancient road to Qanawat. Four tombs dating between the 1st c. BCE and the 2nd c. CE were published from this burial ground: two tumuli and two circular mausolea (cat. 1). One unpublished sarcophagus was found close to Tomb 5 (cat. 2). The Northeast Cemetery extended along a wadi and produced two circular mausolea from the 1st–2nd c. CE (cat. 1) and three Greek lintel inscriptions (cat. 2). The locations of two tombs were not further specified (cat. 2). A stele depicting the bust of a boy also originated from Si’ (cat. 2; Figure 16). Multiple lintels with inscriptions in Nabataean/Aramaic, Greek, or both await publication. Two tumuli and a mausoleum from Si’ yielded sixty-eight finds, predominantly items of personal adornment and unidentifiable fragments of bronze, iron, and other materials.

Suweida55 Suweida, or Dionysias, occupies the northwest flank of Jebel el-’Arab. In 2nd c. CE, it received civic status, coinciding with the diminishing of the territory of Qanawat, in whose territory it perhaps was previously located. The name “Dionysias” is first encountered in the late 3rd–early 4th c. CE. Little is published about Suweida, and no plan exists, despite the fact the town’s elaborate mosaic floors dated to the 3rd and 4th c. CE. The town originally occupied the citadel and eastern lower section, and expanded toward the south and east in perhaps the late 1st–early 2nd c. CE. A section of a city wall perhaps of the 3rd c. CE was preserved over a stretch of 850 m. Tombs were found north and west of the city. The West Cemetery held tumulus graves, one of which was excavated (1st c. BCE, cat. 1). This cemetery 55

RES VII, 5040; Burns 1999 [1992], 226–227; Delhopital 2010; Dentzer 2003, 185; Dentzer et al. 2010; Kalos 1999, 73; Millar 1993, 414; Oenbrink 2006, 64; Sartre-Fauriat 2001, vol. I, 196–198, vol. II, 17.

HOMS

remained in use into the Byzantine period (see cat. 2). A large mausoleum of the early 1st c. CE, known as the Tomb of Hamrath, stood on a hill overlooking the ravine at the northwest of the ancient city (cat. 1; Figure 3). The North Cemetery flanked the road to Qanawat, as described earlier, as part of the Qanawat burial grounds. Excavations by the odeon revealed a cemetery with tumuli and built tombs, with the earliest finds dating to the middle of the 1st c. BCE. The contents of two – a hypogeum and a circular mausoleum – are published (cat. 1). The report states that the preliminary findings suggest that the hypogeum was built in the 2nd or 3rd c. CE and used until the Byzantine era. If correct, this tomb was placed inside the area surrounded by the city wall. Sartre-Fauriat mentions reports of a Nabataean/Aramaic cemetery with tower-tombs on the slopes west of the citadel and on the plateau east of Suweida (cat. 2). Several unpublished Greek inscriptions originated from this site, including one on a stele carrying the date of 96 CE. Another inscription was perhaps composed in an Arabian dialect.

HOMS56

The modern city of Homs, ancient Emesa, lies alongside the Orontes in the fertile Homs plain. A number of dynasts of the Emisenoi appear as client-kings of Rome in the historical sources in the 1st c. BCE and CE, providing military support to the Roman army. This kingdom was presumably located on the upper reaches of the Orontes river, and perhaps inhabited by a recently (partially) settled tribe. Arethusa (Rastan), ca. 10 km north of Homs, served as its main center. The Romans annexed the royal territory between 72 and 78/79 CE. Numerous Greek inscriptions stemmed from the 2nd c. CE, and Emesa, now a city, started minting coins in the middle of the same century. A boundary marker illustrates that the boundary between Emesene and Palmyrene territory was Qasr al-Hair, 90 km southeast of Homs and 60 km west of Palmyra. The two cities were also connected in other ways, and Homs was the gateway to the Mediterranean coast from the Syrian desert and steppe. Caracalla bestowed the colonia title (211–217 CE), and it was the home city of Emperor Septimius Severus’ wife, Julia Domna, whose family produced four Roman emperors in the 3rd c. CE. In the middle of this century, the town also possibly withstood a Sasanian attack, after which a local usurper named Uranius Antonius briefly emerged. Historians often connect a presumed decline to the downfall of Palmyra and Palmyrene trade after the crush of Zenobia’s rebellion in the 270s. Homs was not abandoned, however, and had grown into an important ecclesiastical center in the 5th c. CE. Little is known about the 56

Burns 1999 [1992]; Butcher 2003, 91–92, 140; King 2002, 2002–2003; Kropp 2010; 2013, 24–26, 208–212; Millar 1993, 34, 300–309.

263

264

APPENDIX 1

56. Plan of Homs

archaeology of Homs before the Middle Ages, although the central citadel has yielded Roman pottery (Figure 56). Butcher suggests that the construction of a masonry dam on the Orontes and irrigation projects in the Homs plain coincided with Homs’ consolidation as a settlement in the late 1st c. CE. Homs was a famous center for the worship of the God of the Sun, Elagabalos, and a large temple holding a conical stone is depicted on its coins. Burns states that the current al-Nuri Mosque, or Great Mosque, stands on the site of the Sun Temple and the later church of St. John. Others have placed the famous temple at the top of the tell or even outside the city, at the site of Baalbek.57

The Cemeteries58 In total, twenty-three tombs were published, including five dated to the second half of the 1st c. CE (Online Appendix Homs 1). The five tombs were similar in type and content to the undated examples, and likely more or less contemporary with them.59 The cat. 2 sample held at least 129 graves, 57 58

59

Ball 2000, 39; Burns 1999 [1992], 133; King 2002, 44. IGLS V; al-Maqdissi 1993; Butler 1903, 287; Kropp 2010, 2013; Moussli 1983; Nordiguian 2004; Parlasca 1982, 18; Saliby 1993; Seyrig 1952a, 1952b, 1953; Watzinger 1923. The exception was perhaps the large hypogeum that was built in the north section of the cemetery.

HOMS

the majority consisting of stelae, of which only the inscription was published (Online Appendix Homs 2). The dated examples ranged from the 90s (2) to the 2nd c. CE (75) and yielded only a single example from the 3rd c. CE (223/224 CE). Catacombs dating to the 4th–7th c. CE were found along the road between the Bab el-Dreib and Bab el-Tadmor.60 A sharp drop in dated tombs occurred after the mid 3rd c. CE. This could be related to a contraction of the population, as suggested by Ammianus, or to the cessation of the placing of dates in epitaphs, a trend also seen elsewhere in Syria. Surveys around Homs have identified multiple mausolea prominently placed on high points and associated with villages, dated by the surveyors to the Roman and Byzantine period.61 The cat.1 tombs originated from three locations,which Seyrig assumed were part of a single cemetery. The exact location is not certain; the description suggests that the burial ground extended to the northwest of the ancient town, a location 0.9–1.6 km from the tell and 1.2–1.6 km from the al-Nuri Mosque. The northern section of the excavated burial ground yielded five parallel pitgraves and a hypogeum in an area of ca. 23 × 17 m, spaced several meters apart. The southern section contained sixteen tombs in roughly two clusters: one with ten pit-graves and a possible cist-grave on an east–west and northeast– southwest orientation; the other with five north–south-oriented pit-graves. This section covered ca. 53 × 20 m with tombs between every 2 and 5 m. A tall mausoleum of Sampsigeramos stood close to the railway station and to the other tombs. This mausoleum, dated by a Greek inscription to 78/79 CE, was destroyed in 1911 (Figure 1). The cat. 2 tombs came from various locations in town. A collection of eleven stelae and two inscriptions on stone fragments dating between 108 and 177 CE was found in the Hamidiye Quarter, about 1 km north of the tell. They carried Greek inscriptions, and some were decorated with garlands and rosettes. Basalt was often used. Saliby indicated that hypogea were discovered by the Mosque of Khalid ibn al-Walid, which is in the northern part of this quarter and about 1.36 km north of the tell. A second area that yielded funerary remains was between the Bab el-Dreib and Bab el-Tadmor, ca. 0.8–1.2 km east and northeast of the tell. Here, an undisclosed number of hypogea of possible Byzantine date, eight basalt stelae, and three inscriptions on a basalt fragment were inscribed in Greek, sometimes with a date (120– 166 CE). The Bab el-Sba neighborhood, east of the citadel, represents the third area. Five basalt stelae and a stone fragment came from this area. They were inscribed in Greek, and three carried dates: 95, 125, and 160 CE. Saliby also recorded a hypogeum in this area. Although none of the funerary inscriptions 60

61

Saliby 1993. Saliby dates the origin to the 3rd c. CE, but seemingly based on late 3rd c coins. No other finds suggest a date before the 4th c. CE. Newson et al. 2008–2009, Philip et al. 2005.

265

266

APPENDIX 1

80 70 60 50 40 cat. 1 30

cat. 2

20 10 0

chart 22. Distribution of tomb types, Homs

were found in situ or reported as part of a tomb, they probably point to burial grounds, which by the 2nd c. CE extended north, east, and southeast of the tell. The remainder of the finds came from unknown locations, and included coffins and hypogea,as well as inscribed stelae,lintels,columns,and cippi.Many carried dates, between 90 and 223 CE, with the majority falling in the 2nd c. CE. Chart 22 gives the tomb types of the cat. 1 and 2 assemblage. The pit-graves and mausoleum were close in date; the dating of the other tombs remains uncertain. Two pit-graves and the possible cist-grave contained a terracotta coffin. From the majority of the double pit-graves came fragments of wooden coffins and/or elaborate gold, silver, bronze, iron, and lead decorative and structural fragments of similar coffins. The mausoleum of Sampsigeramos is thus far the only tomb in the Syrian region made of opus reticulatum.The late 1st c.CE tomb was decorated with engaged columns, pediments, garlands, and bucrania on the exterior façade. An early drawing depicts a pyramid-shaped roof, but the basis for this reconstruction remains uncertain. When combining the cat. 1 and cat. 2 material, the variation in tomb types and decoration increased in the 2nd c. CE and later centuries. In total, ninety-nine finds were recovered from eleven tombs: between one and twenty-four per tomb, and on average 9.7 finds per grave (Chart 23, Figures 24, 27). Compared to the rest of the province, a high proportion of the finds were composed of jewelry, clothing, and golden face-covers (62). The total number of vessels was also high (28), particularly those made in materials

HOMS

14 12 10 8 6 4 2 earring bracelet ring pin necklace pendant bell golden facecover golden mask golden wreath other jewelry textile appliques (set) textile button glass unguentarium glass bottle glass flask glass cup glass unknown shape pottery unguentarium pottery small vessel bronze/copper vessel silver vessel wooden vessel faience vessel bronze coin spindle whorl lance umbo silver helmet bronze/copper mirror glass object

0

chart 23. Distribution of grave goods, Homs

other than pottery and glass. The rest of the assemblage produced low numbers when compared to the rest of the province, and no oil lamps were recovered from the tombs. The grave goods from the double pit-graves included many unusual finds, such as a silver and iron helmet with face-mask, golden torque and bulla, large bronze basin, silver jar, copper bucket with wooden vase inside, gilded bronze central part of a shield (umbo), and iron and gold pieces of a lance. Also unusual was the use of turquoise for jewelry. Seyrig notes that some of the jewelry had Iranian influence, and that some of the objects may be older than the construction date of the tomb, perhaps serving as heirlooms. Six pitgraves contained a single individual; the information is lacking about the other tombs. One burial is identified as that of a child, but the basis for this age is not certain. The epigraphic evidence provides a fuller picture about the burying community in Homs. The cat. 1 tombs yielded one inscription, a Greek text memorializing the construction of the mausoleum by Samsigeramos in 78/79 CE. All but four of the cat. 2 entries had inscriptions, placed on stelae, lintels, column drums, sarcophagi, and a cippus. The majority were dated to the 2nd c. CE. The stelae were often decorated with garlands and rosettes, and infrequently with figural sculpture (a standing woman, a seated woman holding fruit and spindles, a seated man carrying a key). The inscriptions were in Greek. Two Nabataean examples are reported as well, but it is not certain whether these were true Nabataean, rather than an Aramaic dialect common for the Homs region. The standard text of the Greek inscriptions consisted of a date, single or double name, an expression of mourning (“lamented”), and a farewell. They specified only the deceased, with the exception of one

267

268

APPENDIX 1

57. Plan of Jebleh

JEBLEH

stele dedicated by a mother to her son, and another one perhaps by a son to his mother. The commemorated community included seventy-two adult men, twenty-two women, three male children, and one female child. Twice, the deceased were specified as soldiers in the Roman army. The area around Homs yielded similar Greek inscriptions of similar date. JEBLEH62

The modern Jebleh, ancient Gabala, is a harbor town on the Syrian coast. The site was inhabited in the Persian period (6th–4th c. BCE) and evidence for the Hellenistic settlement comes from sherds, coins, and other small finds. Coins suggest that Jebleh was part of the territory of Arados in this period. The modern street plan follows a rectilinear layout and is surrounded by a medieval city wall (Figure 57). A Hellenistic cemetery extended north of town. This burial ground was still in use in the Roman period. The harbor area yielded Roman remains, and a theater was constructed in the northeastern part of town. Little else is known about Roman Jebleh.63

The Cemetery64 A Hellenistic–Roman burial ground stretched along the Mediterranean shore north of the ancient settlement, approximately 300 m north of the medieval walls and 500 m northwest of the Roman theater. Eight Roman tombs were fully published, three of which had a Hellenistic origin (200–1 BCE). The remaining five tombs dated to the 1st–3rd c. CE (Online Appendix Jebleh 1). Seven pit-graves of unknown date were dug above Tomb G, and eight more possible hypogea of the 1st c. BCE–1st c. CE were also found in this region. Together with two sarcophagi, they make up the cat. 2 assemblage (Online Appendix Jebleh 2). The cemetery extended over an area of ca. 1.5 km north– south and 200 m east–west. The edges of the cemetery were not indicated in the publications. The tombs cluster in two groups, of which the first (A, B, C, D) lay closest to the settlement, with the second group (E, F, G, H) ca. 300 m further east. The tombs in the first group were spaced on average 22 m apart, whereas those in the second group were spaced 33 m apart. The tomb type assemblage is depicted in Chart 24 (Figure 13c). The simple and undecorated hypogea (2nd and 1st c. BCE–2nd and 3rd c. CE) were of irregular shape, with on average six burial spots – in parallel and perpendicular loculi – and pits in 62

63

64

Abdul-Hak 1958–1959; Badawi 2007; Lund 1993, 29–30; Riis 1960, 130–132; Riis et al. 2004, 61–68; Seyrig 1964, 9–22. Seyrig (1964, 21) connects the site to the Gabala, where Pausanias mentions the presence of a sanctuary of Doto safeguarding a sacred robe (Pausanias 2.1). IGLS IV; Badawi 2007; Riis et al. 2004; Saadé 1985.

269

270

APPENDIX 1

9 8 7 6 5

cat. 1

4

cat. 2

3 2 1 0 pit-grave

hypogeum

jar-burial

coffin

chart 24. Distribution of tomb types, Jebleh 16 14 12 10 8 6 4 2 0

chart 25. Distribution of grave goods, Jebleh

the floor (Figure 9e). The loculi parallel to the main room, possibly arcosolia, did not occur in the earlier, Hellenistic tombs. Fragments of several terracotta coffins were found in the tombs. Lack of stratigraphical reportage hinders the distinction between grave goods from the pre-Roman and Roman periods.The Hellenistic assemblage included pottery vessels (at least six unguentaria, fragments of two Rhodian amphorae, a pitcher, a bowl) and at least eight lamps. At least forty-two artifacts stemming from eight tombs were of certain Roman date (Chart 25). This sample includes a small number of jewelry items and a large amount of glass vessels and lamps. Two tombs yielded at least eight individuals (Hypogeum A and jar-burial B). Both had co-buried individuals, lying next to or on top of each other. Three loculi in hypogeum A contained two individuals; the contents of the other loculi are unknown. One cat. 2 coffin was inscribed with a Greek text in hexameter by the husband of the deceased.

THE LIMESTONE PLATEAU

THE LIMESTONE PLATEAU65

The Limestone Plateau is located in Northwest Syria between the modern centers of Hama, Antakya, and Aleppo. The region covers an area of 140 km north–south by 20–40 km east–west and includes seven hills:Jebel Sim’an in the north, Jebel Wastani, Jebel Doueilli, Jebel A’la, Jebel Barisha, and Jebel Sheikh Barakat in the center, and Jebel Zawiye/Jebel Riha in the south (Figure 58). The average elevation is 400–500 m and the highest point reaches over 800 m. The region was densely inhabited in the Roman, Byzantine, and Early Islamic centuries. Because it was largely abandoned after the 10th c. CE, it is currently known as the region of the Dead Cities. Two towns on the Limestone Plateau, Dana and Sermada, are mentioned in Egyptian and Assyrian texts, but no archaeological remains exist from the centuries immediately preceding the 1st c. CE. The low agricultural potential of this marginal region prohibited permanent settlement in most periods, although seasonal pastoralists likely traversed the region with their flocks. This situation changed radically in the 1st c. CE, with large-scale building activity at several locations. The construction of houses and tombs illustrates that sedentary groups moved into the region. At the same time, sanctuaries arose, often on a monumental scale, and situated on high points in the landscape.66 Tate has estimated that population numbers rose in the 1st–3rd c. CE, with a peak between 110 and 240 CE, and again in the 5th c. CE.67 Approximately 700 ancient sites dating between the 1st and 10th c. CE have been identified in surveys of the region. Most were small villages with courtyard houses built closely together. Some included public buildings such as baths, temples, and inns. Brad and Qatura were larger centers on the Limestone Plateau. The French surveyors of the region debated whether it produced surplus olive products for nearby urban centers, such as Antioch and Apamea, or whether there was a more mixed agricultural economy focused on local production and consumption.68 There is evidence for cadastration of the agricultural fields, a development interpreted as local and related to fiscal control of newly inhabited land, rather than a project of the higher, Roman authority (centuriation). Inscriptions from the 2nd and subsequent centuries, on the other hand, clearly demonstrate the presence of Roman colonists, mostly veterans, on the Limestone Plateau. As a decrease in inscriptions and construction makes clear, economic and demographic growth stagnated in the mid 3rd to 4th c. CE. This could be a result of Persian invasions and the capture of Antioch in 256 and 260. There is also evidence for bubonic plague in the region. Soon afterward, the region reached its zenith in terms of agricultural development and population size. 65 66 68

Strube 1996; Tate 1992, 1997; Tchalenko 1953, 1958; Wilkinson 2003. 67 Steinsapir 2005, 47–55. Tate 1992. Tate 1992, 1997 and Tchalenko 1953.

271

272

APPENDIX 1

58.M ap of the Limestone Plateau

The Cemeteries The explorers identified necropoleis and individual tombs at most settlements on the Limestone Plateau. Of these, seventy-seven tombs from thirty sites could be securely dated to the Roman period and were published in some detail (Online Appendix Limestone Plateau 1). A minimum of ninety-seven tombs were part of the cat. 2 assemblage (Online Appendix Limestone Plateau 2). Although only a small number of tombs were published per site, incidental remarks and photographs demonstrate that larger burial grounds extended

THE LIMESTONE PLATEAU

around the settlements. None of the tombs could be securely dated to the 1st c. CE, and the majority stemmed from the 2nd and first half of the 3rd c. CE, with a concentration of construction in the first half of the 2nd c. CE. Since many tombs were dated based on inscriptions, the chronological patterns were the result of the “epigraphic habit.” In order words, in the 2nd and 3rd c. CE it was most common to place an epitaph on the tomb. Most tombs were associated with villages and towns. Little is known about the earliest phases of the settlements (1st–3rd c. CE) and few maps are available. The relationship between the inhabited space and the burial ground, therefore, cannot always be ascertained. The most common location for a cemetery was south and southeast of the settled area. Agricultural implements, cisterns, and other constructions stood in close proximity to the necropoleis. The tombs were often at some distance – several hundred meters or more – from the residences. The cemeteries associated with small villages were generally located on mountain slopes, rocky outcrops, or wadi slopes. The necropoleis of larger settlements arose closer to the settlement and perhaps in the vicinity of public buildings. A few tombs lay in proximity of isolated large farmhouses. Tchalenko assumes that such tombs were associated with the houses, but this is not certain in every instance. A handful of single tombs and small cemeteries with three to five hypogea could not be linked to a contemporary settlement. A significant portion of the tombs used the natural surroundings to increase visibility. The hypogea, for instance, were often cut in prominent cliff façades. Sarcophagi and mausolea, if not elevated on built podia, frequently stood on rocky outcrops. In the Byzantine period, tombs, often sarcophagi, were frequently placed within or next to ecclesiastical buildings, either in the church (Dehes) or in related buildings (e.g., at Brad and Ruweiha). The assemblage of tomb types from the Limestone Plateau is listed in Chart 26. Over time, the mausolea, sarcophagi, and hypogea became more monumental, reaching their grandest shape in the Byzantine period. Hypogea and sarcophagi remained frequent after 330 CE, and mausoleum construction in particular took off in popularity in the Byzantine period. Although the general tomb types were characteristic of the Syrian province, the funerary architecture of the Limestone Plateau was distinct and followed strong regional trends in shape, tomb markers, and decoration: distyle hypogea, hypogea with decorated façade, elevated sarcophagi, tetrastyle mausolea, and pit-graves with sarcophagus lids. Within the cemeteries, less variation existed, and most sites yielded only a single (contemporary) type of tomb. The exceptional sites were Sardin, Dana-North, and Qatura. The most common form of hypogea had three loculi, resulting in a low average of burial spots per hypogeum. Interior decoration was rare. Most decorative elements were concentrated on the exterior portions of the tombs. Architectural decoration consisting of columns, pilasters, cornices, and paneled stone doors was common. About half of the hypogea included an aboveground

273

274

APPENDIX 1

70 60 50 40 30 cat. 1

20

cat. 2

10 0

chart 26. Distribution of tomb types, Limestone Plateau

marker, either a built structure that was structurally part of the tomb or a marker fully separate from the tomb structure. The most popular form in the first group was an arched porch or awning cut in the bedrock. This type, sometimes adorned with moldings, seats, and benches, recurred all over the region, starting in the late 2nd or 3rd c. CE and continuing into the Byzantine era.69 Variations on this theme, including a built porch with columns supporting a pediment or flat architrave, were popular in the same period.70 The second group of aboveground markers incorporated built structures marking the grave. The most conspicuous examples were the distyle-tombs of the 2nd c. CE, consisting of two tall columns carrying an architrave.71 The tallest example reached 12 m in height. The double columns at Qatura and perhaps Benabil were physically associated with a single hypogeum, whereas those at Sermada and Sitt er-Rum stood on top of several hypogea. Other tomb markers consisted of built structures or low walls.72 A tall, rectangular stele (over 4 m high) with an inscription marked the tomb of Tiberius at Beshindlaye (Figure 42). The hypogeum itself was 7 m away, but inscribed with the same text as the stele. The mausolea mostly dated to the late 2nd c. CE or later. Characteristic of this region was the so-called tetrastyle tomb, consisting of four columns or 69 70

71

72

See examples at Kwaro, Qatura, Sardin, Besindina, and perhaps Moshon and Millis. See examples at Baboutta, Bamuqqa, and Dana. A hybrid form appeared at Frikya, in which a sarcophagus lid surrounded by a little wall closed a small hypogeum with two burial niches. Benebil (2nd c. CE), Qatura (195 CE), Sermada (132–141 CE), Sitt er-Rum (152 CE), and Turin (2nd c. CE). Other possible distyle tombs come from Bursj Heidar (Tchalenko 1953, 37 n. 2) and Turmanin (Tchalenko 1953, 128). An early drawing in de Vogüé’s publication depicts heavy sarcophagus lids covering hypogea at Dana-South on Jebel Zawiye (de Vogüé 1865–1877, plate 78).

THE LIMESTONE PLATEAU

pillars supporting a roof that was often pyramid-shaped. A podium incorporating the burial spots accessible through a door made up the lower part of the tetrastyle tombs. Burial could also occur in coffins placed under the canopy. The construction of this type started in the 3rd c. CE at Brad and continued at least until the 6th c. Tomb 1 at Dana, of the 2nd or early 3rd c. CE, represented an early variant, combining a tetrastyle monument with a hypogeum. The collection of shapes of mausolea expanded further in the Byzantine period, with the addition of temple-tombs and pyramid-shaped roofs.73 The database also includes four pit-graves, but that this type was likely more common than the low count suggests. Extensive burial grounds of pit-graves existed, for instance, at Burdaqli and Turin. Typical of the Limestone Plateau was a cover of a sarcophagus lid (Figure 8c).74 At Besandina, an arched awning was cut over a pit-grave. The original placement of the two sarcophagi in the database is unclear. cat. 2 sarcophagi stood inside mausolea and hypogea or in the open air on top of stone podia or rocky outcrops.75 A graveyard with standing sarcophagi at Serjilla was depicted by de Vogüé.76 Only a handful of funerary stelae are in the database, and this appears to represent a common trend. Funerary stelae were not frequent in the region.77 Several sites yielded rock-reliefs, some associated with a tomb, others with no tomb present (see discussion p. 329). The tombs on the Limestone Plateau did not yield grave goods or skeletal remains. The epitaphs and figural imagery provide information about the deceased. Greek inscriptions were fairly common: forty-three were published from thirty-six tombs, and another four from the cat. 2 assemblage. They were particularly frequent in the 2nd c. CE, but their use continued into the Byzantine era, when (Christian) texts were added. All tomb types carried texts: at the entrances of hypogea and mausolea, on the lids of pit-graves and sarcophagi, and below the sculpted rock-reliefs of Qatura. A few epitaphs were placed inside the communal tombs. One bilingual inscription (of a Roman veteran) in Greek and Latin was recorded; the remainder was Greek. The inscriptions usually included the name (mostly double) of the dedicator and dedicatee(s), and often a date, but otherwise there was little standardization in the content. The exterior inscriptions usually commemorated multiple individuals, and those inside the communal tombs were directed at individual deceased. The majority were foundation inscriptions, recording the founder or sponsor of the tomb. 73

74

75

76 77

Examples at Ruweiha, Frikya, Has, Serjilla, El-Bara, Qal’at Kalota, el-Ghadfeh, and Ma’rata (Butler 1903; Griesheimer 1997a; Tchalenko 1953). Examples at Herbet Kalil. Two inscribed examples at Burdaqli illustrate that the type occurred in the mid 2nd and early 4th c. CE. Also at other sites on the Limestone Plateau, e.g., ’Arsin (Griesheimer 1997a, fig. 22), Bakirha (Prentice 1908b, 70–17), and Teltita (Butler 1903, 107). de Vogüé 1865–1877, plate 86. More examples from other sites are listed by Chéhadeh & Griesheimer (1998) and Peña et al. (1999), perhaps mostly 2nd and 3rd c. CE in date.

275

276

APPENDIX 1

The dedicators were men honoring parents, wives, or sons, or reserving space for themselves. One tomb may have been dedicated by a slave to their masters, and two slaves sponsored a tomb for themselves. Once or twice, a husband and wife team dedicated a tomb, and a woman dedicated a niche at Qatura to a son or husband. The inscriptions mentioning only the deceased were also mostly directed to men (9); the rest were for a mixed-gender group (3) and a woman (1). In comparison to other areas of Syria, relatively few dedications of and to women were preserved. Children were also rarely mentioned, or identified as such in the text. Two priests and two veterans are self-identified.78 The figural imagery consisted of two group scenes of spouses and possibly children, a reclining man, a standing man, and two standing figures. These were placed at the entrance of the communal tomb, with the exception of the reclining scene of Abedrapsas, Amathbabea, and their daughter, which was placed inside their hypogeum. The rock-reliefs in Qatura depicted eight men, two women, and more than eleven individuals whose gender could not be determined.

’Aqrabat79 The small ancient settlement on Jebel Wastani, consisting of several houses, lies ca. 100 m from the modern village of ’Aqrabat. A cemetery with hypogea was nearby on the south slope of a wadi-cliff. At least thirteen hypogea were identified (cat. 1), of which one was dated by an inscription to 321 CE. The hypogea followed more or less the same layout and incorporated three loculi. Another seven hypogea from the same cemetery remain unpublished (cat. 2).

Babutta80 The farmstead at Jebel Barisha was flanked on the south by a hypogeum with built façade of the 2nd c. CE. A second hypogeum of the 2nd or 3rd c. CE was located on the road to Borj Baqirha (cat. 1). A possible temple and a church with a baptisterium of the 5th c. CE were also found at the site.

Baftamun81 Little remains of the ancient settlement on Jebel Wastani, except for a possible quarry. Three groups of tombs covered the south slopes of the valley, and a fourth lay on the north slope. These tombs consisted of hypogea, some with 78

79 81

According to Tchalenko (discussion in Tate 1992, 291), the epigraphy of the 2nd and early 4th c. CE on the Limestone Plateau recorded seven veterans. 80 Peña et al. 1999, 41. Peña et al. 1987, 34–36; Tchalenko 1953, 35 n. 3. Chéhadeh & Griesheimer 1998, 190; Griesheimer 1997a, 166.

THE LIMESTONE PLATEAU

reliefs, and pit-graves (cat. 2). One hypogeum with a decorated façade dated by an inscription to 193 CE was published (cat. 1), but its location was unclear.

Bamuqqa82 In the late 1st c. CE, a farm with side buildings and an olive press was built at Bamuqqa on Jebel Barisha (Figure 6). At some distance, ca. 140 m southeast of the farm and next to a cistern, was a tomb of the 1st or 2nd c. CE (cat. 1). The tomb was a hypogeum with an elaborate vestibule at the end of a large sunken court, surrounded by a high wall crowned by a cornice (Figure 45). In the 3rd–4th c. CE, a village appeared north of the farm, away from the tomb.

Bazliq83 A small village on Jebel Barisha with sparse remains yielded one rectangular mausoleum of the 2nd or 3rd c. CE (cat. 1), situated to the northwest. A stele was found inside the tomb.

Benabil/Banabel84 A large house or farm of perhaps the 2nd c. CE was constructed at Benabil on Jebel el-Al’a. It was enlarged in the 3rd c. CE. Other buildings are Byzantine in date. A necropolis with a dozen or so hypogea of possible 2nd c. CE date extended on the opposite side of the wadi, ca. 200 m south of the village (cat. 2). To the same period dated the distyle tomb rising above the hypogea (cat. 1). The doorway of the hypogeum that lies directly underneath the distyle construction is 4.5 m to the northeast. It is not certain, therefore, whether the distyle marked a single tomb or a collection of tombs.

Besandina85 This small village on Jebel Wastani yielded mostly Byzantine material. A necropolis consisting of rock-cut pit-graves and hypogea extended on the hill slope between 200 and 500 m east of the village. Seven tombs dated to the 1st and 2nd c. CE: three regular hypogea, three hypogea with an arched awning, 82

83 84

85

Burns 1999 [1992], 55; Callot 2007; Griesheimer 1997a, 194; Peña et al. 1987, 52–56; Strube 1996, figs. 49, 50; Tchalenko 1953, 300–318. Peña et al. 1987, 78. Griesheimer 1997a, 185; Mattern 1944, plate XL; Peña et al. 1990, 58–64; Prentice 1908b, 26; Tchalenko 1953, 323. Peña et al. 1999, 63–64.

277

278

APPENDIX 1

and a pit-grave (cat. 1). Inscriptions marked the entrances of three tombs. One hypogeum in the same necropolis dated to 390 CE.

Besindlaya/Bchendlaya86 The site of Besindlaya on Jebel el-Al’a lies 180 m southwest of the modern village and has thus far only yielded a hypogeum marked by a tall stele or pillar, constructed in 134 CE (cat. 1, Figure 42). The hypogeum had a sunken court and decorated vestibule, similar to a tomb in Bamuqqa. An inscription on the architrave repeated the text of the stele. A house of the 3rd or 4th c. CE stood at the eastern edge of the site. Most remains of the site date to the Byzantine period, including churches and olive presses.

Brad/Barad87 Early in the 3rd c. CE, this town on Jebel Sim’an included a temple, an inn (207/208 CE), perhaps a so-called “andron,” and storage facilities connected to olive oil production. In the same period, a quarter with large houses existed to the north of these structures, as well as a bath complex (Figure 6). A tetrastyle mausoleum (2nd–3rd c. CE) stood ca. 80 m southwest of the baths, between them and the rest of the town (cat. 1). The tomb was thus on the outskirts of the settlement, but not necessarily outside it. Tchalenko suggests that the tomb and the bath belonged to the residents of the large houses. Two sets of inscriptions, one from the 2nd c. CE and one from the middle of the 3rd c. CE, may have originally adorned a mausoleum or the decorated façade of a hypogeum (cat.1). In the 4th c. CE, the area between the houses and the other buildings filled in, and a church had replaced the temple by around 400 CE.

Burdaqli88 The small village of Burdaqli in the plain of Dana was situated ca. 1 km southwest of Kis’ala. It yielded several funerary inscriptions from the 2nd c. CE, which remain unpublished. Multiple rock-cut pit-graves (cat. 2) existed in the northwestern part of town, two of which were dated by inscriptions to 164 and 310 CE (cat. 1). An unpublished tomb stood on a rocky outcrop in the vicinity of Burdaqli (cat. 2). 86

87

88

IGLS II; Burns 1999 [1992], 62; Griesheimer 1997a; Peña et al. 1990, 69–71; Tchalenko 1953, 294–295; de Vogüé 1865–1877, pl. 92, 92bis. Burns 1999 [1992], 59–61; Butler 1920, 300; Jarry 1970; Seyrig 1958, 3–4; Tchalenko 1953, 387–388. IGLS II, 287; Strube 1996, 34; Tchalenko 1953, 35, 120–121.

THE LIMESTONE PLATEAU

Dana-North89 A tetrastyle mausoleum from the later 2nd or 3rd c.CE was the only completely preserved monument at the site in the center of the Dana plain (cat. 1). This tomb stood on the northern edge of town, to the west of a large cistern that perhaps once formed part of a bathing complex (2nd c. CE). It overlooked a quarry, in which four arcosolia were carved in the rock (cat. 1). In the database, these arcosolia are counted as part of a single tomb, but this identification is uncertain. One carried an inscription with a date of 324 CE. Tchalenko mentioned another large hypogeum, presumably part of a larger collection of tombs of the 2nd c. CE. The location of this tomb remains unknown. The same is true for the 3rd c. CE hypogeum with decorated porch listed by Griesheimer (cat. 1).

Daousat el-Khadra90 This site on Jebel Wastani has yielded some ancient remains and a necropolis, which lies ca. 100 m east of the modern road. Four hypogea were published. One was covered with relief decoration on the exterior and dated by an inscription to 130 CE (cat. 1).

Deir Seta91 A series of large houses date to the Roman or Late Roman period, but Deir Seta on Jebel Barisha was already inhabited in the 2nd c. CE. One funerary inscription from the mid 2nd c. CE was published (cat. 1). The site also yielded a temple and three churches, one of which incorporated a tomb with a sarcophagus.

Frikya92 Little is known about the settlement of Frikya on Jebel Zawiye. A cemetery extended north of the town and included at least twelve tombs from the 3rd and 4th c. CE. These tombs were relatively spread out, spaced at between 5–10 and 150 m. Three cisterns were also constructed in the cemetery, which was located partly in a quarry. The types and dates of most tombs are unknown (cat. 2). The cat. 1 collection consisted of a small hypogeum topped by a built 89

90 92

IGLS II, 274; Butler 1903, 73–74; Griesheimer 1997a, 177; Prentice 1908b, 100–101; Tchalenko 1953, 117–118. 91 Peña et al. 1999, 34. Peña et al. 1987, 99–101; Tchalenko 1953, 285–286; 1958, 22. IGLS II, 118–123; Butler 1903, 278–284; Griesheimer 1997a, 168, 181, 201–202; 1997b, 300 n. 22; Parlasca 1967, 565–566; Prentice 1908b, 205–213; Tchalenko 1953, 36–37, 255; 1958, 30.

279

280

APPENDIX 1

structure of the 3rd c. CE and a richly sculpted and inscribed hypogeum of the early 4th c. CE in the hillside south of town (cat.1; Figure 29).93

Herbet Kalil94 From this location on Jebel Wastani stems a pit-grave covered with a sarcophagus lid, dating to the late 2nd or 3rd c. CE (cat. 1; Figure 8c). No information is available about nearby habitation.

Jib es-Sifa95 On top of the hill that rose west of this village on the Jebel Wastani extended a burial ground with three hypogea, one of which is published (cat. 1). Several meters from the tomb lay two cisterns. Further east, agricultural presses and more cisterns were found. The town itself yielded some remains from the Roman–Byzantine period.

Kafr ’Aruq96 This small settlement on Jebel Barisha was inhabited in the 2nd c. CE. It yielded a limestone sarcophagus with a funerary inscription from the 3rd c. CE and a hypogeum 5 km from the ruins (cat. 1). A possible temple was constructed in the 2nd and 3rd c. CE.

Kis’ala97 Kis’ala was 2 km west of Dana and included houses from the 1st–4th c.CE,with several presses and cisterns. Many 2nd c. CE hypogea were reported (cat. 2), but an inscription dated only one, to 135 CE (cat. 1).

Kwaro/Kouaro98 This small village on the western edge of Jebel R¯ugˆ /Jebel Wastani covered a surface of ca. 200 × 100 m and extended north of the modern village. A burial ground with hypogea with decorated porches stretched out over an area of 200 m on the slopes of Jebel Doueili, to the west and northwest of the village (cat. 2). No maps exist of the cemetery, but photos indicate that the hypogea 93 94 96 97 98

Tate (1992, 292) lists an inscription dated to 315 CE, but this could not be retraced. 95 Griesheimer 1997a, 172. Peña et al. 1999, 81. Peña et al. 1987, 116–117; Seyrig 1958, 23; Tchalenko 1953, 277, 284; 1958, 23. Seyrig 1958, 21; Tchalenko 1953, 120. IGLS II, 357–360; Griesheimer 1997a, 197 n. 84; 1997b, 299; Prentice 1909, 116–118.

THE LIMESTONE PLATEAU

were dug at different levels in the cliff face. Four carried an inscription, three of which dated to 221–235 CE (cat. 1). Several Byzantine hypogea also originated from the area.

Marina99 This site is located 2 km north of Ma’rret Chelf on Jebel Barisha and 70 m from the modern road. No remains were preserved, except for a burial ground with six tombs: one hypogeum of the 3rd c. CE and three similar tombs of the 3rd or 4th c. CE (cat. 1); a Byzantine hypogeum of the same type; and a rock-cut sarcophagus of unclear date (cat. 2).

Me’ez/Me’ez-Ikhkhnis100 This site lies in the plain between Jebel Barisha and Deir Sim’an. In the 2nd and 3rd c. CE, the small town incorporated a regular street plan and a square with a portico, along which stood a temple (157 CE), an andron (129 CE), and a water reservoir. Epigraphic evidence specified the sponsors of the building program. Settia Secunda provided for the portico, the pavement, and the tiles of the andron. The temple was constructed by Mikkalos Zaarougas. Later architecture included a Byzantine church with a baptistery and tomb, and a second church. One mausoleum (193 CE) stood on the west slope,500 m southeast of the ruins, of which four inscriptions remained (cat. 1). Another possible mausoleum and multiple funerary stelae are included in the cat. 2 assemblage.

Millis101 Little remains of this site on the eastern edge of the plain of Rouj Armenaz in the Jebel Douelli region. Three tombs originated from the area to the southeast of the village: one hypogeum with an arched awning, marked by an inscription of 193 CE (cat. 1), and two possible hypogea of the Byzantine period.

Moshon102 The only find at this site on Jebel Zawiye was a mausoleum with a large arched awning carrying an unpublished inscription of 314 CE (cat. 1). 99

Peña et al. 1987, 165–168. IGLS II, 316–321; Burns 1999 [1992], 158–159; Jarry 1982; Peña et al. 1987, 170–171; Tchalenko 1953, 280–281. 101 IGLS II, 349; Griesheimer 1997a, 180, n. 32; Peña et al. 1990, 217. 102 Griesheimer 1997a, 180 n. 33.

100

281

282

APPENDIX 1

59. Plan of Qatura

Qatura/Katura103 The ruins of this site on Jebel Sim’an covered an area of 250 × 200 m and included a Roman-period residential quarter (Figure 59). The exact location of this quarter is not clear, but it was probably north of the cemetery area. To the south stood a building that has been interpreted as a farm. Tchalenko speculates that this farm housed the owners of the distyle tomb that was erected ca. 50 m to the east of the building (195 CE, cat. 1; Figure 45). A cemetery of contemporaneous date extended ca. 50 m west of the farm. Here, in the cliff 103

IGLS II, 246–251; Burns 1999 [1992], 202–203; Butler 1920, 249–250; Griesheimer 1997a, 185; Prentice 1908b, 127–130; Tchalenko 1953, 189–194.

THE LIMESTONE PLATEAU

façade (up to 8 m high) along the western road to Zerz¯ıta/Zarzita, rock-reliefs of persons were carved (Figure 18). A minimum of thirteen can be identified in photographs, of which five were published (cat. 1, 2). These five reliefs carried commemorative inscriptions below the image, and one epitaph included a date in the 2nd c. CE. The reliefs were not associated with a grave, and Butler believed that pit-graves were located further below them, at the bottom of the ravine.104 Griesheimer and Peña suggested that the reliefs served as cenotaphs and did not mark physical tombs.105 At the edge of the same cliff, a little removed from the niches, a hypogeum was cut in the late 2nd c. CE. Another hypogeum of the mid 3rd c. CE was discovered cut in the quarry, about 110 m north of the farm. Both are in the cat. 1 collection; the cat. 2 database includes another hypogeum.

Ir-Rubbeh106 A small modern village lies on top of a fortified tell in the Jebel ‘Ala region. Several Greek inscriptions and architectural features were reused in later buildings. From this location originated a stele with an inscription of the 2nd or 3rd c CE (cat. 1) and a second example that was not published (cat. 2).

Sardin107 A cemetery was published in association with Sardin on Jebel Barisha, although it lay 2 km southeast of this site. The small cemetery included five tombs: three hypogea with arched porches and inscriptions of the 3rd c. CE (cat. 1), one pit-grave with a sarcophagus lid, and one unpublished hypogeum (cat. 2).

Sermada108 Although Sermada, on the plain of Dana, was known from pre-Roman sources, the standing architectural remains of the ancient town were sparse. Approximately 1 km to the east extended the necropolis, consisting of hypogea with dromoi. A tall distyle tomb of the 2nd c. CE arose on top of the cemetery and in front of a quarry (cat. 1). The monument stood on top of several staircases or dromoi of hypogea (cat. 2), and it is not certain whether it was associated with a single tomb. Another hypogeum from the early 2nd c. CE was situated between these tombs and the village (cat. 1). 104

105 Butler 1920, 249–250. Griesheimer 1997a; Peña et al. 1999. IGLS II, 347; Butler 1908, 6–7; Prentice 1908b, 2. 107 IGLS II, 345–346; Griesheimer 1997a, 180 n. 32, 193; 1997b, 299–301. 108 IGLS II, 284–287; Froment 1930, 182–183; Griesheimer 1997a, 185; Prentice 1908b, 96–97; Tchalenko 1953, 121–124. 106

283

284

APPENDIX 1

Sitt er-Rum109 A farm represented the oldest remains at this site on Jebel Sim’an. A distyle tomb (152 CE) stood ca. 400 m southeast of it, on the south side of the wadi (cat. 1; Figure 45). The area between the farm and the tomb was later filled with a village that included a 4th c. CE church. The location of Sitt er-Rum was located near Ref¯ade, which yielded one inscription of 73/74 CE and several houses of possible 2nd c. CE date. Since the distyle tomb stood about 700 m southeast of Ref¯ade, it may have been associated with this town. Other tombs from Ref¯ade dated to the late 4th c. CE.

Tell ’Aqribin110 The ancient town was close to the tell of the site in the plain of Dana, and yielded buildings from the 6th c. CE. One inscribed sarcophagus (222 CE), the exact location of which remains unclear, stemmed from this site (cat. 1).

Turin111 The ancient settlement on Jebel Wastani mainly yielded remains from the 4th– 7th c. CE. Griesheimer found 112 tombs, including pit-graves with sarcophagus lids, hypogea, sarcophagi, a mausoleum, and unspecified tomb types (cat. 2). At least four of the hypogea were decorated with sculpted busts above the entrance. Little description of the tombs is available, but they may have dated to the Byzantine period. Griesheimer also reports a distyle tomb (perhaps 2nd c. CE) and a hypogeum from the 1st or 2nd c. CE (cat. 1). A cliff wall with reliefs of stylized busts in panels associated with four hypogea of unclear date was found at the same location as the hypogeum (cat. 2). PALMYRA112

The site of Palmyra (modern Tadm¯ur) arose around an oasis of underwater springs in the Central Syrian Desert. A wadi divides the area into a northern and a southern part, and the modern town extends northeast of the northern section. Evidence for the period before the 1st c. BCE is limited (see p. 211). Better evidence for building activity and growth into an urban center stems from the 1st c. CE. Palmyra earned the status of civitas libera after Emperor 109

IGLS II, 243; Burns 1999 [1992], 226; Butler 1920, 260; Griesheimer 1997a, 185; Tchalenko 1953, 194–199; Prentice 1908b, 126–127. 110 IGLS II, 280; Tate 1992, 292; Tchalenko 1953, 125–127. 111 Griesheimer 1997a, 170, 175–176, 184–185, 198–199; Peña et al. 1999, 154–164. 112 Bounni 2005; Butcher 2003; Colledge 1976; Dijkstra 1995; Millar 1993; Schlumberger 1951; Schmidt-Colinet & al-As’ad 2000, 2002, 2003; Schmidt-Colinet et al. 2008, 459–461; Tabaczek 2004.

PALMYRA

Hadrian’s visit in 129 CE, and the title of colonia in the early 3rd c. CE. Its territory extended west toward Homs (Emesa) and north to the Euphrates.113 The bulk of the construction occurred in the 2nd and perhaps the early 3rd c. CE, during which time the town expanded and earlier buildings were refashioned. Building activity mostly ceased during Palmyra’s short war with Rome. The origin of this rebellion should be sought in the earlier 3rd c. CE, when the Romans relied on Palmyra’s military against the rising power of the Sasanians on the eastern border. It was in this context that the fabled Queen Zenobia emerged and turned against the Romans in the 260s. The face-off lasted until 272, when Emperor Aurelian crushed the Palmyrenes and took the city. A few decades later, under Diocletian (ca. 300 CE), a military camp was added to the northwest corner of the city and a city wall was built encircling the settlement; Justinian rebuilt this in the 6th c. CE. The city wall cut off the southern and perhaps oldest section of the city, which may have been abandoned by that time.114 It also incorporated many (abandoned) tombs, as described in the next section, and the military camp covered the Allat Temple. After the conflict with Rome, the use of Palmyrene for inscriptions ceased and the only building activity attested came in the form of churches (6th and perhaps 5th c. CE). The site remained inhabited until at least the Umayyad period.

Pre-Roman Tombs Two pre-Roman tombs were discovered at Palmyra (Figure 5).One was located in the northwest corner of the temenos of the later Baalshamin Temple.115 This communal mudbrick tomb consisted of a central corridor with stacked loculi on both sides and pit-graves in the floor.116 Its construction dates between 175 and 150 BCE. An inscription added in 11 CE was interpreted by the excavator as an epitaph, but by others as the commemoration of the purification of the tomb as it became part of the temple ground.117 The latter interpretation is probably accurate, since the text itself differed from contemporary funerary 113

A boundary marker denoting Palmyrene territory and referring to the time of Germanicus was found 75 km northwest of the city; another, found 60 km west of the city (Qasr al Heir al Gharbi), established the boundary between Palmyrene and Emesene territory in 129/130 CE. 114 115 Schmidt-Colinet & al-As’ad 2002, 161. Fellmann 1970. 116 The type was probably a hypogeum, although it is not certain whether the structure was entirely belowground. 117 Fellmann 1970, 112–114; Kaizer 2010, 24; Teixidor 1979, 22–23. The Palmyrene text reads: “in the month Iyyar, the year 322 [May 11 CE], this tomb was opened and cleaned up: Wahballat son of Mattai son of Gaddarsu son of Mattanai son of Qainu son of ’Atai son of Yedi’ebel opened it and cleaned it up. May he be remembered forever. He and his children in wellbeing!” Underneath, in a different hand: “And remembered be ’Ogeilu son of Maliku son of Hairan. Remembered forever be [ … ] son of Yedi’ebel the Elder, his grandfather. Peace!” (trans. Teixidor 1979, 22–23). According to Fellmann (1970, 115–116), a second inscription on the same loculus has a date of 57/58 CE, but this is doubted by others (Gawlikowski 1974, 238).

285

286

APPENDIX 1

inscriptions.Based on stratigraphic,epigraphic,and material evidence,the tomb was likely abandoned between 50 BCE and 11 CE. It yielded fragments of wooden coffins and 140 grave goods; eleven on average per grave. The finds are listed in Table 4. The most frequent consisted of vessels, items of personal adornment, and lamps. Several alabaster vessels stemmed from the tomb. Multiple interments illustrate that the graves were frequently reused, and a total of sixty-one individuals could be identified. A second Hellenistic tomb was discovered more than 2 km southeast of the Baalshamin Tomb in the area of the later Southeast Cemetery. This pit-grave (Tomb G) held a middle-aged man in a wooden sarcophagus, adorned with a golden ring, two bracelets (gold and bronze), an engraved ring, and beads and pendants of necklaces (gold and agate). Pieces of textile and leather were found by the body (Table 4). Based on C-14 dating of the skeleton, the burial took place between 380 and 160 BCE.118 Tomb G and the Baalshamin Tomb were not only a great distance from each other but also far from the areas with non-funerary contemporary remains: the Baalshamin Tomb was 700 m north of Trench I and the same distance northwest of the Bel Temple. Tomb G lay 1.3 km m southeast of the Bel Temple and 2 km to the northwest of Trench I.

Roman Cemeteries119 Necropoleis surrounded Palmyra on all sides and flanked the main roads to and from the city (Figure 60). At least 476 tombs have been discovered, most of which remain unexcavated and/or unpublished. This study incorporated 115 tombs into the primary database (Online Appendix Palmyra 1); the remainder are listed as cat. 2 (Online Appendix Palmyra 2). The earliest tombs dated to the second half of the 1st c. CE, and the majority were constructed between 50 and 150 CE. Foundation inscriptions indicated that new tombs were constructed up to the early 3rd c. CE, and according to the find-assemblages and inscriptions, the tombs received new burials up to the Aurelian invasion in the 270s. Only one tomb, the Hypogeum of Malikho, had a possible postAurelian invasion inscription (274 or 279 CE), and there is little concrete evidence for the use of the other tombs after 272 CE.120 A series of stelae and 118

Saito 2005a, 34; 2005b, 16–19, 159–162. IGLS XVIII/1; Abdul-Hak 1952; Albertson 1983; Amy & Seyrig 1936; Browning 1979; Cantineau 1932; Colledge 1976; Daszewski 1972; Gawlikowski 1970, 1995; Henning 2003, 2013; Higuchi & Izumi 1994; Higuchi & Saito 2001; Ingholt 1935, 1938, 1966; Machowski 1983; Michalowski 1960, 1961, 1963; Parlasca 1982; Pfister 1937, 1940; Rahmo 1993; Sadurska 1976; Sadurska & Bounni 1994; Saito 1995; Saliby 1992; Schnädelbach 2010; Schmidt-Colinet 1992; Toynbee 1971; Will 1949a; Witecka 1994. 120 Gawlikowski mentions that one hypogeum constructed in the West Cemetery reused a lintel from a tower-tomb and perhaps post-dated the Aurelian invasion. Also, the epitaph of 85b appears to be Christian according to Gawlikowski (1970, 160). 119

PALMYRA

60. Plan of Palmyra

decontextualized inscriptions with reference to the Christian Lord date to the 5th and 6th c. CE.121 Several chronological patterns can be detected, starting with the fact that in the 1st c. BCE the area of the Baalshamin Tomb was no longer used for burial and a new cemetery was created west of the town. This West Cemetery extended perhaps as much as 2 km from the settlement and the Baalshamin Tomb (Figure 5). Over time, the area between the westernmost tombs and the town filled in.In the mid–late 1st c.CE,four additional cemeteries were created north, east, and south of town. Construction activities were fairly constant in the first half of the 2nd c. CE, and coincided with urban development until the middle of the century, after which a sharp decline in tomb building occurred. 121

They were found in the modern town, in the West and North Cemeteries. IGLS XVII, 451–460, 494–511.

287

288

APPENDIX 1

The tombs remained in use, and were frequently enlarged and decorated with sculpture and inscriptions until the 270s. With the construction of Diocletian’s camp and the northern section of the city wall, at least thirteen tombs now stood within the city, and another twelve to sixteen were incorporated in the construction of the wall, likely in Diocletian’s time. None were in use at the time, and it remains unknown where the population of Palmyra was buried after the reconquest by Rome in the 270s. Occasional reuse of older tombs may have occurred, but the population owning the tombs had fled, died, or was too impoverished to finance a (visible) burial. Unfinished loculi in numerous hypogea demonstrate that construction was not finished when the tombs were abandoned. Careful excavation of Tomb C and Tomb F has revealed that perhaps as much as 30% of the loculi and pit-graves were never used. Scholars have divided the burial grounds into four separate entities: the West, North, Southwest, and Southeast Cemeteries. The first two, however, may have been part of a single burial ground. Henning mentions in a footnote the discovery of another cemetery, marked “East Cemetery” on her maps.122 Schnädelbach’s map has four structures in this area (s201–204), three of which may have been tower-tombs. Often overlooked is the collection of pit-graves that was discovered in the 1930s, mostly northeast of the settlement. This may have been a separate burial ground, or it may represent the eastern edge of the North Cemetery. Each cemetery is described in detail in this section.There first follows a summary of the general patterns. The cemeteries covered several square kilometers, and the oldest tombs, in the West Cemetery, stood farthest from the Roman settlement (ca. 1.5–2.0 km), a characteristic shared with the pre-Roman tombs. Starting in the 1st c. CE, the cemeteries grew toward the city or became more crowded. The oldest city or enclosure wall, as reconstructed by Gawlikowski, may have incorporated some tombs in the West and Southeast Cemetery, as discussed later, whereas the second wall system in the north enclosed the entire North Cemetery.

Architecture The oldest tombs were the tower-shaped mausolea, or tower-tombs (Figure 12a,b).123 Tall square stone structures with a stepped lower part and multiple stories, 124 their construction likely started in the second half of the 122

Henning 2013, 11 n. 71. Tower-tombs have been extensively catalogued by Henning (2013) and the reader is referred to her publication for a full description. The cat. 1 sample includes nineteen tower-tombs, which were selected to give a good impression of the different types and chronological development. They are also the ones that yielded data about grave goods or human remains. 124 The square base measured on average 10.64 m; the total height is difficult to reconstruct because of limited preservation. Some towers reached well over 20 m, and the tallest included seven or eight stories. 123

PALMYRA

1st c. BCE, and the last recorded construction of a tower-tomb took place in 120 CE. Finds and inscriptions illustrate that the tower-tombs remained in use at least until the (mid) 2nd c. CE. Burial initially took place in exterior loculi placed in the podium. Over time, burial spots moved to the interior, to tiers with stacks of loculi in the walls. By the mid 1st c. CE, burial also occurred in decorated limestone sarcophagi. The total number of burial spots per tomb also increased. Exterior decoration was limited to the area around and above the entrance and consisted of inscriptions, moldings, and group scenes in low relief. Whereas the earliest tower-tombs were plastered inside, those of the late 1st and 2nd c. CE were decorated with reliefs, pilasters, moldings, and coffered ceilings. Sometimes, painted Victories, garlands, and Bacchic scenes were added. Entrance doors were often decorated with cornice or coffers. Henning has demonstrated that over time, the towers became more regular, were built with new techniques, and became more extensively decorated. Starting in the mid 1st c. CE, they became monumental and more standardized.125 The loculi of early tower-tombs were closed with break stones and mortar, on which the name of the deceased was written. Loculus slabs with portrait busts closed some of the later examples. Characteristic of the West Cemetery were the tower-tombs standing on top of a hypogeum. They were built in the 1st c. CE and used until the late 2nd c. CE. Burial took place both in the hypogeum and in the tower portion of the tomb. The towers were slightly smaller than those of the regular tower-tombs.126 The most common group in the assemblage consisted of hypogea without towers (Chart 27, Figures 9d, 28). The oldest were constructed in the second half of the 1st c. CE, and the type was particularly popular in the first half of the 2nd c. CE. The latest dated construction is 186 CE, and many were in use until the mid 3rd c. CE or slightly later. There is some chronological development in layout, and there was more variation in shape in the 2nd c. CE, but all hypogea followed the same set-up, consisting of an elongated rectangle or T-shape with several connected rooms.127 The hypogea contained many burial spots; the largest had 390 places.128 Burial occurred in stacked loculi, sarcophagi on the tomb floor, and pit-graves in the floor. The latter method is less common, and seems to have been reserved for children or perhaps infants. A large and decorated stone door closed the hypogea. The interior could be plastered, painted, and decorated with moldings and pilasters (Figure 20a). The majority of the Palmyrene funerary busts, decorated sarcophagi, and sculpted 125

Henning 2013, 24, 66. Dimensions: square base on average 9.26 m; between two and five stories. 127 The Tomb of Nasrallat (139 CE) followed a slightly different plan, with four rectangles of equal size. 128 Average dimensions of central room/corridor:17 × 3 m (longest is 21 m).Dromoi or entrance corridors varied widely in length due to limited preservation; the longest was 12 m (width between 2 and 3 m).

126

289

290

APPENDIX 1

160 140 120 100 80 60 cat. 1 40

cat. 2

20 0

chart 27. Distribution of tomb types, Palmyra

group scenes stemmed from hypogea, possibly connected to the fact that the contents of these tombs were less exposed than aboveground mausolea and tower-tombs. The donor of the tomb was often buried at a central location, surrounded by sculptural elements. Many of the hypogea remained unfinished. Probably dating to about the same time as the hypogea, an entirely different cemetery was planned northeast of town: a grave field with pit-graves marked by a stele (Figure 16a). Only the stelae and inscriptions (54) were published, and they may not all have originated from this location. Little is known about the find context of the pit-graves, save for Cantineau’s description: a cemetery of simple graves with a single body in a plaster coffin with some glass vases and jewelry. Close to each pit was a stone stele, of which the pointed lower part was placed vertically in the ground.129 Gawlikowski notes that the coffins were the so-called “slipper-sarcophagi,”130 which, if correct, would be the westernmost find location of a coffin common in Parthian Mesopotamia. The stelae dated between ca. 50 and 150 CE; one carried an inscription of 140 CE (see more description later). Three subgroups can be distinguished: the largest had an inscription and simple molding decoration (twenty-eight in total: eighteen for men, eight for women). The second group depicted a curtain hanging between palms or suspended in midair above the inscription (twenty-two in total:twenty for men, one for a woman). The final group consisted of three stelae portraying 129

Cantineau 1932.

130

Gawlikowski 1970, 34.

PALMYRA

a standing woman (one in front of a curtain) and one with a man standing behind a curtain.131 The Palmyrene inscriptions consisted of a double name and an expression of mourning (“alas,” “lamented”). As such, they are similar to the inscriptions on the individual closing slabs in the communal tombs. A small number of similar stelae were also found in hypogeum tombs in other cemeteries, dating to the 1st c. CE (see later).132 A collection of ten stelae in Latin and one in Greek, belonging to soldiers, was published. One was found in the modern town; the location of the rest is not known (cat. 2). The final tomb type was the stone rectangular mausoleum erected on a podium, often called the temple- or house-tomb (Figure 10a–c). The earliest was built in the mid 2nd c. CE and the latest in 236 CE. Use of house-tombs continued to the Aurelian invasion. Based on the reconstruction of the exterior façade, the mausolea fall into four subgroups. A flat roof and corner pilasters (Corinthian or plain) characterized the first. The second had a similar roof and included a porch carried by multiple columns, as well as a staircase leading to the entrance. The third group combined the porch and staircase with a pediment-shaped roof. The final group resembled temple architecture most closely, although the roof was flat. This last group is represented by Tomb 36, the elaborate façade of which was subdivided into several stories with aediculae, columns, and relief decoration, making it reminiscent of theater architecture and Petra tombs.133 Less is known about the interior of the mausolea. Some included a peristyle surrounded by tiers with loculi, whereas others had a single room with sarcophagi. Doors in the back presumably led to rooms, underground or in the podium, and perhaps held more burial spots. Because only one ground floor of the mausolea of this type was usually preserved, little can be said about the total number of burial spots, except that they could hold dozens or more people. Decoration of the mausolea consisted of intricate vertical and horizontal scrolls on the exterior walls and pilasters. Tomb 36 had shell-shaped niches and depictions of sea-creatures and Dionysus-Baalshamin on the exterior façade, whereas reliefs of a Victory, imago clipeata, an eagle, a theatrical mask, gorgon heads, shells, and sea-creatures were found inside. Some fragments of funerary busts stemmed from the mausolea. Schmidt-Colinet has drawn attention to inspiration from both Roman architecture and Parthian palatial elements for the Palmyrene mausolea.134 131

More of the same type are depicted in Gawlikowski (1970, 34, extreme left), Parlasca (1976, plate 3), Schlumberger (1960, plate X.3), and perhaps Morehart (1956–1957, fig. 23). It is not certain whether these are from the same cemetery. 132 See, for example, a stele with a veil in the Hypogeum of ’Aˆstôr (probably pre-foundation of the tomb; Sadurska & Bounni 1994, 14–22, fig. 1), a standing man in front of curtain (same tomb, fig. 5), a standing man (Hypogeum of Artaban, perhaps a little after 50 CE; Sadurska 1976, 23–40, fig. 3), and a standing woman (Hypogeum of Taim’amed, foundation 94 CE; Sadurska 1976, 113–114, fig. 4). 133 The average dimensions of the podium are 14.6 × 14.5 m. 134 Schmidt-Colinet 1997, 164–166.

291

292

APPENDIX 1

table 7. Distribution of grave goods, Palmyra earring bracelet ring pin amulet necklace pendant bell other jewelry shoe (single) textile button glass bottle glass small jar glass small vessel glass cup glass jug glass unknown shape pottery bottle pottery small vessel pottery plate pottery cup pottery jug pottery pitcher

35 6 25 13 6 32 22 9 57 6 16 2 20 1 4 4 1 58 8 1 2 4 2 3

pottery bowl pottery jar pottery amphora pottery cooking pot pottery incense burner pottery water pot pottery water basin pottery unknown shape lead vessel wicker vessel incense burner pottery incense burner unknown material vessel unknown material bronze coin coin unknown material lamp bronze needle grinding stone terracotta figurine bronze/copper mirror bone comb flute nail unknown material/type

2 5 2 5 5 5 3 8 1 2 7 4 6 7 5 640 1 1 1 1 2 2 9 65

Grave Goods The total amount of grave goods was 1129 from twenty-one tombs, mostly hypogea (Table 7). Oil lamps formed the majority, and many burial spots contained just a single lamp placed in or close to the grave. Compared to other sites in Syria,the Palmyrene tombs included smaller amounts of personal adornment and vessels. Coins were rare, particularly in later tombs (2nd–3rd c. CE). The tower-tombs, including those combined with hypogea, had small numbers of vessels when compared to other sites in Syria. About 30% of the tombs contained vessels. Vessels became more frequent in later tombs, and glass arrived perhaps only in the 2nd c. CE. Of the later hypogea, 75% included glass vessels, but the total numbers remained small, below the numbers of pottery vessels. Most remains of clothing originated from the tower-tombs and towertombs/hypogea. The pit-graves in the Northeast Cemetery may have contained glass vessels and simple jewelry. Two 2nd c. CE hypogea excavated in the 1990s (Tomb C and Tomb F) were relatively undisturbed; these offer a glimpse into the location of items within the tomb (Figures 26, 36, 38). Many individuals in these tombs were buried without (archaeologically traceable) grave goods, and much variation existed in the total amounts per burial spot within the tombs. Tomb F contained 255

PALMYRA

artifacts, with 145 finds coming from thirty-nine loculi (average of 1.8 per loculus and 2.0 per buried individual). The finds per loculus usually consisted of a small number of items of jewelry (often rings and necklaces), fragments of clothing, and a lamp. One glass bottle stood in a loculus; the remains of at least eight others were found dispersed through the tomb and no longer in situ. The pottery was often placed in the central areas of the tomb rather than in a loculus, and consisted of water basins, cooking pots, an incense burner, a plate, and a bowl. Lamps, on the other hand, were placed inside the loculus (fifty-three in twenty-four loculi), on top of the closing slabs, and in the central areas of the tombs (58). The distribution in Tomb C was similar, with a small number of jewelry and lamps inside the loculi.135 The glass vessels were found inside the loculi. The pottery items, such as incense burners, lamps, a water pot, and a bowl, were placed in front of the loculi or in the central area of the tomb.

Sculpture Palmyrene tombs have yielded an extensive collection of sculpture, unfortunately mostly decontextualized (Figures 30a,d, 35). The tombs in cat. 1 have yielded a minimum of 335 sculptural works, in addition to the stelae of the Northeast Cemetery. This collection consists of portrait busts on the closing slabs of the loculi (at least 234), closing slabs with standing figures (at least twenty-three), relief slabs and sarcophagi with reclining group scenes (at least fifty-five), one full statue of a woman, two statuettes of a man and a woman, and nineteen other fragments of figural sculpture. A handful of rectangular stelae come from the tombs, although their original placement remains uncertain. One stele did not depict a figure, just a curtain. The portrait bust and half-figure were the most common types.136 The earliest examples dated to 65 CE; they were found in all communal tomb types. The funerary busts most often depicted men, wearing a long dress with the right arm in a sling and often carrying an item in the other hand. The men were sometimes depicted as priests wearing a priestly cap or modius. Portrait busts of women wore long dresses with a veil over the head and an elaborate headdress, with necklaces, earrings, and bracelets. Heyn has demonstrated that Palmyrene men, women, and children often held gender- and age-specific items in the hand.137 Sometimes, a curtain was suspended behind the busts. Single busts were most common, but multiple ones also existed; these depicted 135

Tomb C contained 110 artifacts in total, of which seventy-two came from twenty loculi. See also Kropp & Raja 2014. 137 Heyn 2010. Woman carried a spindle and distaff, children held birds or grapes, and men grasped scrolls, vessels, and palm branches. 136

293

294

APPENDIX 1

parents with children, brothers, and sometimes cousins, and the additional figures were added in a bust of equal size or as a smaller full figure. Some slabs or stelae depicted a standing individual in long robes, sometimes in front of a curtain, holding items. These most often depicted male children, but occasionally female children and adults were shown as well. In a few cases, the closing slabs were similar to the stelae in the Northeast Cemetery, depicting a curtain with or without a standing adult figure. These seem to have been placed at the same spot as the portraits, closing a loculus, although this is not certain in each case. Sadurska mentions such a stele in the Hypogeum of ’Aštôr, which appeared to pre-date the construction of the tomb, and thus must have originated from elsewhere. The slabs in the communal Palmyrene tombs depicted 139 men, eight-nine women, nine mixed groups, nine male children, and three female children. A second type consisted of forty-four larger flat slabs portraying a figure reclining on a couch or kline, with other individuals to the left and behind them. This type is usually referred to as a “banquet scene,” although “reclining” or “presentation” scene would more accurately describe the depiction. The composition varied, but usually included one or two male figures reclining on a couch with cushions and holding a bowl or cup in the left hand. A woman was seated to the left, and smaller figures or full-sized figures stood behind the reclining male(s). Between the legs of the kline, busts were often depicted. The reclining men wore long robes or so-called “Parthian dress,” consisting of a long-sleeved tunic and trousers. The scenes depicted 103 men, fifty-nine women, nine male children, and three female children, but information about the individuals depicted on many of the slabs is missing. Their earliest occurrence was on the façade of the tower-tomb of Kithot (40 CE; Figure 12b), and they appeared inside the communal tombs in the 2nd c. CE, where they stood in front of or over a burial spot. Inscriptions on the slabs in Palmyrene identified the deceased. The seated female could be the wife or mother of the reclining male; other figures depicted were children and brothers. A related group is formed by stone sarcophagi covered in similar reclining reliefs and other features of Palmyrene life, such as camels. These were introduced in the mid 2nd c. CE. Their description is often unclear. It is not certain, for instance, whether each feature identified in the report as “sarcophagus” was a freestanding stone coffin, a flat relief slab, or a series of such slabs covering loculi. Non-sculptural depictions of persons come in the form of painted portraits in medallions carried by Victories from two hypogea dating to the 2nd–3rd c. CE. Little is known about these, but they presumably depicted the deceased. The coffered ceiling of the tower-tomb of Iamlikh¯o (83 CE or later)

PALMYRA

also contained portrait busts. Many of the figural depictions of Palmyra were published without information about their find context, and the images were not always published in conjunction with the epitaph. The figural sculpture of Palmyra is rich and is in dire need of a full genealogical–spatial analysis, which is beyond the scope of this book.

Human Remains The Palmyrene tombs yielded the skeletal remains of 528 individuals, found in eleven communal tombs. The pit-graves in the Northeast Cemetery reportedly held a single individual, but it is not certain whether this was actually investigated or just assumed. The inscriptions on the stelae of this cemetery usually recorded a single individual, with one or two exceptions, as already described. The mode of burial in the Palmyrene graves displayed greater variation than in the rest of Syria. Three cremation graves were discovered in loculi of Tomb F, the Tomb of Yarhai, and Tomb C. Mummies wrapped in linen, wool, cotton, and silk were found in tower-tombs and in tower-tombs with hypogea. According to early travelers to Palmyra, these mummies were interred in wooden coffins. It is not clear whether this practice was restricted to the towers, but if it were, this might suggest that mummification was common in the earliest decades of the Palmyrene cemeteries (until the mid 2nd c. CE or earlier). At least forty mummies of mixed sex and age were reported from five tombs. They were sometimes covered with exquisite textiles. One bronze pin was also found on a mummy, and a child mummy wore sandals. Recent study has indicated that the bodies of the dead were first dried and subsequently wrapped in cloth and covered with several layers of myrrh paste.138 The sexed skeletal remains represent eighty-five male and sixty-seven female individuals. Skeletons were also aged, but the methods are not always recorded. The sample included 276 adults (of which 137 are certain) and 122 children (of which 95 are certain). All tomb types contained both males and females and both adults and children, who shared not only tombs but also burial spots. About 25% of the loculi contained multiple individuals (between two and six). Children more often accompanied males than females. The back wall of the main room in the hypogeum, which was the most central location and often the most elaborately decorated, contained more shared loculi than other parts of the tomb. Pit-graves in the floors of the hypogeum were often reserved for young children. No strong patterns emerged in the study of the distribution of grave goods, except that single burials – of both sexes and age groups – held a higher numbers of finds than co-burials. 138

Schmidt-Colinet et al. 2000, 56. See also Pfister 1934, 1937, 1940.

295

296

APPENDIX 1

Epitaphs Palmyra has yielded an extensive corpus of funerary inscriptions spanning the entire Roman history of the site (Figures 32, 35). Most were in Palmyrene Aramaic (perhaps around 1300); smaller groups were made in Palmyrene and Greek (ca. 80), Greek (ca. 44), or Latin (ca. 15). The Semitic epitaphs are published separately from the Greek and Latin examples, and at this stage no comprehensive study exists of the funerary epitaphs as a corpus.139 Numerous graffiti in the hypogea are poorly published and not included in the database. They consist of names of the deceased,rosettes,and symbols such as animals and humans.140 The tombs in the database held a minimum of 298 Palmyrene epitaphs, twenty-four in Palmyrene and Greek and two in Greek; the rest are listed as cat. 2 (Tables 5, 6).141 The earliest dated epitaphs were two identical inscriptions on the Tomb of ’Atenatan, carved in 9 BCE. There is no evidence of epigraphy before the Roman period. The two inscriptions or graffiti in the Hellenistic Baalshamin Tomb were added in the Roman period (1st c. CE). Epitaphs were popular in the 1st c. CE, and particularly from the mid 1st to the 2nd c. CE. Their number decreased in the 3rd c. CE. The latest date on an inscription was 274 CE. The number of bilingual inscriptions was always low, and remained steady over time, whereas the number of Palmyrene inscriptions grew exponentially in the 2nd and 3rd c. CE. All tomb types included inscriptions, and the recovery rate was particularly high for stelae and hypogea. The epitaphs included three types: 1) foundation inscriptions (36 in database), 2) transfer inscriptions (42–43 in database), 3) individual inscriptions (247–249 in database). Four inscriptions could not be connected to these categories. The stelae of the pit-graves in the Northeast Cemetery only carried individual inscriptions; the other types included texts of the three types. Foundation texts were the oldest, and mentioned the name of founder(s), who the tomb was meant for, and a date. The language was Palmyrene (56%) or bilingual.142 These texts were usually placed above the doors giving access to communal tombs. The founders were always male, and dedicated to tomb to their descendants, sometimes also specified as male. The dedicator could be a single person or brothers. Sometimes, the tomb was also made in honor of the 139

Greek inscriptions are published in IGLS XVII, 1; Palmyrene inscriptions are published in PAT. 140 Henning 2013, 82–83. 141 For reasons of space, Online Appendix Palmyra 1 provides translations for only 157 inscriptions, which include all the foundation and cession texts of the cat. 1 tombs. 142 One trilingual inscription was found (Latin, Greek, and Palmyrene) in the West Cemetery: “Gaius Virius Alcimus and Titus Statilius Hermes have constructed this funerary monument and hypogeum for themselves and their children, for the honor, for always, the year 368 [56/57 CE]” (Gawlikowski 1998).

PALMYRA

297

father, and in one case it was built for brothers. The second group, the cession texts, and were mostly in Palmyrene (88%). These texts specified which parts of the tomb were sold or leased to others, and included names, a description of the portion ceded, and a date. They were usually located close to the foundation inscriptions, around the entrance door, but could also occur inside. The majority included one or more male owners ceding to one or more male parties (thirty-one inscriptions). Four inscriptions recorded the cession of tomb portions to female parties, and another four female owners sold parts of the tomb to men (3) or to a man and a woman (1). On occasion, freedmen and women are recorded as both the ceding and the receiving parties. The oldest example in the database dates to 147 CE, although Cussini mentions cession texts from the 1st c. CE.143 Around a third of the inscriptions dated to the second half of the 2nd c. CE; the remainder were added between 213 and 274 CE. This means that when monumental tomb building witnessed a sharp decline after the mid 2nd c. CE, the cession inscriptions started to occur. Related to the cession texts was a group of graffiti, painted and scratched inside the tomb. The Hypogeum of ’Abd’astor, for instance, yielded twelve, which according to Ingholt were the names of the owners of the stacks of loculi.144 Similar evidence comes from the Hypogeum of Lishams, where eight tiers of loculi on the left wall and two on the right were marked by the name of Malku son of Shalman. Tower-tomb 15 had graffiti with numbers (“34” and “16”) close to the entrance. Individual inscriptions formed the largest group, and were primarily written in Palmyrene. Only eight (3%) were bilingual. Placed on a stele on top of a pitgrave or inside the tomb on sculpture, coffins, or stone closing slabs, they named the individual deceased and added an expression of sorrow. Few carried a date. The stelae on pit-graves mentioned double name, whereas the other individual inscriptions included two or three names. By contrast, the foundation inscriptions specified at least three names. The texts commemorated a single person, but were sometimes grouped on a single slab or stelae with the names of others (see discussion of sculpture earlier). Individual inscriptions appeared with the introduction of portrait busts in 65 CE, but these may have followed an earlier tradition. In the tower-tombs, names were initially scratched in the plaster or mortar closing the loculi. Palmyrene epitaphs referred to families and possibly to members of larger descent groups, at least initially. The foundation texts presented the male dedicator and his descendants. The dedicator could be alone or could co-sponsor the construction of the tomb with his brothers. Sometimes, the tomb was also made in honor of the father, and in one case of brothers. Many tombs at this site encompassed several generations, and their use spanned decades 143

Cussini 1995, 235.

144

Ingholt 1938.

298

APPENDIX 1

and sometimes centuries. The average number of burial spots in the larger examples was 131, and the largest hypogeum had space for at least 390 people. Texts highlight the burial of brothers, half-brothers, and cousins in these tombs, illustrating that the tombs were destined for larger descent groups than just the immediate family. Abundant evidence exists for tribal structures in Palmyrene society, and the tombs may have housed such clans. The cession inscriptions varied in content, and recorded men and women ceding parts of the tomb to relatives or to people without any obvious familial relation. By the time of the sale of the parts of the tomb, therefore, generally between the mid 2nd and later 3rd c. CE, non-family members appear to have occupied portions of the funerary building. Interior inscriptions commemorated one individual without specification of relation to the rest of the tomb-dwellers, other than what can be gleaned from their double or triple names. Sometimes, additional information was provided, such as the names of the husband, brother, or son. When a single inscription commemorated multiple individuals, they usually concerned parents and children, but occasionally also adult brothers, husbands and wives, male cousins, and young male siblings. The group scenes on the relief-slabs and sarcophagi depicted spouses, parents, children, and adult male siblings. The epitaphs indicate that in Palmyra, extended families were buried along patrilineal descent: parents, unmarried daughters, brothers, and their wives and children. Adult sisters and aunts, when married, would be in the ancestral tomb of the spouse. Slaves and freedmen also featured in the Palmyrene texts. In one or two instances, the deceased was identified as a slave on an interior inscription, and several freedmen were depicted. Seven cession inscriptions dating between 160 and 239 CE referred to the sale of part of the tomb by and to freedmen and freedwomen. A collection of ten stelae in Latin and one in Greek, belonging to soldiers and their families, was published (cat. 2; Figure 35). Reliefs, often in Palmyrene style, accompanied them. Other than soldiers, the group included one veteran and two women, one of whom was a wet-nurse.

West Cemetery The West Cemetery or Valley of the Tombs yielded 141 tombs, of which twenty-six are in the cat. 1 database. This cemetery likely continued as the North Cemetery, but is discussed separately here. This burial ground contained the oldest tombs (from the second half of the 1st c. BCE) and remained in use until at least the 270s CE. Construction was fairly constant until the mid 2nd c. CE, after which it slowed down. This cemetery was also the largest in Palmyra, and yielded the greatest variation in architectural types. Extending

PALMYRA

between two hills along the road to Homs, the distance between the westernmost and easternmost tombs was 1.38 km, and at its widest point it reached 1 km (Figure 60). Four tower-tombs (206–209) stood further west, slightly isolated from the burial ground. If these tombs were part of the same cemetery, its total length would have reached 1.6 km. The relationship between Tombs 206–209 and the settlement also remains uncertain. They stood within the area enclosed by the 1st c. CE wall, but possibly post-dated its abandonment. Gawlikowski dated two of the tombs to the second half of the 1st c. CE, by which time the wall may no longer have been functioning, although Henning proposed a date of before or around the mid 1st c. CE.145 The tombs stood in the area of the so-called “Hellenistic town” investigated by Schmidt-Colinet and al-’Asad, where at least one building (Trench II) yielded layers from the 1st c. BCE to the 3rd c. CE. This building was ca. 100–200 m from the tombs. The oldest tombs of the West Cemetery stood farthest from town (50–1 BCE) and several hundred meters from one another. Most were built isolated on hilltops. The distribution in the 1st c. CE was more irregular, with some tombs grouped on Jebel Umm Belkis and others more dispersed through the wadi and slopes. Two clusters in the central part of the West Cemetery dated to the 2nd c. CE.146 Over time, the density of tombs increased toward the city and toward the road. With the addition of the army camp and new city wall around 300 CE, two tombs were inside the urban space, and at least four tower-tombs were incorporated into the wall construction. The West Cemetery yielded seventy-five tower-tombs, twenty tower-tombs with hypogeum, twenty-four hypogea, sixteen mausolea, and four tombs of unknown shape. The tower-tombs were the oldest, followed by tower-tombs with hypogeum, hypogea, and mausolea. When new types were constructed, however, the older shapes were still in use, and remained so for decades and sometimes centuries. The tower-tombs were built between 50/1 BCE and the early 2nd c. CE, and followed three basic forms. The oldest (50–1 BCE), fourteen to sixteen in total, stood on the western edge of the cemetery and on Jebel Umm Belqis. The loculi of this type were accessible from the exterior façade. The second group (41/43), dating from the late 1st c. BCE to 103 CE, had the loculi on the interior of the tomb. The third type (20) stood on top of a hypogeum. These were built in the 1st c. CE and concentrated in small groups (A, B, 44, 45) (larger group on Umm Belqis). The hypogea without towers concentrated in the west–central section and along the northern slopes of the 145 146

Gawlikowski 1974, 242. Henning 2013, 289–292. In the first centuries, most tombs faced the wadi, presumably facing the western road to the city. The cluster on Umm Belkis faced the city. Two groups of tombs in the center of the valley, some dated to the 2nd c. CE, have their entrances on the opposite side, facing away from the road.

299

300

APPENDIX 1

West Cemetery. Their construction started in the 1st half of the 2nd c. CE. Mausolea were built between 149 and 210 CE, close to the city and the central section of the valley.

North Cemetery The North Cemetery contained a minimum of 150 tombs, of which twelve are in the cat. 1 database. The dated inscriptions ranged from 79 CE to 236 CE, but few tombs in this cemetery were securely dated. Some of the tower-tombs dated to the first half of the 1st c. CE. The cemetery covered a surface of ca. 1.25 × 0.64 km, and the tombs originally lay dispersed north of the city. Most tomb doors opened toward the settlement. Scholars draw the border between the North and West Cemetery on the hill on which the later camp of Diocletian was built. Two tombs in this area (the Hypogeum of ’Alainê and the Anonymous Tomb/Hermes/J), however, cannot be clearly associated with only one of the cemeteries. It is likely that the West and North Cemeteries were part of a single burial ground, which was divided in two parts when the Camp of Diocletian was built. This construction coincided with the erection of a new city wall. At least eleven tombs were incorporated in the course of the wall, and nine others now lay within the walled-in area. The North Cemetery contained fifty-three tower-tombs, one tower-tomb with hypogeum, fifty-five hypogea, thirty-four mausolea, and seven tombs of unknown shape. Only two towers were dated (79 and 118 CE). No spatial clusters were apparent. The hypogea were mostly clustered on the southeast slope of the cemetery, and a handful were dispersed among other tomb types in the eastern part of the cemetery. None was securely dated. The mausolea rose in the 2nd and (early?) 3rd c. CE and are encountered in each section of the burial ground.

Northeast Cemetery Multiple tombs were found during the construction of the new town of Palmyra in 1929–30. Some were discovered in the garden of the current archaeological museum; the exact locations of the others remain unclear. The museum gardens could have marked the eastern edge of the North Cemetery, which lies ca. 200–350 m to the west. However, since the mode of burial differed profoundly from that of the other cemeteries, I treat the tombs here as part of a separate burial ground. This Northeast Cemetery consisted of fiftyfour pit-graves marked by stelae, which are dated in the literature between the mid 1st and 2nd c. CE; one included an inscription of 140 CE. As already mentioned, only the stelae are fully published; according to Cantineau, these stood

PALMYRA

on top of pit-graves containing plaster or terracotta sarcophagi (Figure 16b). Whereas this mode of burial was unique for this cemetery and period, the addition and text of the inscription and the use of individualized figural sculpture correspond to forms of burial elsewhere, in the communal tombs.

Southwest Cemetery The remaining two cemeteries of Palmyra were smaller in size and extended to the south of the oasis, at some distance from the town. The Southwest Cemetery started ca. 17 m southwest of the gate in the southern enclosure wall and stretched out along the road to Damascus. This location was 1.53 km from the Bel Temple and 1.39 km from the central city. It yielded eighty-five tombs, of which twelve are in the cat. 1 database. The earliest tombs dated to the first half of the 1st c. CE, and the majority were built between 50 and 125 CE. The tombs were used until at least 265 CE. The published portion of the cemetery surface covered ca. 300 × 260 m, and most tombs were aligned with the road. The assemblage consisted of twenty-six tower-tombs, four mausolea, and fiftyfive hypogea. This last group was found all over the cemetery surface, with a large cluster on the southeast slope.

Southeast Cemetery The Southeast Cemetery contained at least forty-seven tombs, of which eleven are in the cat. 1 database.147 The dated tombs were constructed between 1/50 and 128 CE, and finds illustrated that they were used until at least the early 3rd c. CE. The location was not new, as Hellenistic Tomb G came from this area, although a gap of 250 years or more separated the digging of Tomb G and the construction of the first Roman tomb. The Southeast Cemetery lay ca. 1.20 km from the Bel Temple and covered a surface of (at least) 500 × 417 m. It included sixteen tower-tombs,twelve hypogea (late 1st and 2nd c.CE),eleven mausolea (2nd c. CE), and eight unknown shapes. Maps reveal several clusters of tombs on hilltops or other elevated positions, consisting of tower-tombs, mausolea, or a combination of the two. A cluster of four possible mausolea stood on a slightly elevated location, and two other possible clusters had two tombs each. Almost all tower-tombs stood on an elevated position, whereas the hypogea crowded the lower parts of the cemetery. The later tombs lay closer to the settlement. 147

Sadurska & Bounni (1994) list three tombs from this necropolis that are not indicated on any map.

301

302

APPENDIX 1

SELENKAHIYE148

The main habitation phase on this prominent tell in the Syrian Upper Euphrates region dates between 2400 and 1900 BCE. Little is known about the site after the early 2nd millennium BCE. The region of Selenkahiye may have been part of the Parthian realm between the 2nd c. BCE and the mid 2nd c. CE (see p. 12). A terrace on top of the mound was formed by mudbrick retaining walls of the Parthian or Roman period (mid 2nd c. BCE–4th c. CE), and contemporary pits were filled with straw, oil lamp fragments, and bone net spacers. The excavators speculate that the Parthian–Roman settlement was located between the necropolis and the summer bed of the Euphrates. In total, fifty-three post-Bronze Age graves were published. Lack of dateable finds makes dating these graves difficult. Zaqzûq states that most tombs date to the middle of the 3rd c. CE, presumably based on coin evidence. Coins, however, rarely make secure indicators of the date of construction, as they could have been several decades or centuries old before being deposited in the tomb. Comparison with pottery from Dura Europos places the construction of some of the graves between the 1st and the mid 3rd c. CE. The second group of tombs was constructed in the late 3rd c. CE, based on pottery and comparative forms nearby in the Syrian Upper Euphrates zone.149 The burial ground at Selenkahiye may have continued or been repurposed in the Byzantine and Early Islamic period. The total number of cat. 1 tombs is forty-five (Online Chart Selenkahiye cat. 1), and a minimum of eight tombs were included in the cat. 2 collection (Online Chart Selenkahiye cat. 2). The cemetery stretched out on the central and northern parts of the mound. No map exists of the burial ground. The cat. 2 graves included a group of tumuli constructed in the 3rd c. CE or later, situated among the houses of the modern village, north of the bottom of the mound. Orientation varied, but the largest group of cat. 1 tombs was laid out on an east–west orientation. The majority of tombs were pit-graves (Chart 28, Figure 8d,e). Multiple large storage jars covered the tops of three pit-graves, one with a Greek graffito. One held a partly preserved limestone sarcophagus. The second group is formed by cist-graves lined by large and smooth limestone blocks, mudbricks, or baked bricks. Three jar-burials held small children (Figure 13d). The cat. 2 tombs add tumuli or pit-graves covered by a large circular earthen mound. The discovery of wooden coffins is also reported, which could have originated from pit- or cist-graves. Iron nails and bronze fragments in the vicinity of the graves may indicate the presence of additional wooden coffins.

148

Van Loon 2001; van Loon & Meijer 1983. Tombs: Bounni 1980, 318; van Loon & Meijer 1983; Zaqzûq 2001. 149 Bounni 1980.

SELENKAHIYE

303

40 35 30 25 20

cat. 1

15

cat. 2

10 5 0 pit-grave

cist-grave

jar-burial

tumulus

unknown shape

chart 28. Distribution of tomb types, Selenkahiye 7 6 5 4 3 2 1 0

chart 29. Distribution of grave goods, Selenkahiye

Approximately 30% of the tombs included a small number of grave goods (Chart 29).Compared to assemblages elsewhere in Syria,the Selenkahiye tombs contained few vessels and more spinning implements and coins, thirteen of which were deposited between the legs of a single individual. The glass bottles originated from later, possibly 3rd c. CE, graves. The pit-graves contained 80% of all finds, including all glass vessels, coins, and spinning tools, whereas the jarand cist-graves were sparsely outfitted with finds. The cat. 2 tombs contained a silver earring. No inscriptions or figural sculpture originated from the cemetery. Three graves contained two individuals; the rest held a single burial, or else the number of burials was not recorded. The age and gender distributions were unusual: females formed 83% of the sexed sample, and 44% of individuals were under 18 years old (28% under 12). If this sample is representative, the cemetery seems to have been reserved for women and children. The adult burials contained more finds than those of the children.

304

APPENDIX 1

TELL KAZEL150

The ancient site of Tell Kazel, sometimes identified with the Simyra of 1st millennium BCE sources, is located on the Syrian coast between the al-Kabir river (ancient Eleutheros) in the north and Tartous in the south. Hellenistic remains were found in Area I in the western half of the tell, where a large mudbrick building and a rural installation were uncovered, dating to the end of the 2nd –early 1st c. BCE. A Hellenistic and Roman cemetery occupied Area II in the southeast quarter of the tell. No Roman settlement has been reported.

The Hellenistic-Roman Cemetery151 In Area II, ninety-six tombs were found, of which forty-five could be dated to the Hellenistic period (300–60 BCE). At least three tombs contained glass from the Roman period (1st –2nd c CE, Online Appendix Tell Kazel 1), and two more were possibly Roman in date (200–1 BCE, Online Appendix Tell Kazel 2). The date of the other tombs is unknown, but since the type and burial rite were similar to those of Hellenistic-Roman tombs, they may have dated to the same period. The edges of the cemetery were not found during the excavation, leaving the total surface area uncertain. The tombs appear to have been located outside the settled (Hellenistic) area, although the size of the contemporary settlement is unclear. As already mentioned, nothing is known about a Roman-period habitation, other than a mention of a Roman tower or castle at or near the site.152 The Hellenistic assemblage included thirty-three cist-graves, seven pit-graves, and five jar-burials. The latter type was reserved for infants. The finds are listed in Table 4. The Roman tombs (1–200 CE) followed the Hellenistic traditions and included four cist-graves and one jar burial. The tombs contained a small number of objects, mostly (glass) vessels and jewelry (Chart 30). The range and type were the same as in the Hellenistic period, except that pottery bottles had been replaced by glass examples.

TYRE153

The modern city of Tyre/Sur (Roman Tyrus) is located 80 km south of Beirut. The original settlement was partly on an island and partly on the mainland (Figure 4). Tyre became an important Phoenician political center in the early 150

Badre 1991; Badre & Gubel 1999–2000; Capet 2003; Dunand & Saliby 1957; Dunand et al. 1964. 151 152 Badre et al. 1990, 1994. Bounni 1997, 275–276. 153 Bikai 1992; Bikai & Bikai 1987; Butcher 2003; Chéhab 1970, 1983; de Jong 2010; Jidejian 1996; Kahwagi-Janho 2007; Lipi´nski 2004, 300; Millar 1993; Poidebard 1939; Sartre 2005.

TYRE

6 5 4 3 2 1 0

chart 30. Distribution of grave goods, Tell Kazel

Iron Age (1200–550 BCE). The burial fields of this period were situated on the mainland. In the al-Bass region, Aubet excavated an Iron Age cemetery (9th–6th c. BCE), consisting of cinerary urns in pits possibly marked by stone stelae.154 Tombs of the 8th c.BCE were discovered at al-Rashidiya,3.5 km south of Tyre.155 The so-called “Tomb of Hiram” stood at a location 6 km southeast of Tyre (Figure 21). This possibly Achaemenid-period (550–330 BCE) tomb consisted of a large stone sarcophagus on a 3 m-high stone platform marking the entrance to a hypogeum.156 Chéhab discovered tombs with stelae dating to the same period at Bursj el-Shemali.157 When Alexander the Great besieged the city in 332/1 BCE, he built a mole and connected the island with the mainland. In subsequent centuries, sediments deposited on either side of the mole turned the island into a peninsula. No architecture remained from the Hellenistic period. Numismatic evidence indicates that Tyre received Roman metropolis status in the late 1st c. CE.158 In 193 CE, when Septimius Severus restructured the provinces of the Near East, the province of Syria Phoenice included Tyre as a main urban center. The same emperor bestowed on the city the title of colonia (Septimia Tyrus Metropolis Colonia) and additional privileges in reward for support during his war against Pescennius Niger. Textual and material evidence indicates that Tyre was an important commercial center for linen and purple dye in the first centuries CE and a local production center for glass. Excavations have yielded archaeological remains from the Roman period on the island, the mainland, and the newly created isthmus.

154

Aubet 2004; Aubet et al. 1998–1999. See also Sader 1991 and Seeden 1991. Described by Chéhab (1942–1943, 86) as large caves in which the deceased were placed, with evidence for secondary burial in large pottery vessels. 156 157 158 Jidejian 1996, 24–28. Chéhab 1934b, 44. Butcher 2003, 101. 155

305

306

APPENDIX 1

61. Map of Tyre and surroundings

The Cemeteries159 Burial grounds stretched out in the plain and foothills to the north, east, and south of the Tyrian peninsula (Figure 61). The cat. 1 sample contains fortyseven tombs, constructed between the 1st and 3rd–4th c. CE and often used 159

Barbet & Vibert-Guigue 1994, 102, 170, plates II, III; Bikai 1992, 30; Bikai et al. 1996; Chéhab, 1934a, 1935, 1968, 1983, 1984, 1985, 1986; de Jong 2010; Dunand 1965; Dussart 1998; Hakimian 1987, 205; Jennings 2004–2005; Jidejian 1996, 163, 166–168, 171–172; Koch & Sichtermann 1982; Le Lasseur 1922a, 1922b; Linant de Bellefonds 1985; Parlasca 1982, 7; Renan 1864–1874, 580–590; Rey-Coquais 1977, 2006; Salamé-Sarkis 1986, 200; Shibasaki et al. 2006, 52–53, figs. 3.19, 3.20; Takase et al. 2003, 326; Venit 2002, 179–180; Ward-Perkins 1969; Will 1946–1948, 160; www.culture.gov.lb/tomboftyre/index.htm?.

TYRE

45 40 35 30 25 20 15 10 5 0

307

cat. 1 cat. 2

chart 31. Distribution of tomb types, Tyre

well into the Byzantine era (Online Appendix Tyre 1). The reports of Renan and other early explorers identify extensive burial grounds of possible Roman date to the south and east of Tyre. These are listed as cat. 2, and represent a minimum of sixty-four tombs (Online Appendix Tyre 2). The painted tomb of Deb’aal, found 14 km east of Tyre, is described separately on p. 241. The types of tomb are listed in Chart 31. Along the main east–west road leading to Tyre extended the urban cemetery (al-Bass). Chéhab published thirty-nine tombs from this location, but excavated more. The cemetery is described separately later. It is unclear how far it extended eastward. Tombs along this road were discovered as far as 1.40 km from the al-Bass Cemetery in the Mashuk area, and may have been part of the same burial ground (Figure 60).160 Renan described the hills east of the plain of Tyre, reaching a distance of 3.6 km between Kabr-Hiram and Mogharet el-Suk, as littered with tombs.161 None was dated, but the discovery of glass unguentaria and lead sarcophagi suggests a Roman or Byzantine date for at least some of them (cat. 2). A necropolis at al-Awatin, ca. 3 km east of Tyre, yielded a hypogeum of the late 1st or 2nd c. CE with elaborately painted decoration of floral and mythological motifs, as well as two graffiti (cat. 1, Figures 20c, 37, 40). A Greek inscription was placed on the left wall.162 The finds from the stacked burial niches (loculi) and a pit-grave in the floor included a terracotta sarcophagus, many pottery bottles, and some coins and lamps (Chart 32). From 160

Renan (1864–1874, 580–584) reported the finds of many tombs at Mashuk, the type of which is not clear, although they included large stone coffins and terracotta sarcophagi; see also Le Lasseur 1922b, 132–134. Macridy (1904) found several Roman caves in the Tell Mashuk area, uncovering many fragments of marble sarcophagi in the sand between Mashuk and Rashidiye. 161 Renan 1864–1874, 591. 162 Dunand 1965. Rey-Coquais (2006, 153) lists four (Greek, untranslated) inscriptions that came from a double tomb (?) close to the one excavated by Dunand; see also Renan 1864–1874, 589–590.

308

APPENDIX 1

120

100

80

60

40

20

0

chart 32. Distribution of grave goods Djel el-’Amed/al-Awatin Tombs, Tyre

a side room that was not used for burial came twelve oil lamps, placed on a pedestal. At least forty-three individuals were buried in the hypogeum, but no information is available about their location, age, or sex. In the 1920s, Le Lasseur discovered a cemetery at Djel el-’Amed, in the foothills approximately 4 km east of Tyre. She published one hypogeum and six pit-graves from this location (cat. 1).163 No map exists of the burial ground. The main room of the hypogeum was painted with polychrome geometric, faunal, and floral motifs, as well as garlands (Figure 20b). Elaborate medallions with flowers, birds, and four busts, possibly personifications of the Winds, covered the ceiling of this room. In addition to loculi and pit-graves in the floor, burial took place in three stone sarcophagi. The hypogeum was cut between the late 1st and mid 2nd c. CE and surrounded by six pit-graves cut in bedrock, dug at a slightly later date (cat.1). The painted tomb and one pit-grave yielded jewelry, vessels, lamps, and coins (Chart 32). The cat. 2 assemblage included many tombs from the region east of Tyre, mostly hypogea and sarcophagi of unknown location. Two necropoleis with an undetermined number of tombs were recorded at al-Awatin and the bottom of Mashuk Hill. South of Tyre, more hypogea with lead and terracotta coffins and a stone sarcophagus with a woman depicted on the lid were found. From 163

Le Lasseur 1922a, 1922b.

TYRE

the area northeast of Tyre originated two stelae with Latin inscriptions. The remainder of the tombs and funerary material come from undisclosed locations. The tombs of the Tyre region yielded eight cat. 1 inscriptions and eighteen cat. 2: two Latin and the rest Greek. Most inscriptions recorded only the name of the deceased: three for women, four for men, one for a male child, and one for a woman and a man. An unprovencanced painted fragment commemorated a 12-year-old boy who seemed to have excelled in wrestling. Three foundation inscriptions were erected by men, for themselves, their father, and their son.

Al-Bass Cemetery In total, thirty-nine tombs are published from the al-Bass Cemetery, two of which did not contain any graves (cat. 1). Between ten and twenty-one funerary enclosures remain unpublished and/or unexcavated (cat. 2). A group of sarcophagi found before the excavations started also likely belonged to the same cemetery, and a stele with a Greek epitaph for a woman was recorded in the same general area (cat. 2). The burial ground was in use between the late 1st and the 7th c. CE, with a heavy concentration of building activity in the 2nd and 3rd c. CE. After the 4th c. CE, few new tombs were constructed, but older ones were reused on a large scale. The cemetery extended on both sides of the main east–west road that led into the city. It covered 1.5 ha but was not completely excavated (Figure 62).164 In the late 1st c. CE, the tombs lay scattered at several meters from the road. This space filled in over time, and by the 3rd c. CE, the burial ground consisted of abutting tombs, with little or no space between the tombs and the road. Several non-funerary buildings also arose in the al-Bass area, such as an aqueduct, a circus, a monumental arch, and the Apollo shrine (1st and 2nd c. CE). The main road was paved in the same period (Figure 4).165 By the Byzantine period, the tombs had expanded into the main road, which was replaced with a smaller paved version. The al-Bass Cemetery yielded one type of tomb, termed a “funerary enclosure”: a large, enclosed space housing multiple graves (Figures 4, 11).166 Inside the enclosures, different burial types occurred side by side, most commonly in the form of loculus graves and freestanding stone sarcophagi placed on a low earth pedestal inside the funerary enclosures, on the roofs of the 164

The total surface area was probably 2.5 ha or more. If the cemetery extended towards Mashuk, 1.4 km east of the city, the surface area would have covered ca. 4.2 ha. 165 There is some evidence that gardens existed at the back end of the tombs, as indicated by basins, canals, and at one location an irrigation system. However, none of these features can be positively connected with the centuries before the Byzantine reuse of the space. 166 Enclosures measured on average 21.23 × 12.46 m, but varied in size. There were an average of twenty-two individual burial spots per enclosure, ranging from two to ninety-six. Each enclosure held funerary platforms (average four, with twelve loculi) and sarcophagi (average 9.5).

309

310

APPENDIX 1

62. Plan of al-Bass Cemetery

TYRE

funerary platforms, or outside along the walls facing the road. Most were carved from local stone, but a substantial number were constructed from more exotic materials imported from Turkey, Greece, and Egypt (Figures 17a, 30b, 40).167 Occasionally, limestone, lead, and terracotta coffins were placed inside the loculi, and in one case two lead sarcophagi filled a larger limestone coffin (all unpublished). Aside from the sarcophagi and loculi, the final type of grave within the funerary enclosures is represented by a small number of underground pit-graves dug into the soil and covered with stone slabs. The thirty-nine published tombs contained a minimum of 825 places for burial: nineteen pit-graves, 357 sarcophagi, and 449 loculi in 145 funerary platforms. Benches and water features (basins, canals, and pipes) were found inside and around the enclosures, the latter probably mostly associated with the latest (Byzantine) phase of the cemetery. Occasionally, a room was found with no graves, and two enclosures did not produce graves (Tomb Complexes 8, 38). At the back of most funerary enclosures extended open areas, sometimes surrounded by low walls and crossed by canals, perhaps indicative of garden areas. It is not clear if these dated to before the Byzantine period. Four altars in the shape of small rectangular blocks situated in front of loculi or near sarcophagi and seven reused column stumps of marble and sandstone perhaps functioned in similar ways. Different forms of decoration adorned the tomb complexes. Lime plaster often covered the walls of the enclosures and the interiors of the loculi. Beveled cornices in relief crowned the funerary platforms, and engaged columns or arches occasionally framed the loculi. At least twentyseven mosaics covered the floors of rooms and, less frequently, the roofs of the funerary platforms. Few are dated, but white mosaics and black-and-white mosaics with geometric patterns were often associated with Roman-period tomb complexes, whereas polychrome examples with geometric and floral motifs came from the Byzantine period. Seven Greek funerary inscriptions come from the al-Bass Cemetery. One, consisting of a date, was placed in a mosaic floor (154/155 CE). A painted inscription refers to the Maioumas festival (4th c. CE) and was possibly not funerary in nature. The rest adorned sarcophagi; these included three foundation inscriptions for a male child, a commemorative text for a servant of the procurator of the province, and a text for a deceased male. A large number of inscriptions (at least 253) dated to the 5th or 6th c. CE. These texts were several centuries younger than the tomb complexes they adorned and represent the Byzantine phase of reuse of the graves.

167

Most sarcophagi (207) were made of local lime- and sandstone, but fifty-eight marble sarcophagi were imported from Proconnesos in Turkey, twelve from Attica, and nineteen from unknown locations. A further fifteen were in gray-purple stone (Assos?) and five in pink granite and gray porphyrite (Egypt); the material of the remainder is not recorded.

311

312

APPENDIX 1

table 8. Distribution of grave goods, al-Bass Cemetery, Tyre earring bracelet ring pin amulet necklace pendant bell gold leaves (set) other jewelry textile belt buckle glass unguentarium glass bottle glass flask glass amphoriskos glass small jar glass small vessel glass goblet glass pitcher glass unknown shape pottery bottle pottery amphoriskos pottery small vessel pottery plate

148 36 30 42 13 41 16 20 8 25 1 2 30 29 10 1 7 12 1 8 48 101 1 8 1

pottery pitcher pottery jar pottery amphora pottery pithos pottery unknown shape bronze/copper vessel bronze coin gold coin silver coin billion coin coin unknown material lamp whorl/loomweight iron scissors mortar pestle lance bronze key bronze weights bronze prism sculpture terracotta figurine bronze/copper mirror nail unknown material/type

5 5 3 1 63 1 142 4 15 3 14 86 1 1 1 1 1 1 1 1 1 3 1 17 55

The graves yielded a total of 6526 finds, most commonly items of gold, silver, and bronze jewelry, glass and pottery vessels, and bronze coins, and less frequently terracotta oil lamps, bone clothing pins, and glass and stone beads. The average was eleven finds per sarcophagus and sixteen per loculus; the remainder came from the vicinity of the tombs. Not all finds have been fully published, and stratigraphic reportage lacks. Because of the high degree of reuse in each grave, it is generally unclear when an object was deposited. The grave goods of 108 loculi in twenty-eight funerary enclosures could be more precisely dated to the 1st–4th c. CE and give an impression of the range of artifacts that ended up in Tyrian tombs. The total number of artifacts was 1066 (Table 8, Figures 24b,c, 25a,b). Compared to the rest of the province, the graves of Tyre were particularly full (an average of nine per burial spot), but the graves were also frequently reused. The tombs contained more items of personal adornment, with many earrings but relatively few items of clothing within that category. Coins made up a large part of the assemblage as well, whereas lamps were few. There are interesting differences between the different forms of burial within the enclosures. The loculi had the fewest items of personal adornment, more lamps, and more pottery compared to the sarcophagus

TYRE

burials. The numbers of finds in the loculi that contained terracotta and lead sarcophagi were in between. Some finds were discovered outside the burial spots, such as pottery and glass bottles, lamps, a few items of jewelry, and a figurine. It is uncertain whether these were the result of robbing activity, or were deposited in front of a sarcophagus or burial platform on purpose. A minimum of 3955 individuals are reported in the excavation reports, stemming from 825 burial spots. The level of reuse of these tombs was substantial, but lack of reportage means we cannot establish at what moment the reuse took place (Roman or Byzantine). Those burial spots that can be dated more precisely to the Roman period on the basis of the grave goods held 913 individuals in sixty-nine loculi (average thirteen, range one to forty-nine). Two individuals were identified as children, based on unknown criteria. No other information is available about the skeletal remains.

313

APPENDIX 2

TOMB TYPES

TOMB TYPES – KNOWN TYPES (CHARTS 1, 3, TABLE 9)

table 9. Tomb types per site (known types) cist- funerary jarpitsarcophagus towergrave enclosure hypogeum burial mausoleum grave in the open tomb tumulus (17) (54) (424) (24) (135) (468) air (10) (181) (82) Apamea

Cat. 1 Cat. 2 Baalbek Cat.1 Cat. 2 Beirut Cat. 1 Cat. 2 Bosra Cat.1 Cat. 2 Deb’aal Cat. 1 Dura Cat. 1 Europos Cat. 2 Hama Cat. 1 Cat. 2 Hauran Cat. 1 Cat. 2 Homs Cat. 1 Cat. 2 Jebleh Cat. 1 Cat. 2 Limestone Cat. 1 Cat. 2 Plateau Palmyra Cat. 1 Cat. 2 Selenkahiye Cat. 1 Cat. 2 Tell Kazel Cat. 1 Cat. 2 Tyre Cat. 1 Cat. 2

2 1

1 4

2 13

1 1

4 1

3

45 3 26 4 4 1 7 7 6 2 3 1 4 7 8 57 60 26 120

1

6

3

3 1

2 1

1 4 10 10

30 27 1

20 250 20

1

3 75

1 6 1 16 49 3

3 1

7 5 11 13 2

7 4 13 54 36

1 19 162 4

1 39 10

2 26

6

Cist-Graves (Figure 8) Cist-graves consist of pits aligned with stones or bricks forming a cist. Tiles, slabs, or earth covered the burial pit. In total, seventeen were discovered, and two are listed in the cat.2 database.They come from most areas of Roman Syria. 314

TOMB TYPES – KNOWN TYPES

Cist-graves made of stone, mudbrick, and baked brick have a long history in the region (Chart 5).1 The type largely disappeared after the 2nd c. CE. One cist-grave contained a terracotta sarcophagus.

Funerary Enclosures (Figures 4, 11) A total of forty-three funerary enclosures were found at Beirut and the al-Bass Cemetery in Tyre, and the same sites yielded between eleven and twenty-two cat. 2 examples. The earliest occurrence of this type was in the late 1st c. CE; its construction ceased after the 4th c. CE, although the use of the funerary enclosures continued well into the Byzantine period (5th or 6th c. CE). The type consisted of a large, enclosed space divided into several rooms or sections. A perimeter wall of limestone and sandstone blocks enclosed a rectangular area without evidence of roofing. The tombs on the flat plain of Tyre were entirely built in stone, whereas the funerary enclosures on the hill slopes around Beirut were partly cut in bedrock.Posts,lintels,thresholds,and sometimes arches facing the road formed doorways into the enclosures. The interior was often paved or covered by a plaster floor, and divided into several compartments by low partition walls. Different burial types occurred side by side, in the form of rock-cut pits, loculus graves, and sarcophagi. The loculus graves were stacked horizontally or vertically in built platforms, and closed with stone slabs. The sarcophagi stood on low earth pedestals inside the funerary enclosures, on the roofs of the funerary platforms, or outside along the walls, facing the road. Most were carved from local stone, but a substantial number were constructed from more exotic materials imported from Turkey, Greece, and Egypt.2 Occasionally, limestone, lead, and terracotta coffins were placed inside the loculi, and in one case two lead sarcophagi filled a larger limestone coffin. Pit-graves could also be dug into the soil of the enclosure and covered with stone slabs. In Beirut, the pit-graves consisted of two or three stacked compartments. The enclosures held on average twenty-two burial spots. Lime plaster often covered the walls of the enclosures and the façades of the funerary platforms, as well as the interiors of the loculi. Beveled cornices in relief crowned the funerary platforms, and engaged columns or arches occasionally framed the loculi.In the al-Bass Cemetery at Tyre,at least twenty-seven mosaics covered the floors of rooms, and less frequently the roofs of the funerary platforms. Few are dated, but white mosaics and black-and-white mosaics with geometric patterns were often associated with Roman-period tomb 1

2

Achaemenid examples come from Tell el-Der (Gasche 1996) and Umm el-Marra (Schwartz et al. 2003, 353). Hellenistic examples were found at Abu Qubur, Nimrud, Seleucia-on-theTigris, Tell ed-Der, and Tell Kazel (Chart 2.5), as well as Sidon (Dunand 1969). See de Jong 2010 for additional discussion of the imported sarcophagi.

315

316

APPENDIX 2

complexes, whereas polychrome examples with geometric and floral motifs came from the Byzantine period. The walls of the enclosures in Beirut were decorated with floral and faunal motives in paint and plaster moldings. One stone with a plaster eagle was discovered near Complex I in Beirut. Funerary enclosures contained relatively few inscriptions from the Roman period; only seven were discovered at Tyre. A large number of inscriptions, at least 253, dating to the 5th or 6th c. CE, are testament to the Byzantine phase of reuse of the al-Bass graves. Aside from graves, the funerary enclosures included additional features, such as benches constructed against the interior walls. Not all space in the enclosures was used for burial, as paved rooms and gardens were sometimes incorporated in the structures. Evidence for water features and gardens in the al-Bass burial ground possibly dated to the Byzantine phase. The funerary enclosure had no predecessors before the Roman period, although some components of the tomb type, such as pit-graves and the use of sculpted sarcophagi, pre-dated the Roman centuries.

Hypogea (Figures 9, 28, 29, 31, 32, 37, 38) The most common form of tomb used for communal burial was the hypogeum, or underground chamber-tomb. The cat. 1 sample includes 117 hypogea, excluding the tower-tomb with hypogea from Palmyra, which were discussed separately. The cat. 2 sample includes a minimum of 307, and demonstrates the widespread regional and diachronic distribution of the type. Hypogea were used well into the Byzantine era. Large hypogea could be in use for 200 years or more. Inscriptions at Palmyra, for instance, indicate that tombs constructed in the late 1st and 2nd c. CE still received burials in the mid 3rd c. CE. Similar evidence comes from the Habbasi Tomb in Hama, built in 101 CE or earlier, and containing sarcophagi of the 3rd c. CE, glass of the 3rd or 4th c. CE, and Christian cross decoration that post-dates 330 CE. Many variations existed in the form and execution of the hypogeum, but the basic type consisted of an entrance corridor (dromos) leading to a small vestibule closed by a door or stone slab. This entrance opened into a central rectangular or square chamber, with rectangular burial niches (loculi) cut in three walls. Burial took place on the floor of these niches, or in coffins hollowed out in the bedrock walls. Most loculi were perpendicular to the room, opening on the short side. In northern Syria, a group of simple hypogea contained parallel loculi, and a small number of hypogea in northern and southern Syria included both parallel and perpendicular loculi. The use of parallel loculi occurred mostly later, after the 1st c. CE, and possibly coincided with the introduction of arched alcoves. In these arcosolia, burial took place in rock-cut coffins lying parallel to the funerary chamber. Burial also occurred in freestanding stone sarcophagi placed on the tomb floor and in pit-graves cut in

TOMB TYPES – KNOWN TYPES

the chamber floor. Many burial spots yielded evidence for wooden, lead, and terracotta sarcophagi. Several hypogea had additional rooms, which could have been added after the initial construction.3 The average number of burial spots was twenty-eight, but this is dominated by the large Palmyrene hypogea, which could hold hundreds of burials. Without these, the average was six burial spots per hypogeum, and three-quarters had fewer than ten spots. Dura Europos, Palmyra, Jebleh, Hama, the Hauran, the hills behind Tyre, and the Limestone Plateau each had a distinct style of hypogeum construction and decoration. The burial space of the hypogeum was located below the surface, but the hypogea often incorporated an aboveground portion (at least 34% of total) in the form of built architecture or a decorated façade. Double columns, stelae, circumference walls, a stepped platform, an exedra, and a small building stood on top of hypogea in the Hauran and the Limestone Plateau, and tumuli covered those at Dura Europos. Rock-reliefs in the shapes of porches, awnings, and stylized temple-fronts embellished the façades of hypogea in the Limestone Plateau. The tumuli pre-dated the Roman period; the other aboveground features were introduced in the 2nd c. CE and subsequent centuries. Various forms of decoration embellished the walls of more than half of the hypogea. The number of decorated tombs, as well as the quantity of decoration per tomb, increased throughout the Roman period: in the late 1st c. CE in Palmyra, and in the 2nd or 3rd c. CE elsewhere. Exterior decoration consisted of moldings, cornices, engaged columns, and the architectural features alredy mentioned. The doors themselves, when preserved, were decorated as well. Exterior decoration was especially common in northern Syria. The interior decoration of hypogea was more diverse and widespread. Walls were plastered and covered with molded cornices and pilasters, as well as figural reliefs. Two regions yielded elaborately painted tombs. Tombs at Palmyra dating to the 2nd–3rd c. CE had extensively painted interior walls, usually depicting Victories and floral motives, and in one case mythological scenes depicting Ganymede and Achilles. Southern Lebanon and northern Palestine yielded a series of painted tombs, mostly dated to the late 1st–2nd c. CE. The subject matter included mythological scenes, as well as architectural motifs framing the loculi and floral and faunal patterns. In Tyre and Hama, the tradition of applying painted decoration to the tomb continued into the Byzantine period. Hypogea had a long history before the Roman period. Hellenistic and Parthian examples come from Jebleh, Dura Europos, Beirut, and perhaps Palmyra (Chart 5). Sometimes, older Iron Age tombs continued to be used in the Hellenistic period. Pre-Roman hypogea also displayed regional 3

Two hypogea in Hama consisted of a single rectangular room with rounded edges and burial on the floor (G XXI, XXVIII), which may have been an older form. Persian-period tombs with a similar layout are listed in Stern 1982, 82–83.

317

318

APPENDIX 2

variation in construction,but were usually not decorated,and were smaller than the Roman examples.4 With the exception of the tumuli on the Parthian tombs at Dura Europos, none of the pre-Roman Parthian and Hellenistic hypogea included an aboveground marker. Some followed a fairly irregular plan, and others were more regular. Burial in loculi was an older phenomenon, but the move to parallel loculi and arcosolia was new in the Roman period.

Jar-Burials (Figure 13) The cat. 1 sample includes nine jar-burials, each consisting of a large to medium-sized jar placed in a pit with a single burial. A total of fifteen are in the cat. 2 database. This type is used for both infant (3) and cremation (5) burials. One grave with two jars at Jebleh held a skeleton of an adolescent and an older individual. The cat. 2 material produced at least thirteen cremation burials from the North Cemetery at Apamea and two jar-burials with remains of children at Dura Europos. The dates of these burials ranged from the 1st c. BCE or CE to the mid 3rd c. CE. They came out of older traditions of placing child burials in jars, and were more frequent beyond the eastern edges of the Roman province in the Parthian–Mesopotamian realm.5

Mausolea (Figures 1, 3, 7, 10) Mausolea are built communal tombs in which burial took place aboveground. The cat. 1 assemblage includes fifty-eight mausolea in addition to the tower-tombs from Palmyra, which are discussed as a separate group. Another seventy-seven are included in the cat. 2 list. Their construction concentrated on Central–West Syria, in the corridor between Antioch, Cyrrhus, and the Euphrates in northern Syria, down to the Hauran in the south and Palmyra in the east. The coastal Levant did not yield mausolea, although the funerary enclosures in this area shared many characteristics. The earliest group of mausolea comes from the Hauran (1st c. BCE–1st c. CE). Their construction increased in subsequent centuries, and mausolea continued to be built in the Hauran, Limestone Plateau, and the Homs area in the Byzantine period; there is no direct evidence for continuity elsewhere. As with the hypogea, these communal tombs likely received burials for several decades or even centuries, but few concrete data are available about the patterns of use. 4

5

Achaemenid tombs at two sites, Amrit and Tyre/Kabr Hiram, did have an aboveground marker. Amrit: Dunand 1953; Dunand et al. 1954–1955; Renan 1864–1874, 59–90; Saliby 1989; Tomb of Hiram: Jidejian 1996, 24–28; Renan 1864–1874, plates 47, 48. Seleucia-on-the-Tigris: Graziosi 1968–1969; Hopkins 1972; Invernizzi 1967; Negro Ponzi 1970–1971, 1972, 2002; Valtz 1986, 1988; Waterman 1931; Yeivin 1933. Tell Sheikh Hamad: Novák 2000. Uruk: Boehmer et al. 1995; Pedde 1995.

TOMB TYPES – KNOWN TYPES

The great diversity in size, shape, and finish of the mausolea highlights strong regional trends in this tomb type. Circular mausolea occurred only in the Hauran and were primarily constructed in the 1st c. BCE–1st c. CE. A small vestibule in the circular walls of these mausolea led to the main and usually rectangular room, where burial took place in loculi and/or on the floor of the chamber. Circular mausolea were sparsely decorated, but on occasion low reliefs and moldings adorned the exterior wall and entrance. The earliest noncircular mausoleum was the Tomb of Hamrath in Suweida (Figure 3), dated between 100 BCE and 25 CE. By the 2nd c. CE, rectangular and square mausolea had replaced the circular types in the Hauran and were common elsewhere in the province. Although each was unique in form, the rectangular mausolea did share common features. The solid stone walls were usually decorated on the exterior with moldings, low reliefs, and engaged columns, focused on the surface around the monumental door. Inside, burial occurred in loculi and probably in sarcophagi. A type of mausoleum characteristic of the Limestone Plateau is the so-called “tetrastyle tomb,” consisting of four columns or pillars supporting a roof that was often pyramid-shaped. Its earliest occurrence was in the 3rd c. CE (Brad), and the type continued until at least the 6th c. CE. Tower-shaped mausolea were square buildings that were higher than wide – at least several stories high. This type is found in Palmyra and the Middle Euphrates zone, and is described separately. A second group was predominant in the Hauran; the Tomb of Samsigeramos at Homs also falls into this category (Figure 1). The latter was the only tomb in Syria built in opus reticulatum. Its exterior architectural decoration, with engaged pilasters, pediments, friezes, and a pyramid-shaped roof, however, resembled that of a mausoleum at Serrin in the Syrian Upper Euphrates region (73 CE) and at Hermel in the Beqa’ Valley (see p. 320).6 The mausolea in Palmyra were constructed between the mid 2nd and mid 3rd c. CE and followed four stylistic groups. Their façades ranged from relatively plain, with a flat roof, to elaborately sculpted, in a manner reminiscent of temple and theater architecture. Schmidt-Colinet has drawn attention to inspiration from both Roman architecture and Parthian palatial elements in the Palmyrene mausolea.7 Temple-shaped mausolea were also found on the Limestone Plateau (4th c. CE), and possibly in the Hauran (3rd c. CE). In a few cases, a mausoleum was combined with a hypogeum, for instance at Qanawat (Q9) and the tower-tombs on top of hypogea at Palmyra. The average number of burial spots was eighteen, but this number is high because of the large Palmyrene examples, without which the average was 8.6.

6

7

Gogräfe 1995. Although the large multistoried building of 43–44 CE in Qal’at Faqra in the Lebanese mountains is often mentioned as an example of a tower-shaped mausoleum, it is unlikely that it was a tomb. Schmidt-Colinet 1997, 164–166.

319

320

APPENDIX 2

Starting in the 2nd c. CE, mausolea received increasingly elaborate decoration. Exterior decoration consisted of architectural features such as columns, pilasters, pediments, aediculae, and on occasion entire temple fronts. The decoration often centered on the door and lintel. The low reliefs representing weapons and armor on the Hamrath tomb at Suweida were unusual (Figure 3). The same is true for a 3rd c. CE mausoleum at Apamea, constructed in brick. Much less is known about the interior decoration of the mausolea, other than in the Palmyrene examples. The interiors of mausolea outside Palmyra were rarely studied or accessible. The mausoleum was a new type in the Roman period: no aboveground tombs date to the Hellenistic period. A possible exception is the 28 m-high tower at Hermel in the northern Beqa’, decorated with corner pilasters and shallow reliefs of hunting scenes on the exterior walls. Proposed dates for this building range from the late 2nd/early 1st c. BCE to the 1st c. CE. Its funerary nature is uncertain,however:before it was heavily restored in the 1930s,descriptions of the building mention that it had a solid interior, and no hypogeum or burial cave was discovered in the vicinity. Because the exterior reliefs resembled those on the Hamrath Tomb in Suweida, the building entered the scholarship as a tomb. However, the hunting scenes on its walls have little in common with the depictions of arms and armor in Suweida. The restored tower at Hermel looks very similar to a drawing of the Tomb of Samsigeramos in Homs made in 1799, to such a degree that one wonders if this drawing was the basis of the Hermel reconstruction.8 Nevertheless, the recent discovery of pit-graves 4 m south of the Hermel building, unfortunately of unknown date, perhaps confirms its funerary nature and repeats a pattern also seen around the Homs tomb.9 If the structure at Hermel was indeed a tomb, and of Hellenistic date, its shape was unique for the period.

Pit-Graves (Figure 8) Pit-graves were the most common type of single tomb. The cat. 1 assemblage includes 170 examples, and the cat. 2 list adds a minimum of 298. The large number of funerary stelae originating from Roman Syria perhaps originally marked pit-graves. Cut into the bedrock or dug into the earth, this type consisted of a single burial in a pit covered with stone slabs, terracotta tiles, a sarcophagus lid, or heaps of earth and rubble. At Besandina, a rock-cut pit-grave 8

9

For instance, Fedak 1990, 148. For the Hermel tomb, see Collart 1973, 144; Cormack 1997b, 348; Nordiguian 2004, 125; Renan 1864–1874, 117; Toynbee 1971, 172; Will 1949b, 273–274. Early accounts: Krencker & Zschietzschmann 1938, 161–162; Perdrizet 1938. For restoration, see Anus 1932, 295–297. Drawing of 1799: Nordiguian 2004. Nordiguian 2004, 125.

TOMB TYPES – KNOWN TYPES

lay under a carved awning and was marked by a stele. An inverted marble sarcophagus covered a burial in a pit in Beirut. The pit-graves at Nawa-tell Umm al-Hauran consisted of a double pit, wherein the smaller burial pit was closed with basalt slabs. Regional variation in construction is demonstrated by storage jars covering three pit-graves at Selenkahiye and other pits in the region. A sarcophagus lid covered pit-graves at Herbet Kalil and Burdaqli, and this was a common mode of closure on the Limestone Plateau and elsewhere in Northwest Syria between the mid 2nd c. CE and the Byzantine period.10 The Northeast Cemetery at Palmyra consisted of fifty-four pit-graves marked by stone stelae (Figure 16a). These graves reportedly held plaster or terracotta sarcophagi. Two pit-graves at Homs contained terracotta coffins, and a lead coffin came from a pit-grave in Apamea. Stone sarcophagi occupied pits in Selenkahiye (1) and Douris (15). The remains of wooden coffins came from at least fourteen pits. Most pit-graves contained a single individual, but this is not always recorded. One example at Selenkahiye contained two individuals, and two stelae in Palmyra commemorated two people each, which suggests co-burial. Pit-graves were an old and common form of burial long before the coming of Rome. Examples from eastern Syria and Mesopotamia under the Seleucids and the Parthians held rectangular or tub-shaped terracotta coffins (Chart 5).11

Sarcophagi in the Open Air (Figure 14) Many sarcophagi were found in Syria, mostly without context. A small group, however, formed its own type: sarcophagi in the open air – at least four groups composed of a total of eight stone sarcophagi stood on a raised position outside and were not associated with other tombs. At Apamea and Bosra, a single sarcophagus rose high on a platform of limestone blocks. At Si’, three sarcophagi placed in triclinium position dominated a triangular basalt pedestal. Apamea yielded a similar group of three coffins. The cat. 2 sample, with six groups, suggests that this type of tomb was particularly common in the Limestone Plateau in the 2nd and 3rd c. CE, continuing into the Byzantine period. When the type of stone was recorded, the coffins were basalt or limestone. They were plain or covered in low relief on the box and lid. Stone coffins in the funerary enclosures of Beirut and Tyre replicated this tomb type, as they stood in an elevated position on funerary platforms, built stone podia, or earth and stone pedestals. This region perhaps represented a link between older practices and the Roman 10

11

Limestone Plateau: Sardin, Turin. Elsewhere: Kasr Naus (north of Beirut; Krencker & Zschietzschmann 1938,19), Gerade (30 km south of Apamea; Mouterde 1949, 22), and perhaps Machnaka (Byblos area; see p. 22). Tell Kannas in the Selenkahiye region: Finet 1977, 81–83. Examples from Mesopotamia come from Tell Sheikh Hamad (Novák 2000) and Nimrud (Oates & Oates 1958).

321

322

APPENDIX 2

type of elevated sarcophagus. The Roman type had no direct predecessors in the region, although it may have been inspired by the nearby Tomb of Hiram (see p. 71, Figure 21).

Tower-Tombs (Figures 12, 46) The enigmatic tower-tombs of Palmyra have received more attention than the other tombs in Roman Syria. Tower-tombs were square mausolea that were higher than wide and had several stories, accessible by a winding interior staircase. The roof was probably flat. The nineteen tower-tombs in the database were situated at Palmyra, where they were constructed between the late 1st c. BCE and the early 2nd c. CE. At least 162 examples from this site are included in the cat. 2 collection. Many stood on elevated positions in the desert landscape around Palmyra. The oldest examples (50–1 BCE) had loculi that were accessible from the exterior façade. Over time, burial moved inside, to tiers with stacked loculi separated by stone or terracotta slabs in the sidewalls and, starting in the mid-2nd c. CE, to sarcophagi. The towers grew in size, or at least base dimensions, which grew from ca. 6 × 6 m in the 1st c. BCE to 12 × 12 m in the early 2nd c. CE. As the burial spots moved to the interior, their number increased as well. The largest tower-tombs held several hundred spots. Whereas the earliest tower-tombs were only plastered inside, those of the late 1st and 2nd c. CE were decorated with reliefs, pilasters, moldings, and coffered ceilings. The loculi of early tower-tombs were closed with breakstones and mortar, on which the name of the deceased was written. Loculus slabs with portrait busts closed some of the later examples. New to the later 1st c. CE was decoration of the exterior with molded cornices and lintels, perhaps including figural representations. Tower-tombs marking hypogea were characteristic of the West Cemetery, although one was found in the North Cemetery. Tombs of this type (twenty-two in the database) were built in the 1st c. CE and used until the late 2nd c. CE. Burial took place in the hypogeum and the tower portion of the tomb, thereby increasing the number of burial spots. The towers were slightly smaller (8.3 × 8.3 m) than those of the regular, contemporary tower-tombs. The interior decoration was similar to that of the regular towers, but also included painted motifs. The façades now sported reclining group scenes, and many were inscribed. The Palmyrene mummies originated from both types of tower-tomb. Nine altars are reported from several tombs, where they were anchored in the floor. The grand and lavishly decorated tower-tombs of the later 1st and 2nd c. CE had no parallels outside of Palmyra. The earliest tower-tombs, however, particularly those with loculi on the exterior, compare to a series of tombs constructed along the Euphrates north and east of Palmyra. None are securely dated or completely preserved. Stone tower-tombs at Dura Europos, el-Susa,

TOMB TYPES – KNOWN TYPES

and Tabuz shared with the Palmyrene examples exterior loculi, a stepped base, and a winding internal staircase. The towers at Dura, tentatively dated between 50 BCE and 50 CE, had decorated façades with corner pilasters and engaged columns, a style that is defined by Clauss as typically Parthian. This was absent from the Palmyrene towers. Close to el-Susa stood the five tower-tombs of Baghuz (Figure 12c). As in Palmyra, they occupied an elevated position in the landscape and were spread over a large area, perhaps reaching 3.5 km in length. These tower-tombs contained exterior and interior loculi, and perhaps space for a sarcophagus. The façades, on the other hand, were similar to those of the Dura and el-Susa tombs, to which Clauss considers them contemporary. The varied floorplans of the Palmyrene tombs corresponded more closely with a group of twenty-six or more tower-tombs at Halabiye. Engaged columns and corner pilasters embellished the façades of these tombs. Busts, a Victory, and an eagle sculpture embellished the interior of at least one tomb in Palmyrene style at Halabiye (#4), according to Toll. He dates the towers to the 3rd–4th c. CE, but this is questioned by Clauss: based on comparison with the other sites, she places their construction in the 1st c. BCE–1st c. CE.12 The earliest towers at Palmyra, therefore, fall within a tradition of funerary architecture common along the Middle Euphrates, perhaps reaching into modern Iraq. The oldest examples date to the mid or late 1st c. BCE. Those on the Euphrates, in Parthian territory, were decked with Parthian-inspired decoration, whereas the Palmyrene examples remained plain. Often, the tower-tombs stood in or close to cemeteries containing other types of tomb, e.g., hypogea covered by tumuli in Baghuz and Dura, perhaps jar-burials and a terracotta sarcophagus at el-Susa, hypogea and pit-graves at Halabiye, and hypogea at Palmyra. Some of these graves were contemporary with the towers, whereas others were older, such as the Bronze Age cemeteries at Baghuz and Susa. In the 1st c. CE, burials in the Palmyrene towers moved to the interior, and at the same time the decoration increased. In the same century, perhaps around 50 CE, tower building ceased at the other sites. As the tombs at Halabiye followed later Palmyrene examples, to some degree, they may date later than the rest. Scholars debate the origin of the tower-tombs. Pre-Hellenistic tombs such as those at Amrit are sometimes mentioned in conjunction with them, but the style of construction and mode of burial were very different.13 A tradition of 12

13

Clauss 2002. Henning (2013, 101–116) maintains Toll’s date. Halabiye: Lauffray 1991, 192–224; Toll 1937. Dura Europos and Baghuz: Clauss 2002; Toll 1946. El-Susa: Geyer & Monchambert 2003. Toll (1946, 147) reports one similar tower-tomb near Qaim in northern Iraq, ca. 20 km southeast of Baghuz, with a stepped base and traces of engaged columns, and another example at Tabus (28 km from Halabiye), with corner pilasters and engaged columns, and one interior loculus. There is confusion in the texts about a tower-tomb he mentions west of Dibseh (at Neshabah), which is in northern Iraq according to Clauss (2002) and in Qal’at Djaber according to Will (1949b). Ball 2000, 366–367; Will 1949b.

323

324

APPENDIX 2

tower building in Phoenician times perhaps had an impact on the tombs (see p. 73). Others have pointed to Persia and Arabia.14 The latter identification is based on the use of the term “nfs” (p. 159). A full discussion of this term, meaning “self/soul” and “funerary monument,” lies outside the scope of this book, but it appeared on a few Palmyrene tower-tombs. It was used to designate a tomb in southern Arabian funerary inscriptions of the 5th c. BCE and later centuries. Clauss and Mouton have pointed out that this region has yielded tower-shaped tombs (Mleha in Oman, end 3rd c. BCE), suggesting a link with Palmyra and a connection to trade routes and the sedentarization/urbanization of nomadic tribes.15 Although this claim is attractive, the evidence is slim, as it is not clear how much the Arabian tombs resembled those at Palmyra, and only a few Palmyrene inscriptions appear to use the term “nfs” to describe the towers.16 To summarize, the tower-tombs at Palmyra and the Euphrates sites, constructed between the second half or the final quarter of the 1st c. BCE and the early 2nd c. CE, formed a distinct group of mausolea, without obvious predecessors.

Tumuli (Figure 15) Tumuli are communal tombs consisting of a circular mound of earth and/or rubble in which pit-graves were dug, or which covered a built grave chamber. The mound could rest on a low stone wall. The cat. 1 database contains only three tumulus-tombs, but the type was more widespread both during and before the Roman period. A minimum of seventy-nine are included in the cat. 2 sample. At least two and possibly more regional traditions in tumulus construction existed. The three cat. 1 tumuli originated from Si’ and Suweida in the Hauran. Each consisted of a rubble mound (8–4 diam) covering a small burial chamber with a central support pillar. Their construction dated to the 1st c. BCE and the 1st c. CE, and one example remained in use until the mid or late 2nd c. CE. Tumuli of possibly similar date come from Bosra and R¯ımet alLohf (cat. 2). Many others are reported from the Hauran, but it remains unclear whether these were true tumulus-tombs or, instead, circular mausolea. The tradition can be traced back in the Hauran and neighboring Golan to before the Bronze Age. One Chalcolithic example from Der’a was reused in the Bronze Age, and possibly again in the 1st c. CE.17 The limited descriptions we have of the tumuli do not allow further comparison of the type across the centuries. The second region that yielded tumulus-tombs was the Syrian Upper Euphrates area around the Taqba dam. Selenkahiye yielded four tumuli from 14 16

17

15 Persia: Watzinger 1932. Arabia: Clauss 2002. Clauss 2002, 178–181; Mouton 1997. At some stage, epitaphs accompanying Palmyrene funerary reliefs also mentioned the term (Colledge 1976; Gawlikowski 1972). Nasrallah 1950. Other tumuli in the Hauran: Sartre-Fauriat 2001, vol. II, 52; 2007b, 239.

TOMB TYPES – UNKNOWN TYPES

325

the 3rd c. CE or later (cat. 2). A stone and earth mound, up to 11.5 m in diameter, covered a mudbrick bench-like structure, into which an oval pit-grave was deepened. Similar tombs with one or multiple pit- and cist-graves originated from the vicinity of Selenkahiye, the earliest of which was perhaps built in the 3rd c. CE. Bronze Age tumuli also dotted the landscape of this region, again indicating a much older origin of these types of tomb. Further downstream, in the area between the confluence with the Balikh and Dura Europos, several unpublished tombs are reported around Mari and Baghuz. The description does not make clear whether these represented true tumuli or, instead, covered hypogea like those in nearby Dura Europos and Halabiye. In any case, their date, between the mid 2nd c. BCE and the 1st c. CE, places these tombs in the Parthian period of eastern Syria.18

TOMB TYPES – UNKNOWN TYPES (CHARTS 2, 4, TABLE 10)

table 10. Tomb types per site (unknown types) Coffin (108) Apamea Baalbek Beirut Bosra Dura Europos Hama Hauran Homs Jebleh Limestone Plateau Palmyra Selenkahiye Tyre

18

Cat. 1 Cat. 2 Cat. 1 Cat. 2 Cat. 1 Cat. 2 Cat. 1 Cat. 2 Cat. 1 Cat. 2 Cat. 2 Cat. 1 Cat. 2 Cat. 2 Cat. 2 Cat. 1 Cat. 2 Cat. 2 Cat. 2 Cat. 2

Rock-relief (13)

48 96 3 19 1 1 7 169 1 1 1 1 37 73

5 5 5 55 22

1 2 3 2 2

6

Stele (474)

5 8

2 4 1 9

Other types (325) 3 17 8 96

3 92 49 1 10 29 4 13

For Dura Europos, see p. 110. Halabiye: Lauffray 1991. One true, but undated, tumulus was excavated 14 km southeast of Bseira/Circesium (Kraeling 1952). Middle Euphrates: Geyer & Monchambert 2003, 163–166. Region of Selenkahiye: examples are reported at Habuba Kabira, Hajj Ali, Mumbaqa, Rumeilah, Tell ’Anab as-Safina, and Tell Qattar (Bounni 1977, 1980; Egami et al. 1979; Zaqzûq 2001). Possible Roman tumuli also came from Msaytba in North–Central Syria (Gatier & Rousset 2010, 161–164). Bronze Age tumuli: Porter 2002.

326

APPENDIX 2

226

96

96

92

41

40 16

limestone

marble

other stone (92)

lead

wood

terracotta

4 basalt

unknown

chart 33. Coffin materials

Coffins (Figures 4, 11, 14, 17, 19d, 30b, 31, 40, 46) The total number of coffins in cat. 1 assemblage is 611, the majority of which (596) were placed in the mausolea, hypogea, or funerary enclosures, whereas others originally filled pit-graves and loculi. Four groups with eight coffins in total formed a separate group, labeled “sarcophagi in the open air,” which are described later. For seven coffins, the original location is unknown. The cat. 2 assemblage includes at least 101. This section describes the entire assemblage of coffins. Chart 33 lists the different materials used for coffins (cat.1).The wooden type was likely the most widespread type, but was rarely preserved. The presence of such coffins is usually deduced from finds of multiple iron nails (some with wood remains attached) and metal coffin fittings.Fragments came from all types of tomb, but the best-preserved examples originated from underground types such as pit-graves and hypogea. Excavation reports from Dura Europos state that each loculus contained a wooden coffin. Pit-graves at Homs yielded elaborately constructed examples with lead, silver, iron, and gold fittings and ornaments, whereas an example from Tomb G XXI in Hama was covered by stucco decoration. Wooden examples were common in the pre-Roman period and originated from Hellenistic and Parthian Dura Europos, Jebel Khalid, the Baalshamin tomb and Tomb G at Palmyra, and tombs at Bey 018 and 040 in Beirut. Terracotta coffins were also encountered throughout the Roman period, in cist-graves, enclosures, hypogea, and pit-graves. There are few studies of these coffins in Roman Syria, which seem to have followed two regional stylistic trends. First, rectangular coffins closed with terracotta tiles were common on the Mediterranean coast.19 Second, rectangular coffins with rounded edges 19

Analysis of examples from Beirut has identified their place of production in eastern Cilicia (Butcher 2003, 200; Stuart 2001). See also Shapiro 1997 for Cilician sarcophagi in Galilee.

TOMB TYPES – UNKNOWN TYPES

(trough-shape) came from Dura Europos and were dated to the Roman period (late 2nd c. CE) by Toll. They were influenced by Mesopotamian–Parthian traditions. Eastern Syria has also yielded several more rounded or tub-shaped sarcophagi in the Middle Euphrates region and Tell Sheikh Hamad, but none could be securely dated to the Roman period (Figure 19d). The tub shape was common in Mesopotamia, however, and was discovered in Hellenistic cemeteries of Uruk. Gawlikowski notes that the pit-graves of the Northeast Cemetery at Palmyra contained so-called “slipper-shaped coffins”; if this is correct, this cemetery would be the westernmost find location of a coffin common in Parthian Mesopotamia.20 Both eastern Syria and the coastal Levant had much older and more diverse modes of burial in terracotta examples, as evidenced by anthropoid coffins at Arados on the Syrian coast (late 6th–4th c. BCE), torpedo-shapes at Neo-Assyrian Tell Kn¯ediˆg in northeastern Syria, and the aforementioned tub-shaped terracotta coffins in Hellenistic and Parthian eastern Syria.21 Lead coffins were a new addition to the assemblage in the mid to late 2nd c. CE, and retained their popularity into the Byzantine period. Predominant in the western part of the province, they were found in hypogea, pit-graves, and funerary enclosures. Several examples from the Deb’aal tomb were placed inside a stone sarcophagus. The thirty-one examples in the database originated from Deb’aal, Apamea, Beirut, and Tyre (seventeen tombs). Production centers presumably arose along the Levantine coast.22 Relief decoration covered the lids and boxes of the rectangular or trapezoid coffins, consisting of geometric and floral motives, depictions of architecture (temple façade, columns), masks, sphinxes, lions, and figural motifs of standing or seated persons. Stone sarcophagi were discovered in all types of tomb except tumuli and cist-graves (Figures 4, 11, 17, 30b, 40, 46). The cat. 1 sample included 410 stone coffins, the majority of which (357) originated from the al-Bass Cemetery at Tyre.The cat.2 assemblage yielded more stone coffins from all other sites except Dura Europos, Tell Kazel, and Selenkahiye. There is little evidence for the use of stone sarcophagi before the late 1st c. CE. On the other hand, since the use

20

21

22

Gawlikowski 1970, 34. Slipper sarcophagi were found in Parthian Uruk and Seleucia-onthe-Tigris (Graziosi 1968–1969, 49; Hopkins 1972, 40–46, 92; Invernizzi 1967, 21–22; Negro Ponzi 1970–1971, 35; 2002, 72; Valtz 1986, 16–18; Yeivin 1933, 56–62). Rectangular terracotta coffins are known from Parthian Tell Sheikh Hamad (Novák 2000), Assur (Andrae & Lenzen 1933), Babylon (Reuther 1926), and Hellenistic Nimrud (Oates & Oates 1958). Tub-shaped terracotta coffins: Geyer & Monchambert 2003, 164. Uruk: tub-sarcophagi (8) and reversed tub-sarcophagi (19) belonged to the Hellenistic period (Boehmer et al. 1995; Pedde 1995). Dating of sarcophagi by Toll 1946, 95–100. Kulemann-Ossen & Martin 2000, 495; Lembke 2001, 16. Other examples are listed in Nunn 2001, 398. More examples originated from Sidon, around Hama, and Latakia (Chéhab 1934a, 1935). For lead sarcophagi from the region, see also Koch & Sichtermann 1982 and Rahmani 1999.

327

328

APPENDIX 2

of stone examples was a long-standing tradition in the Near East, this could be the result of excavation bias. A plain sarcophagus from Beirut (Bey 040) perhaps dated to before the 1st c. CE. The evidence does clearly demonstrate that the use of stone (decorated) coffins rose in popularity later in the 1st c. CE, and especially starting in the mid 2nd c. CE. Most coffins were made of local lime- and sandstone or basalt. At least 111 coffins in the cat. 1 sample were made of non-local (i.e., imported) stone, the majority of marble. The earliest imports date perhaps to the mid 2nd c. CE. Two were found in Beirut; the rest stem from Tyre. Marble sarcophagi were imported from Proconnesos in Turkey (58), Attica in Greece (12), Egypt (5: pink granite and gray porphyrite), and unknown locations (19). Another fifteen were in gray-purple stone, perhaps Assos stone (Turkey). They were usually found at coastal sites with easy harbor access to the sea, but sometimes travelled further inland, as illustrated by a Pentelic marble sarcophagus at Bosra (150–260 CE, cat. 2) and a Proconnesian marble sarcophagus at Rastan (ancient Arethusa) near Homs (first half of the 3rd c. CE).23 Popular decoration of the sarcophagi consisted of moldings, rosettes, and an uninscribed tabella ansata. Some of the imported sarcophagi had elaborate reliefs with depictions from Greek mythology, such as the life of Achilles and Bacchic scenes. Reclining figures carved in the round topped eight lids. The assemblage also included three strigil sarcophagi and six lids with roof-tile imitation. Garland sarcophagi complete with bucrania, ram’s and bull’s heads, flowers, fruit, and Medusa faces framed by heavy garlands formed a group of sixteen coffins. Other Proconnesian imports (21–23) were still in so-called “quarry state,” with stylized garlands and disks in relief panels. Most came from the al-Bass Cemetery in Tyre and dated to the 3rd c. CE. The locally produced coffins were plain, followed the decoration schemes of the imported coffins, or followed other stylistic patterns. The latter group displayed strong regional styles. Group reclining scenes, for instance, covered examples from Palmyra, as did busts, camels, and other features typical of life in this city. These sculpted coffins appear to have been mid to late 2nd c. CE. Although this motif occurred elsewhere, characteristic of basalt coffins from Bosra are lion heads in relief. Panels with disks and pelta shields covered sarcophagi at Douris and Baalbek. Pre-Roman stone coffins primarily originated from the Levantine coast. Several cemeteries in Sidon, for instance, yielded stone anthropoid sarcophagi from the late 6th–mid 4th c. BCE, some in marble imported from Egypt. From 23

Bosra: Sartre-Fauriat 2001, vol. I, 218. Restan: Gatier 1997–1998. More Proconnesian coffins were found in Beirut (cat. 2) and Byblos (Ward-Perkins 1969, 142–143). Assos stone coffins were found near Sidon (Contenau 1920, 152; Ward-Perkins 1969, 128) and Tripoli (Koch & Sichtermann 1982, 562). From Rome: Tyre, Sidon, perhaps Beirut (Koch & Sichtermann 1982, 562). A marble child sarcophagus or ossuary was reported from Jebleh (cat. 2 Jebleh – 9). Koch (1977, 388–389) lists two sarcophagi, found in Tyre and Beirut, originating from Rome.

TOMB TYPES – UNKNOWN TYPES

the same site and period stemmed imported marble Greek and Lycian sarcophagi dating to the late 5th and 4th c. BCE.24 This group includes the Alexander sarcophagus, which, as evidenced by its subject matter, dated to the Hellenistic period (325–300 BCE). So far, this is the only stone coffin securely dated to this period.

Rock-Reliefs (Figure 18) The next tomb type is difficult to define or place in the larger context of funerary practices in Roman Syria. A cliff face at Qatura on the Limestone Plateau contained at least thirteen reliefs depicting seated or standing people in a round or rectangular frame. The individuals wore long robes and appeared alone or in groups of two to four. One relief portrayed an eagle over a reclining person. Inscriptions ran along the bottom of five reliefs, one with a date in the 2nd c. CE. These five are part of the cat. 1 assemblage; the remainder are listed as cat. 2. The framed reliefs were not associated with graves, but the inscriptions were similar to those crowning tombs elsewhere in the region. Butler believed that pit-graves were located underneath the niches in the bottom of the ravine, whereas others considered the reliefs to be cenotaphs.25 The type of decoration, the combination of figural relief and text, and the text of the inscriptions find close parallel in funerary stelae from the Limestone Plateau and elsewhere in the province, and confirm their funerary or commemorative nature. Relief sculpture on rocky outcrops associated with cemeteries existed elsewhere in Northwest Syria, the Lebanese mountains, and the region of Hama and Damascus, but their publication record is poor, and few are dated. These reliefs fall into five stylistic groups. The first includes the Qatura reliefs and consisted of frontal depictions of seated, standing, or reclining persons (single or in a group) in framed niches or aediculae. Examples from the Limestone Plateau come from el-Hallougha (Jebel Wastani), where groups of seated and standing people decorated the entrances of at least three hypogea. At Baftamun (Jebel Wastani), four panels close to the entrance of a series of hypogea of low relief depicted a seated couple, three standing couples, and two seated couples. Outside the Limestone Plateau, at Abila (Souq Wadi Barada, near Damascus), two persons sat in an aedicula next to a hypogeum entrance. A pit-grave at the same cemetery was marked by a relief of a bust in a frame on the rock face. The site of Hirbet Qanafar (Hermon/Beqa’) included a relief of a seated couple with a floating head or bust between them, next to a hypogeum. The 24

25

Hamdy Bey & Reinach 1892; Jidejian 1971, 121, 131–137; Smith & Ertu˘g 2001, 71–90. Anthropoid coffins in local basalt and imported marble were discovered in Arwad and Arados on the Levantine coast (Elayi & Haykal 1996; Lembke 2001, 6–16). For instance, Butler 1920, 249–250; Griesheimer 1997a, 170; Peña et al. 1999, 154–163.

329

330

APPENDIX 2

Qatura group dates to the 2nd c. CE, but information about the date of the other reliefs is lacking.If they are of similar date,they highlight a trend common throughout the hillside from South to North Syria. A different type of rock-relief is found in the Orontes area. The cliff walls surrounding hypogea at Bdama, ca. 45 km northwest of Apamea, had at least twenty-three reliefs in the shape of stelae, consisting of a projecting rectangle with pediment or arched top. Some included inscriptions, dated between 161/171 and 231/232 CE, and one depicted a reclining scene. The mountains east of Byblos contained a third group of rock-reliefs. At Ghineh, reliefs were carved immediately on top of an opening to rock-cut tomb(s): a man with a short tunic and spear pointed toward a bear, a man in similar dress with two dogs, and a seated woman in a niche. At nearby Jrapta, a relief depicted a sacrificial scene with four people in profile (one male with priestly cap, one female, two children) around an altar; a female bust rose above the altar niche. Tombs, perhaps hypogea, opened several meters away from the reliefs. Reliefs at Machnaka in the same region depicted three scenes: a profile of a seated figure (woman) in an aedicula, flanked by two standing figures; a standing man depicted frontally, also in an aedicula and flanked by two standing figures; and a standing figure in a rounded niche. Pit-graves were cut in the vicinity. Parlasca has dated the reliefs in this group to the Hellenistic period, but without explaining his reasoning.26 The depiction in profile of some of the Machnaka reliefs compares to stelae from Hellenistic Umm el-’Amed, but the type of pit-grave at this site, covered by a sarcophagus lid, was common only in the Roman period. The fourth group of rock-reliefs depicted stylized stelae or cone-shaped projections. The Hellenistic cemetery at Umm el-’Amed may have yielded examples, as did nearby Tannoura, where these reliefs were juxtaposed with a relief of a soldier. El-Ferzol, in the same region, produced a hypogeum adorned with a cone-shaped relief in a niche above the entrance. Scholars often consider these aniconic shapes of pillars, cones, and pyramids to represent nfs/nefesh, or the “soul” of the deceased, but they do not elaborate upon this point (p. 159). At Wadi Qana, east of Tyre, similar cones adorned a cliff wall, together with groups of mostly frontally depicted standing individuals in long robes and stelae of quadrangular shape with rounded tops. No tombs were discovered in the vicinity of these reliefs. Based on the clothing, Kaoukabani dates the Qana group to the late Hellenistic–early Roman period.27 The final group of rockreliefs consisted of arched niches with crude and stylized human busts, found

26

27

Parlasca 1982, 7. El-Hallougha, Baftamun: Peña et al. 1999, 44–45. Abila, Hirbet Qanafar: Mouterde 1951, 47–48, 84–87. Bdama: Mouterde 1949. Ghineh, Jrapta, Machnaka: Renan 1864–1874, 238, 284–194. Kaoukabani 1971.

TOMB TYPES – UNKNOWN TYPES

at Turin on the Limestone Plateau and Rasm en-Nahal on the southwest edge of the Jabboul salt lake. The Turin examples were associated with tombs, but their date remains unknown.28 To summarize, a trend of relief-carvings of elaborates scenes near tombs dating to the Hellenistic and/or Roman period existed in Central Lebanon (Byblos), and one of depictions of people, cones, and stelae existed in South Lebanon. A different type of relief showing standing, reclining, and sitting persons was found in the vicinity of tombs in the northern hillside of Syria, perhaps extending southward to Damascus. The examples from Turin, Rasm en-Nahal, and Bdama perhaps fell in the same tradition. The dated examples from this group fell in the 2nd and 3rd c. CE. This sculptural tradition requires more extensive research than that presented here, but some conclusions can be drawn. In several areas of the Roman province, cliff walls depicted iconic and aniconic scenes. Their spatial association with tombs, pit-graves, and hypogea suggests a funerary nature, but we cannot be certain about their exact function. It is possible that these sculptural traditions drew on older – or, at least, Hellenistic – practices.

Stelae (Figures 16, 30c, 33, 34) Roman Syria yielded more than 474 stone stelae, of which the majority are known only by their inscription.Funerary stelae were common throughout the Roman province,except for the region east of the Euphrates,where the handful of extant inscriptions belonged to Roman soldiers.29 The majority of the stelae date to the 2nd and 3rd c. CE, and their number declined after the 3rd c. CE, although examples from the 4th and 5th c. CE come from the Hama region, Limestone Plateau, and Bosra. In some cases, the original placement of the stelae is recorded. A group of fifty-four stood atop pit-graves in the Northeast Cemetery of Palmyra (1/50–150 CE). Other stelae were associated with a pitgrave at Besandina or marked a hypogeum elsewhere on the Limestone Plateau. Stelae not only marked single and communal tombs but were also found inside them. The large hypogea in Palmyra yielded stelae that were similar to those in the Northeast Cemetery but likely functioned as closing slabs of loculi. At Umm el-Jimal in North Jordan, several rows containing twenty-three stelae in total were placed against an exterior wall of a partly sunken mausoleum. 28

29

Umm el-’Amed: Chéhab 1955; Dunand & Duru 1962. Tannoura, El-Ferzol: Mouterde 1951, 31–32, 52–54. Turin: Griesheimer 1997a, 170, fig. 5. Rasm en-Nahal: Mouterde & Poidebard 1945, 75 (pt. 1), plate XLII (pt. 2). These include one Latin funerary stele each at Dura Europos (chart Dura Europos cat.1–11), Sura (perhaps second half of the 1st c. CE; Konrad 2001, 9), and Nisibis/Qamishly (probably early 3rd c. CE; Gatier 1988, 227–229). A Greek stele commemorating a child was also found in Qamishly (Cumont 1933, 385).

331

332

APPENDIX 2

According to Butler, these stelae, dating after the 1st c. CE, were in situ.30 The shape of the stone sometimes provides an indication for its usage. The pointed base of Palmyrene stelae, for example, could be inserted into the soil, and the undecorated lower third portion of the stelae at Apamea was buried. It remains uncertain whether the stelae physically belonged to a grave or served a more commemorative function and were originally erected in a non-funerary setting (see discussion on cenotaphs, p. 158). The funerary stelae were made in stone, but the type of stone is rarely recorded. The majority of the cat. 1 stelae consisted of an image above or next to an inscription. The remainder consisted of only an inscription or, infrequently, a single image. The stelae were rectangular, with a flat, rounded, or triangular top. Other shapes included cippi (short square or round pillar), column drums, and altars, the latter often belonging to soldiers of the Roman army. The decoration was applied in low relief. Based on the decoration and the contents of the inscription, the stelae can be divided into several groups. First, Apamea yielded thirty-seven military stelae dating to the early–mid 3rd c. CE. These stelae, with a pediment-shaped top, depicted the deceased soldier with arms or horses above a panel containing a Latin inscription. Other military stelae with Latin inscriptions were discovered at Dura Europos, Baalbek (250–300 CE), Bosra (2nd–4th c. CE), and Beirut (an early example: 25– 75 CE). Cat. 2 examples are reported from Apamea, Baalbek, Bosra, Homs, and Tyre. The second group of stelae came from the Northeast Cemetery at Palmyra. This group of the mid 1st to mid 2nd c. CE consisted of fiftyfour stelae, twenty-eight of which had an inscription in Palmyrene only. The remainder depicted a curtain hanging between palms or suspended in midair above the inscription. Three stelae portrayed standing men and women, sometimes in combination with a curtain. Outside Palmyra, the cat. 1 stelae varied in content and location. Two stood atop a hypogeum of the 2nd and 3rd c. CE on the Limestone Plateau, one in the form of a tall square pillar. Less is known about the stelae with Greek inscriptions found at Apamea, Baalbek, and four sites on the Limestone Plateau: Bazliq, Besandina, Deir Seta, and Ir-Rubbeh. The cat. 2 material and the stelae from other locations in Syria demonstrate strong regional diversity throughout the Roman province. A distinct group from Sidon (2nd–3rd c. CE) consisted of cippi with a stylized decoration of geometric and flower patterns on top and an inscription in the lower part. Another group depicting double or triple busts originated from the area between Belkis/Seleucia-on-the-Euphrates and Membidj/Hierapolis 30

Butler 1920, 209–210; Sartre-Fauriat 2001, vol. I, 206–208. One hypogeum at Khirbet Senbalte on the Limestone Plateau contained two stelae inserted into slits in front of the hypogeum (Peña et al. 1999, 124). Perhaps the same type was found at Baftamun (Peña et al. 1999, 44–45).

TOMB TYPES – UNKNOWN TYPES

(perhaps late 1st–3rd c. CE). The region around Antioch yielded stelae from the 1st–2nd c. CE with reclining and standing single persons. The Hauran yielded ca. 2000 basalt stelae, mostly unpublished, often representing stylized busts of single individuals or groups of two or three persons (mostly 2nd and 3rd c. CE).31 The use of stone funerary stelae pre-dates the establishment of the Roman province, but this practice was not widespread (Chart 5). Examples from the Hellenistic period displayed regional variation, such as a group of painted stelae from Sidon belonging to Greek mercenaries of the Seleucid or Ptolemaic army (2nd c. BCE).32 Four rectangular stelae with rounded tops depicting standing women and men and a winged sun-disk, with a Phoenician inscription below the paneled decoration, come from a location approximately 2 km south of Umm el-’Amed (south of Tyre). These are dated 250–150 BCE.33 Parlasca and Balty list additional stelae with Greek inscriptions from the coastal Levant, the Antioch area, and Apamea (2), but their Hellenistic date remains uncertain.34 The same is true for two stelae from a cemetery in the hills at Bursj el-Shemali, 2 km east of Tyre. One of these stelae depicted two framed columns and the second a person with a raised hand. The latter was found at the entrance of a rock-cut chamber tomb or hypogeum. Chéhab dated them to the 3rd or 2nd c. BCE, whereas Sader placed the first in the first half of the 5th c. BCE. An unusual stele originating from the vicinity of Tyre depicted two female mourners around an anthropoid sarcophagus. The image was crowned by a laurel wreath and had traces of blue and red paint. Seyrig is unsure of its date and places it somewhere between the 6th and 3rd c. BCE, but Parlasca lists it as a Hellenistic stele. A stele inscribed in Carthaginian Punic was found southeast of Tyre and is dated by Sader to the 4th–3rd c. BCE.35 There were stylistic parallels between the Hellenistic stelae in the Antioch area and the later examples

31

32 33

34 35

Sartre-Fauriat 2001, vol. I, 245–267, vol. II, 103–122. See also Weber 2006. Other types are listed in Parlasca 1982 and Rey-Coquais 1998. Antioch region: Kondoleon 2000, 139–141 and Weir 2001. Belkis-Membidj: Parlasca 1982. Sidon: Gatier 2001; see also Parlasca 1982, Taf. 22.1. Jidejian 1971, 70–71; Macridy 1904, 547–548; Mendel 1912, 258–259. Umm el-’Amed: Chéhab 1955; Dunand & Duru 1962. See also Renan 1864–1874, 749. Because of similarities with stelae found in a sanctuary at Umm el-’Amed, it is not certain whether these were strictly funerary or more commemorative in nature. In addition to these examples, twelve similar stelae came from unknown contexts in Umm el-’Amed and seven from the temple area, but their funerary nature is uncertain. Iron Age stelae were found in the Syro-Hittite territories in North Syria and Southeast Anatolia (9th–7th c. BCE), Sidon (6th–4th c. BCE), and the Iron Age cemetery (9th–6th c. BCE) at Tyre (Bonatz 2000, 189; Kurtz & Boardman 1971, 322). Balty 1981, 208; Parlasca 1982. See also Seyrig 1939. Sader 2005, 80–81. Bursj el-Shemali: Chéhab 1934b; Sader 2005, 76–77. Tyre: Parlasca 1982, 8, Taf. 5.4; Seyrig 1940, 120–122.

333

334

APPENDIX 2

from Apamea and Dura (see also Chapter 4). Whereas the Palmyrene stelae had some affinity with reliefs of pre-Roman standing figures, the depiction of a curtain was new. Also novel to the Roman period was the inclusion of reclining figures and busts. Group scenes were uncommon before the Roman period.

LIST OF ONLINE APPENDICES

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. 32.

Online Appendix Palmyra 2, #9 (Tomb 206) Online Appendix Palmyra 2, #9 (Tomb 207) Online Appendix Palmyra 2, #9 (Tomb 208) Online Appendix Palmyra 2, #9 (Tomb 209) Online Appendix Limestone Plateau 1, #28 (Tomb 1/Brad) Online Appendix Hauran 2, #59 (Philippeion/Shahb¯a) Online Appendix Limestone Plateau 1, #27 (Tomb of Tiberius/Beshindlaya) Online Appendix Limestone Plateau 1, #44 (Tomb 1/Herbet Kalil) Online Appendix Limestone Plateau 1, #59 (Tomb 1/Moshon Online Appendix Limestone Plateau 1, #75 (Sarcophagus 1 (IGLS 505)/ Tell ’Aqribin) Online Appendix Dura Europos 1, #11 (Stele 1) Online Appendix Limestone Plateau 1, #15 (Tomb 2/Baboutta) Online Appendix Limestone Plateau 1, #26 (Tomb 8/Besandina) Online Appendix Limestone Plateau 1, #27 (Tomb of Tiberius/Beshindlaye) Online Appendix Limestone Plateau 1, #18 (Mausoleum 1/Bazliq) Online Appendix Bosra 2, #207 (Sarcophagus 5) Online Appendix Apamea 2, #10 (Sculpture) Online Appendix Hauran 2, #54 (Stele 25) Online Appendix Baalbek 2, #76 (Bust 1) Online Appendix Baalbek 2, #90 (2010–53) Online Appendix Baalbek 2, #93 (Tomb 3) Online Appendix Bosra 2, #203 (IGLS 9092) Online Appendix Bosra 2, #206 (IGLS 9392 Online Appendix Tyre 2, #27 (Inscription 111) Online Appendix Beirut 2, #25 (Hypogeum 3) Online Appendix Hama 1, #12 (G XXI) Online Appendix Limestone Plateau 1, #77 (Tomb 1/Turin) Online Appendix Palmyra 1, #4 (Hypogeum of Zabda) Online Appendix Deb’aal 1 Online Appendix Tyre 1, #26 (Complex 18) Online Appendix Hauran 1, #8 (Tomb 51/Naw¯a-tell Umm el-Haur¯an) Online Appendix Palmyra 1, #111 (Tomb C) 335

336

LIST OF ONLINE APPENDICES

33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 60. 61. 62. 63. 64. 65. 66. 67. 68. 69. 70. 71. 72.

Online Appendix Tyre 1, #1 (Hypogeum 1/Djel el-’Amed) Online Appendix Hauran 1, #50 (Tomb 1-a/Shahb¯a) Online Appendix Hama 1, #11 (G XV) Online Appendix Palmyra 1, #2 (Hypogeum of Yarhai) Online Appendix Selenkahiye 1, #2 (R 21.4) Online Appendix Homs 1, #3 (Tomb 1) Online Appendix Beirut 1, #1 (Complex 1) Online Appendix Hauran 1, #17 (Tomb 1/Naw¯a-tell Umm el-Haur¯an) Online Appendix Hauran 1, #22 (Tomb 133/Naw¯a-tell Umm el-Haur¯an) Online Appendix Beirut 1, #2 (Complex 2) Online Appendix Beirut 1, #13 (Sarcophagus 4) Online Appendix Dura Europos 2, #11 (Tomb 11). Online Appendix Tyre 1, #8 (Hypogeum 1/al-Awatin) Online Appendix Tyre 1, #35 (Complex 28) Online Appendix Palmyra 1, #93 (Hypogeum of ’Abd ’astor) Online Appendix Palmyra 1, #101 (Hypogeum of the Three Brothers) Online Appendix Palmyra 1, #13 (Tomb of Elahbel) Online Appendix Palmyra 1, #15 (Tomb of Iamlikh¯o) Online Appendix Apamea 1, #9 (Hypogeum 1) Online Appendix Limestone Plateau 1, #62 (Qatura/Tomb of Aemilius Reginus) Online Appendix Palmyra 1, #11 (Tomb of ’Atenatan) Online Appendix Palmyra 1, #20 (Tower-Tomb 046) Online Appendix Palmyra 1, #24 (Tower-Tomb of Elasha) Online Appendix Limestone Plateau 1, #74 (Tomb 1/Sitt er-Rum) Online Appendix Limestone Plateau 1, #49 (Tomb of Zoilos/Kwaro) Online Appendix Limestone Plateau 1, #61 (Tomb of T. Flavius Julianus/ Qatura) Online Appendix Limestone Plateau 1, #38 (Tomb 2/Daousat) Online Appendix Limestone Plateau 1, #42 (Tomb of Abedrapsas/Frikya) Online Appendix Limestone Plateau 1, #32 (Pit-Grave 2/Burdaqli) Online Appendix Limestone Plateau 1, #47 (Tomb 1/Kafr ’Aruq) Online Appendix Bosra 2, #210 (IGLS 9405) Online Appendix Palmyra 1, #87 (Stele 49) Online Appendix Palmyra 1, #88 (Stele 50) Online Appendix Hama 1, X15 (Habbasi Tomb) Online Appendix Selenkahiye 1, #39 (T 06.4–3) Online Appendix Baalbek 1, #1 (Tomb 1) Online Appendix Apamea 2, #9 (Stele 2) Online Appendix Bosra 2, #6 (IGLS 9434–9435) Online Appendix Jebleh 2, #9 (Sarcophagus 1) Online Appendix Homs 2, 122 (IGLS 2404)

LIST OF ONLINE APPENDICES

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.

Online Appendix Homs 2, #95 (IGLS 2371) Online Appendix Bosra 2, #130 (IGLS 9276) Online Appendix Beirut, #15 (Stele 1) Online Appendix Palmyra 1, #99 (Hypogeum of Nasrallat) Online Appendix Palmyra 1, #98 (Hypogeum of Malikho/Malkû) Online Appendix Palmyra 1, #101 (Hypogeum of the Three Brothers) Online Appendix Hauran 2, #66 (IGLS 428/Shahb¯a) Online Appendix Hauran 2, #76 (IGLS 427/Shahb¯a) Online Appendix Palmyra 1, #107 (Hypogeum of Artabn/Artaban) Online Appendix Bosra 2, #5 (IGLS 9179) Online Appendix Limestone Plateau 1, #29 (Tomb 2/Brad) Online Appendix Limestone Plateau 1, #72 (Hypogeum 1/Sermada) Online Appendix Tyre 1, #24 (Complex 16) Online Appendix Limestone Plateau 1, #62 (Tomb of Aemilius Reginus/ Qatura) Online Appendix Palmyra 2, #38 (IGLS 484) Online Appendix Palmyra 2, #39 (IGLS 485) Online Appendix Palmyra 2, #40 (IGLS 486) Online Appendix Palmyra 2, #41 (IGLS 487) Online Appendix Palmyra 2, #42 (IGLS 488) Online Appendix Palmyra 2, #43 (IGLS 489) Online Appendix Hauran 1, #49 (Tomb 1/R¯ımet al-Lohf) Online Appendix Apamea 1, #25 (Stele 10) Online Appendix Limestone Plateau 1, #57 (Tomb 1/Me’ez) Online Appendix Bosra 1, #8 (Tomb 8) Online Appendix Baalbek 1, #2 (Stele 1) Online Appendix Apamea 1, #44 (Stele 29) Online Appendix Homs 2, #128 (IGLS 2455) Online Appendix Bosra 2, #40 (IGLS 9363) Online Appendix Apamea 1, #19 (Stele 4) Online Appendix Apamea 1, #42 (stele 27) Online Appendix Baalbek 1, #3 (Stele 2) Online Appendix Hauran 1, #60 (Tomb of Hamrath/Suweida) Online Appendix Palmyra 1, #29 (Marona Tomb) Online Appendix Palmyra 1, #23 (Tower-Tomb 068) Online Appendix Palmyra 1, #38 (Tomb of Malikho) Online Appendix Homs 1, #2 (Tomb of Sampsigeramos)

337

BIBLIOGRAPHY

Abdul-Hak, S. 1952, “L’Hypogée de Taai a Palmyre,” Les Annales Archéologiques de Syrie, vol. 2, pp. 193–251. Abdul-Hak, S. 1954–1955, “Rapport préliminaire sur des objects provenant de la nécropole romaine située a proximité de Nawa (Hauran),” Les Annales Archéologiques de Syrie, vol. 4–5, pp. 163–188. Abdul-Hak, S. 1958–1959, “Découvertes archéologiques récentes dans les sites GrécoRomains de Syrie,” Les Annales Archéologiques de Syrie, vol. 8–9, pp. 81–100. Adams, R.M. 1965, Land behind Baghdad. A History of Settlement on the Diyala Plains, The University of Chicago Press, Chicago & London. al-As’ad, K. & Schmidt-Colinet, A. 1995, “Kulturbegegnung im Grenzbereich” in Palmyra: Kulturbegegnung im Grenzbereich, ed. A. Schmidt-Colinet, Verlag Philipp von Zabern, Mainz am Rhein, pp. 28–53. Albertson, F.C. 1983, “A Palmyrene funerary bust of a Roman cavalryman,” Damaszener Mitteilungen, vol. 12, pp. 141–153. Alcock, S.E. 1989, “Roman imperialism in the Greek landscape,”Journal of Roman Archaeology, vol. 93.2, pp. 5–34. Alcock, S.E. 1993, Graecia Capta. The Landscapes of Roman Greece, Cambridge University Press, Cambridge. Alexandridis, A. 2010, “Neutral bodies? Female portrait statue types from the late republic to the second century CE” in Material Culture and Social Identities in the Ancient World, eds. S. Hales & T. Hodos, Cambridge University Press, Cambridge, pp. 252–279. Allara, A. 1991, “1987 season at Dura-Europos, Syria. New informations on the chronology of the site” in Golf – Archäologie, eds. K.

Schippmann, A. Herling, & J. Salles, Verlag Marie L. Leidorf, Buch am Erlbach, pp. 19–26. al-Maqdissi, M. 1993, “Chronique des activités archéologiques en Syrie (I),” Syria, vol. 70, no. 3/4, pp. 443–560. Amy, R. & Seyrig, H. 1936, “Recherches dans la nécropole de Palmyre,” Syria, vol. 17, pp. 229–266. Anderson-Stojanovi´c, V.R. 1987, “The chronology and function of ceramic unguentaria,” American Journal of Archaeology, vol. 91, pp. 105–122. Andrae, W. & Lenzen, H. 1933, Die Partherstadt Assur, J.C. Hinrich’sche Buchhandlung, Leipzig. Anus,F.1932,“La protection des monuments historiques en Syrie et au liban,” Syria, vol. 13, no. 3, pp. 293–299. Appadurai, A. 1996, Modernity at Large: Cultural Dimensions of Globalization, University of Minnesota Press, Minneapolis, MN. Aubet, M.E. 2004, The Phoenician Cemetery of Tyre-al Bass. Excavations 1997–1999, Directorate General of Antiquites, Ministry of Culture, Lebanon, Beirut. Aubet,M.E.,Nunez,F.J.,& Trelliso,L.1998–1999, “The Phoenician cemetery of Tyre al-Bass,” Bulletin d’Archéologie et d’Architecture Libanaises, vol. 3, pp. 267–94. Avni, G., Dahari, U., & Kloner, A. 2008, The Necropolis of Bet Guvrin-Eleutheropolis, Israel Antiquities Authority, Jerusalem. Badawi, M. 2007, “Huit tombes Hellénistiques et Romaines à Jablé,” Syria, vol. 84, pp. 185–204. Badre, L. 1991, “Kazel (in Harvey Weiss, Archaeology of Syria),” American Journal of Archaeology, vol. 95, pp. 734–735.

339

340

BIBLIOGRAPHY

Badre, L. & Gubel, E. 1999–2000, “Tell Kazel, Syria. Excavations of the AUB Museum, 1993–1998. Third preliminary report,” Berytus, vol. 44, pp. 123–203. Badre, L., Gubel, E., al-Maqdissi, M., & Sader, H. 1990, “Tell Kazel, Syria. Excavations of the AUB Museum, 1985–1987. Preliminary report,” Berytus, vol. 38, pp. 9–124. Ball, W. 2000, Rome in the East. The Transformation of an Empire, Routledge, London & New York. Balty, J. (ed.) 1969, Apamée de Syrie. Bilan des recherches archéologiques 1965–1968,Centre Belge de recherches archéologiques à Apamée de Syrie, Brussels. Balty, J. (ed.) 1984, Apamée de Syrie. Bilan des recherches archéologiques 1973–1979, Centre Belge de recherches archéologiques à Apamée de Syrie, Brussels. Balty, J. & Balty, J.C. 1977, “Apamée de Syrie, archéologie et histoire. I. Des origines à la Tétrarchie” in Aufstieg und Niedergang der Römischen Welt, eds. H. Temporini & W. Haase, Walter de Gruyter, Berlin & New York, pp. 103–134. Balty, J.C. 1969, “L’Église à atrium de la grande colonnade” in Apamée de Syrie. Bilan des recherches archéologiques 1965–1968, ed. J. Balty, Centre Belge de recherches archéologiques à Apamée de Syrie, Brussels, pp. 75–80. Balty, J.C. 1972, “Le groupe Episcopal d’Apamée, dit ‘cathédrale de l’est’; premières recherches” in Apamée de Syrie. Bilan des recherches archéologiques 1969–1971, eds. J.C. Balty & J. Balty, Centre Belge de recherches archéologiques à Apamée de Syrie, Brussels, pp. 187–208. Balty, J.C. 1977, “Les grandes étapes de l’urbanisme d’Apamée-sur-l’Oronte,” Ktema, vol. 2, pp. 3–16. Balty, J.C. 1981, Guide d’Apamée, Centre Belge de recherches archéologiques à Apamée de Syrie, Brussels. Balty, J.C. 1988, “Apamea in Syria in the second and third centuries A.D.,”Journal of Roman Studies, vol. 78, pp. 91–104. Balty, J.C. 1994, “Grande colonnade et quartiers nord d’Apamée à la fin de l’époque hellénistique,” Comptes-rendus des séances de l’Académie

des inscriptions et belles-lettres, vol. 138, no. 1, pp. 77–101. Balty, J.C. & Balty, J. (eds.) 1972, Apamée de Syrie. Bilan des recherches archéologiques 1969–1971, Centre Belge de recherches archéologiques à Apamée de Syrie, Brussels. Balty, J.C. & van Rengen, W. 1993, Apamée de Syrie. Quartiers d’hiver de la IIe Légion Parthique. Monuments funéraires de la nécropole militaire, VUB Press, Brussels. Barbet,A.& Vibert-Guigue,C.1994,Les peintures des nécropoles romaines d’Abila et du nord de la Jordanie, Institut française d’archéologique du proche-orient, Beirut. Barrett, J.C. 1990, “The monumentality of death: the character of Early Bronze Age mortuary mounds in southern Britain,” World Archaeology, vol. 22, no. 2, pp. 179–189. Barrett, J.C. 1996, “The living, the dead, and the ancestors: Neolithic and Early Bronze Age mortuary practices” in Contemporary Archaeology in Theory, eds. R.W. Preucel & I. Hodder, Blackwell Publishing, Oxford & Malden, pp. 394–412. Bauman, Z. 1998, “On glocalization: or globalization for some, localization for some others,” Thesis Eleven, vol. 54, pp. 37–49. Baur, P.V.C. & Rostovtzeff, M.I. (eds.) 1929, The Excavations at Dura-Europos. Preliminary Report of the First Season of Work, Spring 1928, Yale University Press, New Haven, CT. Bhabha,H.K.1994,The Location of Culture,Routledge, London & New York. Bielfeldt, R. 2003, “Orest im Medusengrab. Ein Versuch zum Betrachter,” Mitteilungen des Deutschen Archaeologischen Instituts Römische Abteilung, vol. 110, pp. 117–150. Bikai, P.M. 1992, “Classical Tyre” in The Heritage of Tyre: Essays on the History, Archaeology, and Preservation of Tyre, ed. M.S. Joukowsky, Kendall-Hunt Publishing Co., Dubuque, IA, pp. 61–68. Bikai, P.M. & Bikai, P. 1987, “Tyre at the end of the twentieth century,” Berytus, vol. 35, pp. 67–96. Bikai, P.M., Fulco, W.J., & Marchand, J. 1996, Tyre: The Shrine of Apollo, National Press, Amman.

BIBLIOGRAPHY

Bodel, J. 2000, “Dealing with the dead. Undertakers, executioners and potter’s fields in ancient Rome” in Death and Disease in the Ancient City, eds. V.M. Hope & E. Marshall, Routledge, London & New York, pp. 128–151. Bodel, J. (ed.) 2001, Epigraphic Evidence. Ancient History from Inscriptions, Routledge, London & New York. Boehmer, R.M., Pedde, F., & Salje, B. 1995, Ausgrabungen in Uruk-Warka: Endberichte. Bd. 10 Uruk, die Gräber, Verlag Philipp von Zabern, Mainz am Rhein. Bonatz, D. 2000, “Syro-Hittite funerary monuments. A phenomenon of tradition or innovation?” in Essays on Syria in the Iron Age, ed. G. Brunnens, Peeters Press, Louvain & Paris, pp. 189–210. Bopp, E. 2008, “Jmarrin – Die Siedlungsstruktur” in Hauran III. L’habitat dans les campagnes de Syrie du Sud aux époques classique et médiévale, ed. P. Clauss-Balty, IFPO, Damascus, pp. 169–182. Borg, B.E. 1998, “Der zierlichste Anblick der Welt … ” Ägyptische Porträtmumien, Verlag Philipp von Zabern, Mainz am Rhein. Borg, B.E. (ed.) 2004, Paideia: The World of the Second Sophistic, Walter de Gruyter, Berlin. Bounni, A. 1977, “Preliminary report on the archeological excavations at Tell al-’Abd and ’Anab al-Safinah (Euphrates) 1971–72” in Archeological Reports from the Tabqa Dam Project – Euphrates Valley, Syria, ed. D.N. Freedman, ASOR, Cambridge, pp. 49–61. Bounni, A. 1980, “Les tombes à tumuli du Moyen-Euphrate” in Le Moyen-Euphrate. Zone de contacts et d’échanges, ed. J.Cl. Margueron, Brill, Leiden, pp. 315–325. Bounni,A.1997,“Tell Kazel”in The Oxford Encyclopedia of Archaeology in the Near East, vol. 3, ed. E.M. Meyer, Oxford University Press, Oxford, pp. 275–276. Bounni, A. 2005, “Vierzig Jahre Syrische Ausgrabungen in Palmyra” in Palmyra. Kulturbegegnung im Grenzbereich, ed. A. Schmidt-Colinet, Philipp von Zabern, Mainz am Rhein, pp. 13–21.

341 Bounni, A. & Saliby, N. 1956, “Preliminary study of the cemetery at Naw¯a” [in Arabic], Les Annales Archéologiques de Syrie, vol. 6, pp. 8–37. Bowersock, G.W. 1994, Studies on the Eastern Roman Empire, Keip, Goldbach. Brody, L.R. & Hoffman, G.L. 2011, Dura Europos. Crossroads of Antiquity, McMullen Museum of Art, Boston, MA. Browning, I. 1979, Palmyra, Chatto & Windus, London. Buchet, L. 2012, “Europos-Doura, analyse des restes humains” in Europos-Doura. Varia I, eds. P. Leriche & G. Coqueugniot, Presses de l’ifpo, Beirut, pp. 191–200. Burns, R. 1999 [1992], Monuments of Syria. An Historical Guide, I.B. Tauris, London & New York. Butcher, K. 2003, Roman Syria and the Near East, The J. Paul Getty Museum, Los Angeles. Butcher,K.& Thorpe,R.1997,“A note on excavations in central Beirut 1994–96,” Journal of Roman Archaeology, vol. 10, pp. 291–306. Butler, H.C. 1903, Publications of an American Archaeological Expedition to Syria in 1899–1900. Part 2. Architecture and Other Arts. The Century Co., New York. Butler, H.C. 1908, Publications of the Princeton University Archaeological Expedition to Syria in 1904– 1905. Division II. Ancient Architecture in Syria. Section B. Northern Syria. Part 1, Brill, Leiden. Butler, H.C. 1914, Publications of the Princeton University Archaeological Expedition to Syria in 1904– 1905 and 1909. Division 2, Architecture, Section A, Southern Syria. Part 4. Bosra, Brill, Leiden. Butler, H.C. 1915, Publications of the Princeton University Archaeological Expedition to Syria in 1904– 1905 and 1909. Division 2, Architecture, Section A, Southern Syria. Part 5. Hauran Plain and Djebel Hauran, Brill, Leiden. Butler, H.C. 1916, Publications of the Princeton University Archaeological Expedition to Syria in 1904– 1905 and 1909. Division 2, Architecture, Section A, Southern Syria. Part 6. Si, Brill, Leiden. Butler, H.C. 1920, Publications of the Princeton University Archaeological Expedition to Syria in 1904– 1905 and 1909. Division 2, Architecture, Section B, Northern Syria, Brill, Leiden.

342

BIBLIOGRAPHY

Cagnat, R. 1925, “Nouvelles inscriptions de Syrie,” Comptes rendus de l’Académie des inscriptions et belles lettres, vol. 69, no. 3, pp. 150–153. Callot, O. 2007, “Bamuqqa, histoire d’un village banal,” Mélanges de l’université Saint Joseph, vol. 60, pp. 127–134. Callot, O. & Marcillet-Jaubert, J. 1984, “HautsLieux de Syrie du Nord” in Temples et sanctuaires, ed. G. Roux, Maison de l’Orient, Lyon, pp. 185–202. Cantineau, J. 1932, Inventaire des inscriptions de palmyre. Fasc. VIII Les dépots des antiquités, Publications du Musée National Syrien de Damas, Beirut. Carington Smith, J. 1984, “Pilate’s wife?” Antichthon, vol. 18, pp. 102–107. Carroll, M. 2013, “‘The insignia of women.’ Dress, gender and identity on the Roman funerary monument of Regina from Arbeia,” The Archaeological Journal, vol. 169, pp. 281–311. Carroll, M. & Wild, J.P. 2012, Dressing the Dead in Classical Antiquity, Amberley, Gloucestershire. Casana, J.J. 2007, “Structural transformations in settlement systems of the northern Levant,” American Journal of Archaeology, vol. 112, pp. 195–221. Chéhab, M.H. 1934a, “Sarcophages en plomb du musée national Libanais,” Syria, vol. 15, pp. 337–350. Chéhab, M.H. 1934b, “Trois stèles trouvées en Phénicie,” Berytus, vol. 1, pp. 44–46. Chéhab, M.H. 1935, “Sarcophages en plomb du musée national Libanais,” Syria, vol. 16, pp. 59–60. Chéhab, M.H. 1942–1943, “Chronique,” Bulletin du Musée de Beyrouth, vol. 4, p. 87. Chéhab, M.H. 1944–1945, “Chronique,” Bulletin du Musée de Beyrouth, vol. 7, pp. 109–120. Chéhab, M.H. 1955, “Chronique,” Bulletin du Musée de Beyrouth, vol. 12, pp. 47–58. Chéhab, M.H. 1968, “Sarcophages à relief de Tyr,” Bulletin du Musée de Beyrouth, vol. 21. Chéhab, M.H. 1970, Tyr: histoire, topographie, fouilles, Société d’Impression et d’Edition Libanaise, Beirut. Chéhab, M.H. 1983, “Fouilles de Tyr; La Nécropole, I: L’Arc de Triomphe,” Bulletin du Musée de Beyrouth, vol. 33.

Chéhab, M.H. 1984, “Fouilles de Tyr; La Nécropole, II: Descriptions des fouilles,” Bulletin du Musée de Beyrouth, vol. 34. Chéhab,M.H.1985,“Fouilles de Tyr;La Nécropole, III: Descriptions des fouilles,” Bulletin du Musée de Beyrouth, vol. 35. Chéhab,M.H.1986,“Fouilles de Tyr,La Nécropole, IV: Description des fouilles,” Bulletin du Musée de Beyrouth, vol. 36. Chéhadeh, K. & Griesheimer, M. 1998, “Les reliefs funéraires du tombeau du Prêtre Rapsônès (Babulin, Syrie du Nord),” Syria, vol. 75, pp. 171–192. Christensen, A.P. & Johansen, C.F. 1971, Hama. Fouilles et recherches de la fondation Carlsberg 1931–1938. Les poteries hellénistiques et les terres sigillées orientales, Nationalmuseet, Copenhagen. Christensen, A.P., Thomsen, R., & Ploug, G. (eds.) 1986, Hama. Fouilles et recherches 1931– 1938 III 3 The Graeco-Roman objects of clay, the coins and the Necropolis, Nationalmuseet, Copenhagen. Clauss, P. 2002, “Les tours funéraires du Djebel Baghouz dans l’histoire de la tour funéraire syrienne,” Syria, vol. 79, pp. 155–194. Collart, P. 1973, “La tour de Qalaat Fakra,” Syria, vol. 50, no. 1, pp. 137–161. Colledge, M.A.R. 1976, The Art of Palmyra, Thames and Hudson Ltd., London. Colledge, M.A.R. 1977, Parthian art, Cornell University Press, Ithaca, NY. Connerton, P. 1989, How Societies Remember, Cambridge University Press, Cambridge. Contenau, G. 1920, “Mission archéologique à Sidon (1914),” Syria, vol. 1, no. 2, pp. 108–154. Cormack, S. 1997a, “Funerary monuments and mortuary practice in Roman Asia Minor” in The Early Roman Empire in the East, ed. S.E. Alcock, Oxbow Books, Oxford, pp. 137–156. Cormack, S. 1997b, “Funerary monuments” in The Oxford Encyclopedia of Archaeology in the Near East vol. II, ed. E.M. Meyer, Oxford University Press, Oxford, pp. 347–350. Cormack, S. 2004, The Space of Death in Roman Asia Minor, Phoibos Verlag, Vienna. Coulston, J. 2007, “Art, culture, and service: the depiction of soldiers on funerary monuments

BIBLIOGRAPHY

of the 3rd c AD” in The Impact of the Roman Army (200 BC–AD 476): Economic, Social, Political, Religious, and Cultural Aspects: Proceedings of the Sixth Workshop of the International Network Impact of Empire (Roman Empire, 200 BC– AD 476), Capri, March 29–April 2, 2005, eds. O. Hekster, G.d. Kleijn, L.d. Blois, & E. Lo Cascio, Brill, Leiden, pp. 529–561. Cumont, F. 1929, “Un sarcophage d’enfant trouvé a Beyrouth,” Syria, vol. 10, pp. 217–219. Cumont, F. 1933, “Deux monuments des cultes solaires,” Syria, vol. 14.4, pp. 381–395. Curtis, J. 1976, “Parthian gold from Nineveh,” The British Museum Yearbook, pp. 47–66. Curtis, J. 1995, “Gold face-masks in the ancient Near East” in The Archaeology of Death in the Ancient Near East, eds. S. Campbell & A. Green, Oxbow Books, Oxford, pp. 226–231. Curvers, H.H. & Stuart, B. 1997, “The BCD Infrastructure Archaeology Project, 1995,” Bulletin d’Archéologie et d’Architecture Libanaises, vol. 2, pp. 167–205. Curvers, H.H. & Stuart, B. 1998–1999, “The BCD Archaeology Project,” Bulletin d’Archéologie et d’Architecture Libanaises, vol. 3, p. 14. Curvers, H.H. & Stuart, B. 2004, “Beirut Central District Archaeology Project 1994–2003” in A Decade of Archaeology and History in the Lebanon, ed. C. Doumet-Serhal, The Lebanese British Friends of the National Museum, Beirut, pp. 248–265. Curvers, H.H. & Stuart, B. 2005, “The BCD Archaeology Project 2000–2006,” Bulletin d’Archéologie et d’Architecture Libanaises, vol. 9, pp. 189–222. Cussini, E. 1995, “Transfer of property at Palmyra,” ARAM-Periodical, vol. 7, pp. 233–250. Cussini, E. 2005, “Beyond the spindle: investigating the role of Palmyrene women” in A Journey to Palmyra. Collected Essays to Remember Delbert R. Hillers, ed. E. Cussini, Brill, Leiden, pp. 26–39. Dahari, U. (ed.) 1998, Liban. L’autre rive, Flammarion, Paris. Dalley, S. 1993, “Anointing in Ancient Mesopotamia”in The Oil of Gladness.Anointing in the

343 Christian Tradition, eds. M. Dudley & G. Rowell, SPCK, London, pp. 19–34. Darrous, N. & Rohmer, J. 2004, “ChahbaPhilippopolis (Hauran): Essai de Synthèse Archéologique et Historique,” Syria, vol. 81, pp. 5–42. Daszewski, W.A. 1972, “Les Fouilles Polonaises à Palmyre en 1968 et 1969,” Les Annales Archéologiques de Syrie, vol. 22, pp. 129–150. Davie, M.F. 1987, “Maps and the historical topography of Beirut,” Berytus, vol. 35, pp. 141–164. de Jong, L. 2001, “Aspects of Roman burial practices in Beirut: excavations at BEY 022,” ARAM-Periodical, vol. 13, pp. 293–312. de Jong, L. 2007, Becoming a Roman Province: An Analysis of Funerary Practices in Roman Syria in the Context of Empire, unpublished dissertation, Stanford University. de Jong, L. 2010, “Performing death in Tyre: the life and afterlife of a Roman cemetery in the province of Syria,” American Journal of Archaeology, vol. 114, no. 4, pp. 597–630. de Jong, L. 2011, “Mallowan’s Marshes. The archaeology of the Balikh Valley under Roman and Byzantine rule,” Zeitschrift für Orient-Archäologie, vol. 4, pp. 262–282. de Jong, L. 2011 [2015], “The tombs of Roman Baalbek,” Bulletin d’Archéologie et d’Architecture Libanaises, vol. 15, pp. 239–256. de Jong, L. 2012, “Resettling the steppe: the archaeology of the Balikh Valley in the Early Islamic period” in Proceedings of the 7th International Congress on the Archaeology of the Ancient Near East. Volume 2, eds. R. Matthews & J. Curtis, Harrassowitz Verlag, Wiesbaden, pp. 517–531. de Jong, L. 2014, “Die Nekropolen von Baalbek” in Baalbek – Heliopolis. 10 000 Jahre Stadtgeschichte, eds. M. van Ess & K. Rheidt, Philipp von Zabern, Mainz, pp. 46–51. de Jong, L. in press, “Funerary traditions” in Blackwell Companion to the Hellenistic and Roman Near East, ed. T. Kaizer, WileyBlackwell, Malden, MA. de Jong, L. & Palermo, R. in press, “Living on the edge: the Roman Empire in the North Mesopotamian steppe” in The Archaeology of

344

BIBLIOGRAPHY

Empires – Investigating the Transformation of Rural and Peripheral Landscapes and Societies in Old World Empires, eds. B.S. Düring & T. Stek, Cambridge, Cambridge University Press. de Vogüé, M., Le Comte 1865–1877, Syrie centrale: architecture civile et religieuse du Ier au VIIe siècle, J. Baudry, Paris. Delhopital, N. 2010, “Étude archéoanthropologique de deux tombes de Suweida (Syria)” in Hauran V. La Syrie du Sud du Néolithique à l’Antiquité tardive, eds. M. alMaqdissi, F. Braemer, & J. Dentzer, IFPO, Beirut, pp. 395–409. Dentzer, J. (ed.) 1985, Hauran I: recherches archéologiques sur la Syrie du Sud à l’époque hellénistique et romaine. Volume 1, Paul Geuthner, Paris. Dentzer, J. (ed.) 1986, Hauran I: recherches archéologiques sur la Syrie du Sud à l’époque hellénistique et romaine. Volume 2, Paul Geuthner, Paris. Dentzer, J. 2003, “Le Jebel al-’Arab à l’époque: Repères chronologiques et Identité Culturelle” in Kulturkonflikte im Vorderen Orient an der Wende vom Hellenismus zur Römischen Kaiserzeit, eds. K.S. Freyberger, A. Henning, & H. von Hesberg, Verlag Marie Leidorf, Rahden/Westf., pp. 181–217. Dentzer, J. 2007, “Le développement urbain à Bosra” in Bosra. Aux portes de l’Arabie, eds. J. Dentzer-Feydy, M. Vallerin, T. Fournet, R. Mukdad, & A. Mukdad, IFPO, Beirut, pp. 13–15. Dentzer, J., Blanc, P., Fournet, T., Kalos, M., & Renel, F. 2010, “Formation et développement des villes en Syrie du Sud de l’époque Hellénistique à l’époque Byzantine: les exemples de Bosra, Suweida, Shabha” in Hauran V. La Syrie du Sud du Néolithique à l’Antiquité tardive, eds. M. al-Maqdissi, F. Braemer, & J. Dentzer, IFPO, Beirut, pp. 139–169. Dentzer-Feydy, J., Vallerin, M., Fournet, T., Mukdad, R., & Mukdad, A. 2007, Bosra. Aux portes de l’Arabie, IFPO, Beirut. Derks, T. 1998, Gods, Temples and Ritual Practices. The Transformation of Religious Ideas and Values of Gaul, Amsterdam University Press, Amsterdam.

Díaz-Andreu, M., Lucy, S., Babi´c, S., & Edwards, D.N. 2005, The Archaeology of Identity. Approaches to Gender, Age, Status, Ethnicity and Religion, Routledge, London & New York. Dietler, M. 2005, “The archaeology of colonialization and the colonization of archaeology. Theoretical challenges from an ancient Mediterranean colonial encounter” in The Archaeology of Colonial Encounters, ed. G.J. Stein, School of American Research Press, Santa Fe, NM, pp. 33–68. Dijkstra, K. 1995, Life and Loyalty: A Study in the Socio-Religious Culture of Syria and Mesopotamia in the Graeco-Roman Period based on Epigraphical Evidence, Brill, Leiden. Downey,S.B.1977,The Stone and Plaster Sculpture, Institute of Archaeology, University of California, Los Angeles. Downey, S.B. 1988, Mesopotamian Religious Architecture. Alexander through the Parthians, Princeton University Press, Princeton, NJ. Downey, S.B. 2000, “The transformation of Seleucid Dura-Europos” in Romanization and the City, ed. E. Fentress, Journal of Roman Archaeology Supplementary Series, Portsmouth, pp. 154–172. Drijvers, H.J.W. 1982, “After life in Palmyrene religion” in La soteriologia dei culti orientali nell’Impero romano, eds. U. Bianchi & M.J. Vermaseren, Brill, Leiden, pp. 709–733. du Mesnil du Buisson, R. 1924–1925, “Recherches archéologiques à Beyrouth,” Bulletin de la société française des fouilles archéologiques, vol. 6, pp. 81–134. Dudley, M. & Rowell, G. (eds.) 1993, The Oil of Gladness. Anointing in the Christian Tradition, SPCK, London. Dunand, M. 1953, “Recherches archéologiques dans la région de Marathus. Note préliminaire,” Les Annales Archéologiques de Syrie, vol. 3, pp. 165–170. Dunand, M. 1965, “Tombe peinte dans la campagne de Tyr,” Bulletin du Musée de Beyrouth, vol. 18, pp. 5–51. Dunand, M. 1969, “Rapport sur les fouilles de Sidon en 1967–1968,”Bulletin du Musée de Beyrouth, vol. 22, pp. 101–107.

BIBLIOGRAPHY

Dunand, M. & Duru, R. 1962, Oumm el-’Amed. Une ville de l’époque hellénistique aux échelles de Tyr, Librairie d’amérique et d’orient Adrien maisonnauve, Paris. Dunand, M. & Saliby, N. 1957, “A la recherche de Simyra,” Les Annales Archéologiques de Syrie, vol. 7, pp. 1–16. Dunand, M., Saliby, N., & Khirichian, A. 1954– 1955, “Les fouilles d’Amrith en 1954. Rapport préliminaire,” Les Annales Archéologiques de Syrie, vol. 4–5, pp. 189–204. Dunand, M., Bounni, A., & Saliby, N. 1964, “Fouilles de Tell Kazel, Rapport Préliminaire,” Les Annales Archéologiques de Syrie, vol. 14, pp. 1–14. Dussart, O. 1998, Le verre en Jordanie et en Syrie du Sud, IFAPO, Beirut. Dussaud, R. 1902, “Inscription Nabatéo-Arabe d’en-Nemȃra,” Revue Archéologique, vol. 41, pp. 409–421. Edwell, P.M. 2008, Between Rome and Persia: The Middle Euphrates, Mesopotamia and Palmyra under Roman Control, Routledge, London & New York. Egami, N., Masuda, S., & Iwasaki, T. 1979, Rumeilah and Mishrifat: Excavations of Hellenistic Sites in the Euphrates Basin, 1974–1978, The Ancient Orient Museum, Tokyo. Elayi,J.& Haykal,M.R.1996,Nouvelles découvertes sur les usages funéraires des Phéniciens d’Arwad, Gabalda, Paris. Fappas, I. 2011, “Exchange of ideas in the eastern Mediterranean during the 14th and 13th centuries BC: the case of perfumed oil use and ideology” in Intercultural Contacts in the Ancient Mediterranean, eds. K. Duistermaat & I. Regulski, Uitgeverij Peeters, Leuven, pp. 495–510. Fedak, J. 1990, Monumental Tombs of the Hellenistic Age, University of Toronto Press, Toronto. Fejfer, J. 2006, Roman Portraits in Context, Walter de Gruyter, Berlin. Fellmann, R. 1970, Le sanctuaire de Baalshamin à Palmyre V: Die Grabanlagen, Institut Suisse de Rome, Rome. Fick, S.M.E. 1999, “Gesichter aus gold, die den Glanz der Sonne widerspiegeln” in Vom Steinbruch zum Jupitertempel von Heliopolis/ Baalbek (Libanon), ed. E.M. Ruprechtsberger,

345 NORDICO-Museum der Stadt Linz, Linz, pp. 77–97. Finet, A. 1977, “Bilan provisoire des fouilles belges du Tell Kannâs” in Archeological Reports from the Tabqa Dam Project – Euphrates Valley, Syria, ed. D.N. Freedman, ASOR, Cambridge, pp. 79–95. Finlayson, C. 2002–2003, “Veil, turban, and headpiece: funerary portraits and female status at Palmyra,” Les Annales Archéologiques Arabes Syriennes, vol. 45–46, pp. 221–235. Foss, C. 1997, “Syria in transition, A.D. 550–750: an archaeological approach,” Dumbarton Oaks Papers, vol. 51, pp. 189–269. Freyberger, K.S. 1992, “Die Bauten und Bildwerke von Philippopolis,” Damaszener Mitteilungen, vol. 6, pp. 293–311. Freyberger, K.S. & Ragette, F. 1999, “Stadt des Jupiter Heliopolitanus” in Baalbek. Im Bann Römischer Monumentalarchitektur, eds. M. van Ess & T. Weber, Verlag Philipp von Zabern, Mainz am Rhein, pp. 45–67. Frier, B. 2000, “Demography” in The Cambridge Ancient History, eds. A.K. Bowman, P. Garnsey, & D. Rathbone, Cambridge University Press, Cambridge, pp. 787–816. Froment, L. 1930, “Carte touristique et archéologique du caza de H.a¯ rem,” Syria, vol. 11, no. 3, pp. 280–292. Gasche, H. 1991, “Tell ed-Der 1985–1987. Les Vestiges Meso-Babyloniens,” Northern Akkad Project Reports, vol. 6, pp. 9–92. Gasche, H. 1996, “Les tombes Achéménides Tardives et Seleucides de Tell ed-Der, Abu Qubur et Mahmudiyah,” Northern Akkad Project Reports, vol. 10, pp. 39–84. Gasche, H., Warburton, D.A., Warburton, R.C., Amandry, M., Van Zeist, W., & Gautier, A. 1989, “Abu Qubur 1987–1988, Chantier F. La Résidence Achéménide,” Northern Akkad Project Reports, vol. 4, pp. 3–43. Gasche, H., Pons, N., Verhoeven, K., & Warburton, D.A. 1991, “Fouilles d’Abu Qubur. Quatrième campagne (1990),” Northern Akkad Project Reports, vol. 7, pp. 3–74. Gatier, P. 1988, “Inscriptions latines et reliefs du Nord de la Syrie,” Syria, vol. 65, no. 1/2, pp. 217–229.

346

BIBLIOGRAPHY

Gatier, P. 1997–1998, “Un sarcophage de la nécropole de Restan,” Annales d’Histoire et d’Archéologie, vol. 8–9, pp. 35–47. Gatier, P. 2001, “Inscriptions funéraires grecques de Sidon,”Bulletin des musées et monuments lyonnais, pp. 15–23. Gatier,P.& Rousset,M.2010,“Temples Romains et mausolées de la Syrie Centrale (1)” in Entre nomades et sédentaires. Prospections en Syrie du Nord et en Jordanie du Sud, eds. P. Gatier & B. Geyer, Maison de l’Orient et de la Méditerranée, Lyon, pp. 147–167. Gawlikowski, M. 1970, Monuments funéraires de Palmyre, PWN Edition Scientifique de Pologne, Warsaw. Gawlikowski, M. 1972, “La notion de tombeau en Syrie romaine,” Berytus, vol. 21, pp. 5–15. Gawlikowski, M. 1974, “Les défenses de Palmyre,” Syria, vol. 51, no. 3/4, pp. 231–242. Gawlikowski, M. 1988, “La route de I’Euphrate d’Isidore à Julien” in Géographie historique au Proche-Orient (Syrie, Phénicie, Arabie, grecques, romaines, byzantines). Actes de la Table Ronde de Valbonne, 16–18 septembre 1985, ed. P.-L. Gatier, Éditions du CNRS, Paris, pp. 77–98. Gawlikowski, M. 1995, “Tempel, Gräber und Kasernen. Die polnischen Ausgrabungen im Diokletianslager” in Palmyra. Kulturbegegnung im Grenzbereich, ed. A. Schmidt-Colinet, Verlag Philipp von Zabern, Mainz am Rhein, pp. 21–27. Gawlikowski, M. 1998, “Deux publicains et leur tombeau,” Syria, vol. 75, pp. 145–151. Gawlikowski, M. 2005, “Tempel, Gräber und Kasernen. Die Polnischen Ausgrabungen im Diokletianslager” in Palmyra. Kulturbegegnung im Grenzbereich, ed. A. Schmidt-Colinet, Philipp von Zabern, Mainz am Rhein, pp. 22–28. Gee, R. 2008, “From corpse to ancestor: the role of tombside dining in the transformation of the body in ancient Rome” in The Materiality of Death: Bodies, Burials, Beliefs, eds. F. Fahlander & T. Oestigaard, Archaeopress, Oxford, pp. 59–68. George,M.2005,The Roman Family in the Empire: Rome, Italy, and Beyond, Oxford University Press, Oxford and New York.

Gessert, G. 2004, “Myth as consolatio: Medea on Roman sarcophagi,”Greece & Rome,vol.51,no. 2, pp. 217–249. Geyer, B. & Monchambert, J. 2003, La basse valleé de l’Euphrate Syrien. Du Néolithique à l’avénement de l’Islam, IFAPO, Beirut. Giddens, A. 1990, The Consequences of Modernity, Stanford University Press, Stanford. Gillespie, S.D. 2001, “Personhood, agency, and mortuary ritual: a case study from the ancient Maya,” Journal of Anthropological Archaeology, vol. 20, pp. 73–112. Gogräfe, R. 1995, “Die Grabtürme von Sirrin (Osroëne),” Damaszener Mitteilungen, vol. 8, pp. 165–201. Gogräfe, R. & Chehadé, J. 1999, “Die Waffen führenden Gräber aus Chisphin im Golan,” Journal of Roman Military Equipment Studies,vol. 10, pp. 73–80. Goldhill, S. 2001, Being Greek under Rome: Cultural Identity, the Second Sophistic and the Development of Empire, Cambridge University Press, Cambridge. Gowland, R. 2006, “Ageing the past: examining age identity from funerary evidence” in Social Archaeology of Funerary Remains, eds. R. Gowland & C. Knüsel, Oxbow Books, Oxford, pp. 143–154. Graf, D.F. 1992a, “The Syrian Hauran,” Journal of Roman Archaeology, vol. 5, pp. 450–466. Graf, D.F. 1992b, “Hellenisation and the decapolis,” ARAM-periodical, vol. 4, pp. 1–48. Grainger, J.D. 1990, The cities of Seleucid Syria, Oxford University Press, Oxford. Grainger, J.D. 1991, Hellenistic Phoenicia, Clarendon Press, Oxford. Gramsch, A. 2013, “Treating bodies: transformative and communicative practices” in The Oxford Handbook of the Archaeology of Death and Burial, eds. L. Nilsson Stutz & S. Tarlow, Oxford University Press, Oxford, pp. 459–474. Graziosi, G. 1968–1969, “Excavations in squares CLXXI, 54/55/64/65/66,” Mesopotamia, vol. 3–4, pp. 43–52. Green, D. 2008, “Sweet spices in the tomb: an initial study on the use of perfume in Jewish burials” in Commemorating the Dead. Texts and

BIBLIOGRAPHY

Artifacts in Context, eds. L. Brink & D. Green, Walter de Gruyter, Berlin, pp. 145–173. Griesheimer,M.1997a,“Cimetières et tombeaux des villages de la Syrie du Nord,” Syria, vol. 74, pp. 165–211. Griesheimer, M. 1997b, “Sociabilité et Rites Funéraires. Les porches à banquettes des maisons et des tombeaux du Massif Calcaire” in Les Maisons dans la Syrie Antique du IIIe millénaire aux débuts de l’Islam. Pratiques et représentations de l’espace domestique, eds. C. Castel, M. al-Maqdissi, & F. Villeneuve, IFAPO, Beirut, pp. 297–304. Gruen, E.S. 2011, Cultural Identity in the Ancient Mediterranean, Getty Research Institute, Los Angeles. Hajjar, J. 1965, “Un hypogée Romain a Deb’aal dans la Région de Tyr,” Bulletin du Musée de Beyrouth, vol. 18, pp. 61–104. Hakimian,S.1987,“Une archéologie parallèle:les découvertes clandestines et fortunes au Liban,” Berytus, vol. 35, pp. 199–209. Hamdy Bey, O. & Reinach, T. 1892, Une nécropole royale a Sidon. Fouilles de Hamdy Bey, Ernest Leroux, Paris. Harter-Uibopuu, K. 2014, “Tote soll man ruhen lassen … Verbote und Strafen zur Sicherung von Gräbern am Beispiel der Inschriften von Ephesos” in Der Beitrag Kleinasiens zur Kulturund Geistesgeschichte der griechisch-römischen Antike, ed. J. Fischer, Verlag der Österreichischen Akademie der Wissenschaften, Vienna, pp. 157–180. Hauser, S.R. 2012, “The Arsacid (Parthian) Empire” in A Companion to the Archaeology of the Ancient Near East Wiley-Blackwell, Malden, MA, pp. 1001–1020. Healey, J.F. 1995, “Death in West Semitic Texts: Ugarit and Nabataea” in The Archaeology of Death in the Ancient Near East, eds. S. Campbell & A. Green, Oxbow Books, Oxford, pp. 188–191. Healey, J.F. 2009, Aramaic Inscriptions and Documents of the Roman Period, Oxford University Press, Oxford. Heintz, F. 1998, “Circus curses and their archaeological contexts,” Journal of Roman Archaeology, vol. 11, pp. 337–342.

347 Heinz, M., Gross, A., John, E., Kirsch, L., Kulemann-Ossen, S., Van Lengerich, L., Leschke, C., Nieling, J., Von Rüden, C., & Wagner, E. 2004, “Kamid el-Loz in the Beqa’a plain/Lebanon. Excavations in 2001, 2002 and 2004,” Bulletin d’Archéologie et d’Architecture Libanaises, vol. 8, pp. 83–117. Henning, A. 2003, “Individueller Anspruch und gesellschaftliche Orientierung” in Kulturkonflikte im Vorderen Orient an der Wende vom Hellenismus zur Römischen Kaiserzeit, eds. K.S. Freyberger, A. Henning, & H. von Hesberg, Verlag Marie Leidorf, Rahden/Westf., pp. 95–108. Henning, A. 2013, Die Turmgräber von Palmyra, Verlag Marie Leidorf, Rahden/Westf. Hertz, R. 1907, “Contribution à une étude sur la répresentation collective de la mort,” Année sociologique, vol. 10, pp. 48–137. Heyn, M.K. 2010, “Gesture and identity in the funerary art of Palmyra,” American Journal of Archaeology, vol. 114, no. 4, pp. 631–661. Heyn, M.K. 2012, “Case study VI: female portraiture in Palmyra” in A Companion to Women in the Ancient World, Wiley-Blackwell, Malden, MA, pp. 439–441. Higuchi, T. & Izumi, T. 1994, Tombs A and C. Southeast Necropolis Palmyra, Syria, Research Center for Silk Roadology, Nara. Higuchi, T. & Saito, K. 2001, Tomb F – Tomb of BWLH and BWRP – Southeast Necropolis, Palmyra,Syria,Research Center for Silk Roadology, Nara. Hingley, R. 2005, Globalizing Roman Culture: Unity, Diversity and Empire, Routledge, London & New York. Hitchner, R.B. 2008, “Globalization avant la lettre: globalization and the history of the Roman Empire,” New Global Studies, vol. 2, no. 2, pp. 1–12. Hodder, I. 1986, Reading the Past. Current Approaches to Interpretation in Archaeology, Cambridge University Press, Cambridge. Hodder, I. 1995, Theory and Practice in Archaeology, Routledge, London & New York. Hodos, T. 2010, “Local and global perspectives in the study of social and cultural identities” in Material Culture and Social Identities in the

348

BIBLIOGRAPHY

Ancient World, eds. S. Hales & T. Hodos, Cambridge University Press, Cambridge, pp. 3–31. Hodos, T. 2014, “Global, local and in between: connectivity and the Mediterranean” in Globalisation and the Roman World: World History, Connectivity and Material Culture, eds. M. Pitts & M.J. Versluys, Cambridge University Press, Cambridge, pp. 240–254. Hoffmann, G. 1878, “Über eine am Euphrat gefundene Mumie mit goldener Gesichtsmaske,” Archäologische Zeitung, vol. 36, pp. 25–27. Hope, V.M. 1997, “Words and pictures: the interpretation of Romano-British tombstones,” Britannia, vol. 28, pp. 245–258. Hope, V.M. 2000, “Inscription and sculpture: the construction of identity in the military tombstones of Roman Mainz” in Epigraphy of Death: Studies in the History and Society of Greece and Rome, ed. G. Oliver, Liverpool University Press, Liverpool, pp. 155–185. Hope, V.M. 2001, Constructing Identity: The Roman Funerary Monuments of Aquileia. Mainz and Nîmes, IS 960 edn., Hadrian Books, Oxford. Hope, V.M. 2003a, “Trophies and tombstones: commemorating the Roman soldier,” World Archaeology, vol. 35, no. 1, pp. 79–97. Hope, V.M. 2003b, “Remembering Rome. Memory, funerary monuments and the Roman soldier” in Archaeologies of Remembrance, ed. H. Williams, Kluwer, New York, pp. 113–140. Hope, V.M. 2011, “Remembering to mourn: personal mementos of the dead in ancient Rome” in Memory and Mourning: Studies on Roman Death, eds. H.V. Huskinson & J. Huskinson, Oxbow Books, Oxford, pp. 176–195. Hopkins, C. 1972, Topography and Architecture of Seleucia on the Tigris, The University of Michigan, Ann Arbor, MI. Inda, J.X. & Rosaldo, R. 2008, The Anthropology of Globalization: A Reader, Blackwell, Malden, MA. Ingholt, H. 1934, Rapport préliminaire sur la première campagne des fouilles de Hama, Levin & Munksgaard, Ejnar Munksgaard, Copenhagen.

Ingholt, H. 1935, “Five dated tombs from Palmyra,” Berytus, vol. 2, pp. 57–120. Ingholt, H. 1938, “Inscriptions and sculptures from Palmyra,” Berytus, vol. 5, no. 2, pp. 93–140. Ingholt, H. 1940, Rapport préliminaire sur sept campagnes des fouilles a Hama en Syrie, Enjar Munksgaard, Copenhagen. Ingholt, H. 1942, “The Danish excavations at Hama on the Orontes,” American Journal of Archaeology, vol. XLVI, no. 4, pp. 469–476. Ingholt, H. 1966, “Some sculptures from the tomb of Malkû at Palmyra” in Mélanges offerts à Kazimierz Michalowski, eds. M.L. Bernhard, R.B. Bandinelli, P. Devambez, et al., PWN, Warsaw, pp. 457–476. Invernizzi, A. 1967, “The excavations of Tell ’Umayr,” Mesopotamia, vol. 2, pp. 9–32. Jackson, H. & Littleton, J. 2002, “Appendix: inventory of graves excavated in 1996 and 1997” in Jebel Khalid on the Euphrates. Report on Excavations 1986–1996, eds. G.W. Clarke, P.J. Connor, L. Crewe, et al., Mediterranean Archaeology Supplement, Sydney, pp. 71–100. James, S. 2001, “Soldiers and civilians: identity and interaction in Roman Britain” in Britons and Romans: Advancing an Archaeological Agenda, eds. S. James & M. Millett, CBA Research Report 125, York, pp. 187–209. James, S. 2007, “New light on the Roman military base at Dura-Europos: interim report on a pilot season of fieldwork in 2005” in The Late Roman Army in the Near East from Diocletian to the Arab Conquest: Proceedings of a Colloquium Held at Potenza, Acerenza and Matera, Italy/May 2005, eds. A. Lewin & P. Pellegrini, Archaeopress, Oxford, pp. 29–47. Jarry, J. 1970, “Inscriptions Arabes, Syriaques et Grecques du Massif du Bélus en Syrie du Nord,” Annales Islamologiques, vol. 9, pp. 187–212. Jarry,J.1982,“Nouvelles Inscriptions du Syrie du Nord,” Zeitschrift für Papyrologie und Epigraphik (ZPE), vol. 47, pp. 73–104. Jennings,S.2004–2005,“Vessel Glass from Beirut. Bey 006, 007, and 045,” Berytus, vol. 48–49. Jennings, J. 2011, Globalizations and the Ancient World, Cambridge University Press, Cambridge & New York.

BIBLIOGRAPHY

Jennings, S. & Abdallah, J. 2001, “Roman and later blown glass from the AUB excavations in Beirut (sites Bey 006, 007 and 045),” ARAMperiodical, vol. 12, pp. 237–264. Jidejian, N. 1971, Sidon through the Ages, Dar elMashreq Publishers, Beirut. Jidejian, N. 1993, Beyrouth à travers les ages, Dar el-Mashreq Publishers, Beirut. Jidejian,N.1996 Tyr à travers les ages,Librarie Orientale, Beirut. Jidejian, N. 1998 [1975], Baalbek – Heliopolis – “City of the Sun,” Dar el-Mashreq Publishers, Beirut. Jiménez, A. 2008, “A critical approach to the concept of resistance: new ‘traditional’ rituals and objects in funerary contexts of Roman Baetica” in TRAC 2007: Proceedings of the 17th Annual Theoretical Roman Archaeology Meetings, London 2007, eds. C. Fedwick, M. Wiggings, & D. Wythe, Oxbow, Oxford, pp. 15–30. Jones, S. 1997, The Archaeology of Ethnicity, Routledge, London & New York. Jones Hall, L. 2004, Roman Berytus. Beirut in Late Antiquity, Routledge, London & New York. Joyce, R.A. 2001, “Burying the dead at Tlatilco: social memory and social identities” in Social Memory, Identity, and Death: Anthropological Perspectives on Mortuary Rituals, ed. M.S. Chesson, American Anthropological Association, Arlington, VA, pp. 12–26. Kahwagi-Janho, H. 2007, “Les bains des Bleus de I’hippodrome de Tyr,” Bulletin d’Archéologie et d’Architecture Libanaises, vol. 11, pp. 25–56. Kaizer, T. 2002, The Religious Life of Palmyra, Franz Steiner Verlag, Stuttgart. Kaizer, T. 2010, “Funerary cults at Palmyra” in Cultural Messages in the Graeco-Roman World, eds. O. Hekster & S.T.A.M. Mols, BABESCH Suppl. 15 edn., Peeters, Leuven, pp. 23–31. Kalos, M. 1999, “Soada-Dionysias, une ville sous la ville,” Le Monde de la Bible, vol. 118, p. 73. Kaoukabani, B. 1971, “Les monuments de Wadi Cana,” Bulletin du Musée de Beyrouth, vol. 24, pp. 27–37. Kearns, E. 2010, Ancient Greek Religion: A Sourcebook, Wiley-Blackwell, Malden, MA. Keay, S. 1997, “Urban transformation and cultural change” in The Archaeology of Iberia. The

349 Dynamics of Change, eds. M. Díaz-Andreu & S. Keay, Routledge, London and New York, pp. 192–210. Keay, S. & Terrenato, N. (eds.) 2001, Italy and the West. Comparative Issues in Romanization, Oxbow Books, Oxford. Kennedy, D.L. 2007, Gerasa and the Decapolis. A “Virtual Island” in Northwest Jordan, Duckworth, London. Kennedy, D.L. & Freeman, P. 1995, “Southern Hauran survey,” Levant, vol. 27, pp. 39–73. Khalil, L.A. 2001, “Glass vessels and miniature jugs from Khirbat Yajuz Cemetery, Jordan,” Levant, vol. 33, pp. 127–138. King, G.R.D. 2002, “Archaeological fieldwork at the Citadel of Homs, Syria: 1995–1999,” Levant, vol. 34, pp. 39–58. King, G.R.D. 2002–2003, “The Homs Citadel excavations,” Les Annales Archéologiques Arabes Syriennes, vol. 45–46, pp. 415–424. Kleiner, D.E.E. 1987, Roman Imperial Funerary Altars with Portraits, Giorgio Bretschneider Editore, Rome. Kloner, A. & Zissu, B. 2006, The Necropolis of Jerusalem in the Second Temple Period, Peeters, Leuven. Koch, G. 1977, “Sarkophage im römischen Syrien,” Archäologischer Anzeiger, pp. 388–395. Koch, G. & Sichtermann, H. 1982, Römische Sarkophage, C.H. Beck’sche Verlagsbuchhandlung, Munich. Kondoleon, C. (ed.) 2000, Antioch: The Lost Ancient City, Princeton University Press, Princeton, NJ. Konrad, M. 2001, Resafa V: Der spätrömische Limes in Syrien: Archäologische Untersuchungen an den Grenzkastellen von Sura, Tetrapyrgium, Cholle und in Resafa, Philipp von Zabern, Mainz am Rhein. Konrad, M. 2003, “Römisches Militär in den Orientprovinzen – Defensivmaßnahme oder politisches Instrument?” in Kulturkonflikte im Vorderen Orient an der Wende vom Hellenismus zur Römischen Kaiserzeit, eds. K.S. Freyberger, A. Henning, & H. von Hesberg, Verlag Marie Leidorf, Rahden/Westf., pp. 237–249. Konrad, M. 2004, “Grabformen und Totenkult als Quellen zur Bevölkerungsstruktur im

350

BIBLIOGRAPHY

Römischen Syrien” in Lokale Identitäten in Randgebieten des Römischen Reiches, ed. A. Schmidt-Colinet, Phoibos Verlag, Vienna, pp. 133–146. Konrad, M. 2007, “Gräber und Bestattungssitten in den Römischen Provinzen Syria, Osrhoene und Mesopotamia. Einheimische und Fremde zwischen orientalischem und römischem Grabbrauch” in Körpergräber des 1.–3. Jahrhunderts in der römischen Welt: internationales Kolloquium, Frankfurt am Main, 19.– 20. November 2004, ed. A. Faber, Archäologisches Museum Frankfurt, Frankfurt am Main, pp. 77–97. Konrad, M. 2008, “Roman military fortifications along the eastern desert frontier: Settlement continuities and changes in North Syria, 4th– 8th centuries A.D.” in Residences, Castles, Settlements. Transformation Processes from Late Antiquity to Early Islam in Bilad al-Sham. Proceedings of the International Conference held at Damascus, 5–9 November 2006, eds. K. Bartl & A.A. Moaz, Verlag Marie Leidorf, Rahden/Westf., pp. 433–453. Kosmin, P.J. 2014, The Land of the Elephant Kings, Harvard University Press, Cambridge, MA. Kraeling, C.H. 1952, “Report on a sounding in the Jezireh,” Les Annales Archéologiques de Syrie, vol. 2, pp. 252–258. Krencker, D. & Zschietzschmann, W. 1938, Römische Tempel in Syrien, Walter de Gruyter, Berlin. Krencker, D., von Lüpke, T., & Winnefeld, H. 1923, Baalbek. Ergebnisse der Ausgrabungen und Untersuchungen in den Jahren 1898 bis 1905, zweiter band, W. de Gruyter, Berlin and Leipzig. Kropp, A.J.M. 2010, “Earrings, Nefesh and Opus Reticulatum: self-representation of the royal house of Emesa in the first century AD” in Kingdoms and Principalities in the Roman Near East, eds. T. Kaizer & M. Facella, Franz Steiner Verlag, Stuttgart, pp. 199–216. Kropp, A.J.M. 2013, Images & Monuments of Near Eastern Dynasts, 100 BC–AD 100, Oxford University Press, Oxford. Kropp, A.J.M. & Raja, R. 2014, “The Palmyra portrait project,” Syria, vol. 91, pp. 393–408.

Kühn, D. 2005, Totengedenken bei den Nabatäern und im Alten Testament, Ugarit-Verlag, Münster. Kühne, H. 2005, Magdalu/Magdala. Tall Š¯eh Hamad von der Postassyrischen zeit bis zur Römischen kaiserzeit, Dietrich Reimer Verlag, Berlin. Kuhrt, A. & Sherwin-White, S. (eds.) 1987, Hellenism in the East. The Interaction of Greek and Non-Greek Civilizations from Syria to Central Asia after Alexander, Duckworth, London. Kulemann-Ossen, S. & Martin, L. 2000, “Tall Kn¯ediˆg. A rural site in the Iron Age” in Essays on Syria in the Iron Age, ed. G. Brunnens, Peeters Press, Leuven, pp. 487–503. Kurtz, D.C. & Boardman, J. 1971, Greek Burial Customs, Cornell University Press, Ithaca, NY. Lauffray, J. 1944–1945, “Forums et monuments de Béryte,” Bulletin du Musée de Beyrouth, vol. 7–8, pp. 13–80. Lauffray, J. 1977, “Beyrouth Archéologie et Histoire, époques gréco-romaines. I. Période hellénistique et Haut-Empire romain” in Aufstieg und Niedergang der Römischen Welt, eds. H. Temporini & W. Haase, Walter de Gruyter, Berlin & New York, pp. 135–163. Lauffray, J. 1991, Halabiyya-Zenobia. Place forte du limes oriental et la haute-mésopotamie au IVe siècle. Tome II, Librairie Orientaliste Paul Geuthner, Paris. Le Lasseur, D. 1922a, “Mission archéologique à Tyr,” Syria, vol. 3, no. 1, pp. 1–26. Le Lasseur, D. 1922b, “Mission archéologique à Tyr,” Syria, vol. 3, no. 2, pp. 116–133. Lembke, K. 2001, Phönizische anthropoide Sarkophage, Philipp von Zabern, Mainz am Rhein. Lepetz, S. & Van Andringa, W. 2011, “Publius Vesonius Phileros vivos monumentum fecit. Investigations in a sector of the Porta Nocera cemetery in Roman Pompeii” in Living through the dead. Burial and commemoration in the Classical World, eds. M. Carroll & J. Rempel, Oxbow Books, Oxford, pp. 110–133. Leriche, P. 2001, “Doura Europos – Bilan de la campagne 1998,” Les Annales Archéologiques Arabes Syriennes, vol. 44, pp. 107–130. Leriche, P. 2002, “The Greeks in the Orient: from Syria to Bactria” in The Greeks beyond the

BIBLIOGRAPHY

Aegean: From Marseilles to Bactria, ed. V. Karageorghis, Alexander S. Onassis Public Benefit Foundation (USA), New York, pp. 78–103. Leriche, P. 2003, “Europos-Doura Hellénistique” in La Syrie Hellénistique Topoi, Lyon, pp. 171–191. Leriche, P. 2012, “Europos-Doura. Quinze années de travaux de la mission FrancoSyrienne (1986–2001)” in Europos-Doura. Varia I, eds. P. Leriche & G. Coqueugniot, Presses de l’ifpo, Beirut, pp. 11–46. Leriche,P.& Gelin,M.(eds.) 1997,Doura-Europos. Etudes IV 1991–1993, IFAPO, Beirut. Leriche, P., Al-Mahmoud, A., Mouton, B., Alabe, F., Saliou, C., Allara, A., Matheson, S.B., & Downey, S.B. (eds.) 1992, Doura-Europos. Etudes 1990, Librairie Orientalist Paul Geuthner, Paris. Levison, J.R. 2002, “The Roman character of funerals in the writings of Josephus,” Journal for the Study of Judaism, vol. 33, no. 3, pp. 245–277. Lightfoot, J.L., 2003, Lucian. On the Syrian Goddess, Oxford University Press, Oxford. Linant de Bellefonds, P. 1985, Sarcophages Attiques de la Nécropole de Tyr. Une étude iconographique, Éditions Recherche sur les Civilisations, Paris. Lindsay, H. 2000, “Death-pollution and funerals in the city of Rome” in Death and Disease in the Ancient City, eds. V.M. Hope & E. Marshall, Routledge, London & New York, pp. 152–173. Lipi´nski, E. 2004, Itineraria Phoenicia, Uitgeverij Peeters en Departement Oosterse Studies, Leuven. Littleton, J. & Frohlich, B. 2002, “Excavations of the cemetery – 1996 and 1997” in Jebel Khalid on the Euphrates. Report on Excavations 1986– 1996,eds.G.W.Clarke,P.J.Connor,L.Crewe,et al., Mediterranean Archaeology Supplement, Sydney, pp. 49–70. Littleton, J., Frohlich, B., & Clarke, G.W. 1996– 1997, “Preliminary excavation of the Jebel Khalid necropolis,” Mediterranean Archaeology, vol. 9–10, pp. 187–205. Littmann, E. 1914, Publications of the Princeton University Archaeological Expedition to Syria in 1904– 1905 and 1909. Div. IV. Section A. Nabataean

351 Inscriptions from the Southern Haurân. Brill, Leiden. Lönnqvist, M., Lönnqvist, K., Whiting, M.S., Törmä, M., Nunez, M., & Okkonen, J. 2005, “Documenting, identifying and protecting a late Roman-Byzantine fort at Tabus on the Euphrates, International Cooperation to Save the World’s Cultural Heritage, Proceedings of the XX International Symposium CIPA 2005, Turin (Italy) 26 September – 01 October 2005, vol. 1” in The International Archives of Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol. XXXVI-5/C34, CIPA, ed. S. Dequal, The ICOMOS/ISPRS Committee for Documentation of Cultural Heritage, pp. 427–432. Luck, G. 2006, Arcana Mundi: Magic and the Occult in the Greek and Roman Worlds: A Collection of Ancient Texts, Johns Hopkins University Press, Baltimore, MD. Lund, J. 1993, “The archaeological evidence for the transition from the Persian period to the Hellenistic period in Northwestern Syria,” Transeuphratène, vol. 6, pp. 13–45. Lund, J. 2003, “Hama in the Early Hellenistic period. A review of the archaeological evidence,” Topoi, suppl. 4, pp. 253–268. Macdonald, M.C.A. 2006, “Burial between the desert and the sown: cave-tombs and inscriptions near Dayr al-Kahf in Jordan,” Damaszener Mitteilungen, vol. 15, pp. 273–301. MacDonald, W.L. 1986, The Architecture of the Roman Empire. Volume II. An Urban Appraisal, Yale University Press, New Haven, CT & London. Machowski, C. 1983, “Recherches sur Ie tombeau de A’ailamî et Zebîdâ,” Damaszener Mitteilungen, vol. 1, pp. 175–187. MacMullen, R. 1982, “The epigraphic habit in the Roman Empire,”American Journal of Philology, vol. 103, pp. 233–246. Macridy, T. 1904, “A travers les nécropoles sidoniennes,” Revue Biblique, vol. 13, pp. 547–572. Masséglia, J. 2013, “Phrygians in relief: trends in self-representation” in Roman Phrygia Culture and Society, ed. P. Thonemann, Cambridge University Press, Cambridge, pp. 95–123.

352

BIBLIOGRAPHY

Matheson, S.B. 1992, “The tenth season at Dura-Europos 1936–1938,” Syria, vol. 69, pp. 121–140. Mattern, J. 1944, A traves les villes mortes de Haute Syrie, Imprimerie Catholique, Beirut. Mattingly, D.J. 2004, “Being Roman: expressing identity in a provincial setting,” Journal of Roman Archaeology, vol. 17, pp. 5–25. Mayence, F. 1938, “La VIe campagne de fouilles a Apamée,” Bulletin des musées royaux d’art et d’histoire, vol. 10, pp. 98–113. McCane, B.R. 2003, Roll Back the Stone: Death and Burial in the World of Jesus, Trinity Press International, Harrisburg, PA. McKenzie, J. 1990, The Architecture of Petra, published for the British Academy, the British School of Archaeology in Jerusalem, and the British Institute at Amman for Archaeology and History by Oxford University Press, Oxford & New York. Mendel, G. 1912, Catalogue des Sculptures: Grecques,Romaines et Byzantines,Musées Impériaux Ottomans, Istanbul. Meurdac, M. & Albanèse, A. 1938, “A travers les nécropoles gréco-romaines de Sidon,” Bulletin du Musée de Beyrouth, vol. 2, pp. 73–98. Meyer, E.A. 1990, “Explaining the epigraphic habit in the Roman Empire: the evidence of epitaphs,” Journal of Roman Studies, vol. 80, pp. 74–96. Michalowski, K. 1960, Palmyre. Fouilles Polonaises 1959, PWN, Warsaw. Michalowski, K. 1961, “Fouilles Polonaises a Palmyre 1961,” Les Annales Archéologiques de Syrie, vol. 10, p. 63. Michalowski, K. 1963, Palmyre. Fouilles Polonaises 1961, PWN, Warsaw. Milik, J.T. 1958, “Nouvelles inscriptions nabatéennes,” Syria, vol. 35, no. 3/4, pp. 227–251. Millar, F. 1987, “The Problem of Hellenistic Syria” in Hellenism in the East. The Interaction of Greek and Non-Greek Civilizations from Syria to Central Asia after Alexander, eds. A. Kuhrt & S. Sherwin-White, Duckworth, London, pp. 110–133. Millar, F. 1990, “The Roman coloniae of the Near East: a study of cultural relations”

in Roman Eastern Policy and Other Studies in Roman History, eds. H. Solin & M. Kajava, Finnish Society of Sciences and Letters, Helsinki, pp. 7–58. Millar, F. 1993, The Roman Near East 31 BC–AD 337, Cambridge University Press, Cambridge. Millar, F. 1998, “Dura Europos under Parthian rule”in Das Partherreich und seine Zeugnisse.The Arsacid Empire; Sources and Documentation, ed. J. Wiesehöfer, Franz Steiner Verlag, Stuttgart, pp. 406–431. Millett, M. 1990, The Romanization of Britain: An Essay in Archaeological Interpretation,Cambridge University Press, Cambridge. Morehart, M. 1956–1957, “Early sculpture at Palmyra,” Berytus, vol. 12, no. 1, pp. 53–87. Morris, I. 1992, Death-Ritual and Social Structure in Classical Antiquity, Cambridge University Press, Cambridge. Moussli, M. 1983, “Griechische Inschriften aus Emesa und Laodicea ad Libanum,” Philologus, vol. 127, pp. 254–261. Mouterde,R.1929,“Sarcophages de plomb trouvés en Syrie,” Syria, vol. 10, pp. 238–251. Mouterde, R. 1949, “A travers l’Apamène,” Mélanges de l’université Saint-Joseph, vol. 28, no. 3–42. Mouterde, R. 1951, “Antiquités de l’Hermon et de la Beqâ’,” Mélanges de l’université SaintJoseph, vol. 29, pp. 21–89. Mouterde, R. 1964, “Régards sur Beyrouth Phénicienne, Hellénistique et Romaine,” Mélanges de l’université Saint-Joseph, vol. 40, pp. 149–190. Mouterde, R. & Poidebard, A. 1945, Les Limes de Chalcis. Organisation de la Steppe en Haute Syrie Romaine, Paul Gauthner, Paris. Mouton, M. 1997, “Les tours funéraires d’Arabie, nefesh monumentales,” Syria, vol. 74, pp. 81–98. Myers, E.A. 2010, The Ituraeans and the Roman Near East: Reassessing the Sources, Cambridge University Press, Cambridge, New York. Nasrallah, J. 1950, “Tumulus de l’Age du Bronze dans le Hauran,” Syria, vol. 27, no. 3/4, pp. 314–331. Nederveen Pieterse, J. 2014, “Ancient Rome and globalisation: decentring Rome” in

BIBLIOGRAPHY

Globalisation and the Roman World: World History, Connectivity and Material Culture, eds. M. Pitts & M.J. Versluys, Cambridge University Press, Cambridge, pp. 225–239. Negro Ponzi, M. 1970–1971, “Excavations in squares CLXXI, 54/55/63/64/74,” Mesopotamia, vol. 5–6, pp. 31–39. Negro Ponzi, M. 1972, “The excavations in the Agora,” Mesopotamia, vol. 7, pp. 17–25. Negro Ponzi, M. 2002, “The glassware from Seleucia (Central Iraq),” Parthica, vol. 4, pp. 63–156. Nehmé, L. 2007, “La langue et l’écriture des Nabatéens” in Bosra. Aux portes de l’Arabie, eds. J. Dentzer-Feydy, M. Vallerin, T. Fournet, R. Mukdad, & A. Mukdad, IFPO, Beirut, pp. 16–18. Nehmé, L. 2010, “Les inscriptions nabatéenes du Hawran” in Hauran V. La Syrie du Sud du Néolithique à l’Antiquité tardive, eds. M. alMaqdissi, F. Braemer, & J. Dentzer, IFPO, Beirut, pp. 451–492. Newson, P., Abdulkarim, M., McPhillips, S., Mills, P., Reynolds, P., & Philip, G. 2008– 2009, “Landscape study of Dar es-Salaam and the basalt region north-west of Homs, Syria. Report on work undertaken during 2005– 2007,” Berytus, vol. 51–52, pp. 9–35. Ng, D.Y. 2015, “Asia Minor” in The Oxford Handbook of Roman Sculpture, eds. E.A. Friedland, M.G. Sobocinski, & E.K. Gazda, Oxford University Press, New York, pp. 538–551. Nilsson Stutz, L. & Tarlow, S. 2013, “Beautiful things and bones of desire: emerging issues in the archaeology of death and burial” in The Oxford Handbook of the Archaeology of Death and Burial, eds. L. Nilsson Stutz & S. Tarlow, Oxford University Press, Oxford, pp. 1–11. Nordiguian, L. 2004, Le voyage archéologique en Syrie et au Liban de Michel Jullien et Paul Soulerin en 1888, Presses de l’Université Saint-Joseph, Beyrouth. Novák, M. (ed.) 2000, Der parthisch-römische Friedhof von Tall Seh Hamad/Magdala. Teil I, Dietrich Reimer Verlag, Berlin. Nunn, A. 2001, “Nekropolen und Gräber in Phönizien, Syrien und Jordanien zur

353 Achämenidenzeit,” Ugarit-Forschungen, vol. 32, pp. 389–463. Oates, D. & Oates, J. 1958, “Nimrud 1957: the Hellenistic settlement,” Iraq, vol. 20, pp. 114–157. Oenbrink, W. 2000, “Qanawat. Untersuchungen zu den Nekropolen und Grabbauten. Ein Vorbericht,” Damaszener Mitteilungen, vol. 12, pp. 231–243. Oenbrink, W. 2003, “Späthellenistischfrühkaiserzeitliche Rundgräber der Nekropolen Qanawats” in Kulturkonflikte im Vorderen Orient an der Wende vom Hellenismus zur Römischen Kaiserzeit, eds. K.S. Freyberger, A. Henning, & H. von Hesberg, Verlag Marie Leidorf, Rahden/Westf., pp. 75–93. Oenbrink, W. 2006, “Späthellenistischfrühkaiserzeitliche Rundgräber in den Nekropolen von el-Qanawat/Kanatha. Eine Leitform der indigen-südsyrischen Grabarchitektur des Haur¯an,” Zeitschrift des deutschen Palästina-Vereins, vol. 122, pp. 61–83. Oenbrink, W. 2010, “Integration und Representation städlicher und ländlicher Eliten am Beispiel der Grabarchitektur Südsyriens: Die Ausgrabungen in den Nekropolen von alQanawat” in Hauran V. La Syrie du Sud du Néolithique à l’Antiquité tardive, eds. M. alMaqdissi, F. Braemer, & J. Dentzer, IFPO, Beirut, pp. 369–382. Oettel, A. 2000, “Charonspfennig und Totenglöckchen,” Altorientalische Forschungen, vol. 27, no. 1, pp. 106–120. O’Hea, M. 1993, “Glass from areas XXXIV and XXIX (Hellenistic-Abbasid)” in The Eleventh and Twelfth Seasons of Excavations at Pella (Tabaqat Fahl) 1989–1990, eds. A. Walmsley, P. Macumber, P. Edwards, S. Bourke, & P. Watson, Annual of the Department of Antiquities of Jordan (ADAJ), Amman, pp. 221–227. Olszewksi, M. 2001, “Le langage symbolique dans la décoration à scènes mythologiques et son sens dans les tombes peintes de l’Orient Romain. Nouvelle approche” in La peinture funéraire antique: IVe siècle av. J.-C.–IVe siècle ap. J.-C.: actes du VIIe Colloque de l’Association internationale pour la peinture murale antique

354

BIBLIOGRAPHY

(AIPMA), 6–10 octobre 1998, Saint-Romain-enGal, Vienne, ed. A. Barbet, Editions Errance, Paris, pp. 155–162. Parker, R. 1996 [1983], Miasma. Pollution and Purification in Early Greek Religion, Oxford, Oxford University Press. Parker Pearson, M. 2008 [1999], The Archaeology of Death and Burial, Texas A&M University, College Station, TX. Parlasca, K. 1967, “Zur Syrischen Kunst der frühen Kaiserzeit,” Archäologischer Anzeiger, pp. 547–568. Parlasca, K. 1976, “Probleme Palmyrenischer Grabreliefs. Chronologie und interpretation” in Palmyre, bilan et perspectives: colloque de Strasbourg (18–20 octobre 1973) AECR, Strasbourg, pp. 33–43. Parlasca, K. 1982, Syrische Grabreliefs hellenistischer und römischer Zeit. Fundgruppe und Probleme, Verlag Philipp von Zabern, Mainz am Rhein. Parlasca, K. 2006, “Die Skulpturen aus dem Habb˘asi-Grab im Hama,” Damaszener Mitteilungen, vol. 15, pp. 219–225. Patterson, J.R. 1992, “Patronage, collegia and burial in Imperial Rome” in Death in Towns: Urban Responses to the Dying and the Dead 100– 1600, ed. S. Basset, Leicester University Press, Leicester, pp. 15–27. Pedde, F. 1995, “Parthische Gräber: Friedhof oder Hausbestattungen?” in Beiträge zur Kulturgeschichte Vorderasiens. Festschrift für Rainer Michael Boehmer, eds. R. Dittmann, U. Finkbeiner, & H. Hauptmann, Verlag Philipp von Zabern, Mainz am Rhein, pp. 537–544. Peña, I., Castellana, P., & Fernández, R. 1987, Inventaire du Jébel Baricha. Recherches archéologique dans la région des Villes Mortes de la Syrie du Nord, Franciscan Printing Press, Jerusalem. Peña, I., Castellana, P., & Fernández, R. 1990, Inventaire du Jébel el-A’la. Recherches archéologique dans la région des Villes Mortes de la Syrie du Nord, Franciscan Printing Press, Milan. Peña, I., Castellana, P., & Fernández, R. 1999, Inventaire du Jébel Wastani. Recherches archéologique dans la région des Villes Mortes de

la Syrie du Nord, Franciscan Printing Press, Milan. Perdrizet, P. 1938, “Le monument de Hermel,” Syria, vol. 19, pp. 47–71. Pérez-Arantegui, J., Paz-Peralta, J., & OrtizPalomar, E. 1996, “Analysis of the products contained in two Roman glass unguentaria from the colony of Celsa (Spain),” Journal of Archaeological Science (JAS), vol. 23, pp. 649–655. Perring, D. 2001, “Beirut in antiquity: some research directions suggested by recent excavations in the souks,” ARAM-Periodical, vol. 13, pp. 129–140. Petersen, L. 2003, Die römische Nekropole Douris/Baalbek, unpublished MA thesis. Pfister, R. 1934, Textiles de Palmyre découverts par le Service des antiquités du Haut-commissariat de la République Française dans la nécropole de Palmyre, Les éditions d’art et d’histoire, Paris. Pfister, R. 1937, Nouveaux textiles de Palmyre découverts par le Service des antiquités du Hautcommissariat de la République Française dans la nécropole de Palmyre, Les éditions d’art et d’histoire, Paris. Pfister, R. 1940, Textiles de Palmyre découverts par le Service des antiquités du Haut-commissariat de la République Française dans la nécropole de Palmyre III, Les éditions d’art et d’histoire, Paris. Pfister, R. & Bellinger, L. 1945, The excavations at Dura-Europos. Final report IV, part II, Yale University Press, New Haven, CT. Philip, G., Abdulkarim, M., Newson, P., Beck, A., Bridgland, D., Bshesh, M., Shaw, A., Westaway, R., & Wilkinson, K. 2005, “Settlement and landscape development in the Homs region, Syria. Report on work undertaken during 2001–2003,” Levant, vol. 37, pp. 21–42. Pitts, M. 2008, “Globalizing the local in Roman Britain: an anthropological approach to social change,” Journal of Anthropological Archaeology, vol. 27, no. 4, pp. 493–506. Pitts, M. & Versluys, M.J. (eds.) 2014, Globalisation and the Roman World: World History, Connectivity and Material Culture, Cambridge University Press, Cambridge. Ploug, G. 1985, Hama. Fouilles et recherches de la fondation Carlsberg 1931–1938 III. The GraecoRoman Town, Nationalmuseet, Copenhagen.

BIBLIOGRAPHY

Poidebard, A. 1939, “Tyr. Un grand port disparu. Recherches aériennes et sous-marines 1934– 1936,” Bibliothèque archéologique et historique, vol. 29, pp. 1–78. Pollard, N. 1996, “The Roman army as ‘total institution’ in the Near East? Dura-Europos as a case study” in The Roman army in the East, ed. D.L. Kennedy, Journal of Roman Archaeology Supplementary Series, Portsmouth, pp. 211–227. Porter, A. 2002, “The dynamics of death. Ancestors, pastoralism and the origins of a third millennium city in Syria. Bulletin of the American Schools of Oriental Research,” Bulletin of the American Schools of Oriental Research (BASOR), vol. 325, pp. 1–36. Prentice, W.K. 1908a, Publications of an American Archaeological Expedition to Syria in 1899–1900. Part 3.Greek and Latin inscriptions,The Century Co., New York. Prentice, W.K. 1908b, Publications of the Princeton University Archaeological Expedition to Syria in 1904–1905. Division III. Greek and Latin Inscriptions. Section B Northern Syria. Part 1. The ’Ala and Kasr Ibn Wardan, Brill, Leiden. Prentice, W.K. 1909, Publications of the Princeton University Archaeological Expedition to Syria in 1904–1905. Division III. Greek and Latin Inscriptions. Section B Northern Syria. Part 3. The Djebel Riha and the Djebel Wastaneh, Brill, Leiden. Purcell, N. 1987, “Tomb and suburb” in Römische Gräberstrassen, Selbstdarstellung, Status, Standard, eds. H. von Hesberg & P. Zanker, Verlag der Bayerischen Akademie der Wissenschaften, Munich, pp. 25–41. Rahmani, L.Y. 1999, A Catalogue of Roman and Byzantine Lead Coffins from Israel, Israel Antiquities Authority, Jerusalem. Rahmo, R. 1993, Les momies de Palmyre, Imprimeries SIDAWY, Damascus. Renan, E. 1864–1874, Mission de Phénicie, Imprimerie impériale, Paris. Renel, F. 2010, “Tumuli, simpula et banquet funéraires à Suweida: un témoignage sur l’hellénisation des élites au 1er siècle en Syrie du Sud” in Hauran V. La Syrie du Sud du Néolithique à l’Antiquité tardive, eds. M. al-

355 Maqdissi, F. Braemer, & J. Dentzer, IFPO, Beirut, pp. 383–394. Renfrew, C. 1976, “Megaliths, territories and populations” in Acculturation and Continuity in Atlantic Europe, ed. S.J. De Laet, De Tempel, Bruges, pp. 198–220. Reuther, O. 1926, Die Innenstadt von Babylon, J.C. Hinrichs, Leipzig. Revell, L. 2009, Roman Imperialism and Local Identity, Cambridge University Press, Cambridge. Rey-Coquais, J. 1977, “Inscriptions de la nécropole,” Bulletin du Musée de Beyrouth, vol. 29. Rey-Coquais, J. 1998, “Deux stèles inscrites de Syrie du Nord,” Syria, vol. 75, pp. 193–200. Rey-Coquais, J. 2006, Inscriptions Grecques et Latines de Tyr – BAAL Hors-série III, Ministère de la Culture, Direction Générale des Antiquités, Beirut. Ribechini, E., Modugno, F., Colombini, M.P., & Evershed, R.P. 2008, “Gas chromatographic and mass spectrometric investigations of organic residues from Roman glass unguentaria,”Journal of Chromatography A,vol.1183,no. 1–2, pp. 158–169. Riis, P.J. 1960, “L’activité de la mission archéologique danoise sur la côte phenicienne en 1959,” Les Annales Archéologiques de Syrie, vol. 10, pp. 111–132. Riis, P.J. & Buhl, M. 2007, Hama Fouilles et Recherches 1931–1938 I 2. Bronze Age Graves in Ham¯a and its Neighbourhood, Nationalmuseet, Copenhagen. Riis, P.J., Thuesen, I., Lund, J., & Riis, T. 2004, ˘ Topographical Studies in the Gabla Plain, Det Kongelige Danske Videnskabernes Selskab, commission agent: C.A. Reitzel, Copenhagen. Riggs, C. 2005, The Beautiful Burial in Roman Egypt: Art, Identity, and Funerary Religion, Oxford University Press, Oxford, New York. Robb, J. 2007, “Burial treatment as transformations of bodily ideology” in Performing Death. Social Analysis of Funerary Traditions in the Ancient Near East and Mediterranean, ed. N. Laneri, The Oriental Institute of the University of Chicago, Chicago, IL, pp. 287–297. Robertson, R. 1992, Globalization: Social Theory and Global Culture, Sage, London.

356

BIBLIOGRAPHY

Röllig, W. 2004, “Semitische Inschriften auf Grabdenkmälern Syriens und der Levante. Formale und inhaltliche Aspekte” in Sepulkralund Votivdenkmäler östlicher Mittelmeergebiete (7. Jh. v. Chr. – 1. Jh. n. Chr), eds. R. Bol & D. Kreikenbom, Bibliopolis, Möhnesee, pp. 23–32. Rostovzteff, M.I. 1937, “Rapport sur les fouilles de Doura-Europos, campagne de 1936–37,” Comptes rendus de l’Académie des inscriptions et belles lettres, vol. 2, pp. 195–204. Rothe, U. 2013, “Whose fashion? Men, women and Roman culture as reflected in dress in the cities of the Roman north-west” in Women and the Roman city in the Latin West, eds. E.A. Hemelrijk & G. Woolf, Brill, Leiden, pp. 243–268. Roymans, N. 1996, “The sword or the plough. Regional dynamics in the Romanisation of Belgic Gaul and the Rhineland area” in From the Sword to the Plough. Three Studies on the Earliest Romanisation of Northern Gaul, ed. N. Roymans, Amsterdam University Press, Amsterdam, pp. 9–126. Rüpke, J. 2007, Religion of the Romans, Polity, Cambridge. Ruprechtsberger, E.M. 1999, Vom Steinbruch zum Jupitertempel von Heliopolis/Baalbek (Libanon), NORDICO-Museum der Stadt Linz, Linz. Saadé, G. 1985, “Découverte de caveaux funéraires à Jablé,” Syria, vol. 62, pp. 346–348. Sachet,I.2010,“Libations funéraire aux frontières de l’Orient romain: le cas de la Nabatène” in Bestattungsrituale und Totenkult in der römischen Kaiserzeit = Rites funéraires et culte des morts aux temps impériales, eds. J. Rüpke & J. Scheid, Franz Steiner Verlag, Stuttgart, pp. 157–174. Sader, H. 1991, “Phoenician stelae from Tyre,” Berytus, vol. 39, pp. 101–126. Sader, H. 2005, Iron Age Funerary Stelae from Lebanon, Bellaterra, Barcelona. Sadurska, A. 1976, “Nouvelles recherches dans la nécropole ouest de Palmyre” in Palmyre: Bilan et Perspectives: Colloque de Strasbourg (18–20 Octobre 1973) AECR, Strasbourg, pp. 11–32. Sadurska, A. & Bounni, A. 1994, Les sculptures funéraires de Palmyre, Giorgio Bretschneider, Rome.

Saghieh-Beydoun, M. 2005, “Urban planning in a seaport city. Beirut for Hellenistic to Byzantine period” in La Mobilité des Paysages Portuaires Antiques du Liban, eds. C. Morhange & M. Saghieh-Beydoun, Ministère de la Culture Direction Générale des Antiquités, Beirut, pp. 147–183. Saghieh-Beydoun, M., ’Allam, M., ’Ala ’Eddine, A., & Abdulhosn, S. 1998–1999, “Bey 004. The monumental street “Cardo Maximus” and the replanning of Roman Berytus,” Bulletin d’Archéologie et d’Architecture Libanaises, vol. 3, pp. 95–126. Saidah, R. 1969, “Archaeology in the Lebanon 1968–1969,” Berytus, vol. 18, pp. 119–142. Saito, K. 1995, “Excavation at Southeast Necropolis in Palmyra from 1990 to 1995,” ARAMperiodical, vol. 7, pp. 19–28. Saito, K. 2005a, “Die Arbeiten der Japanischen Mission in der Südost-Nekropole” in Palmyra. Kulturbegegnung im Grenzbereich, ed. A. Schmidt-Colinet, Philipp von Zabern, Mainz am Rhein, pp. 32–35. Saito, K. (ed.) 2005b, The Study of Palmyrene Funerary Practices and Social Backgrounds in Palmyra, Archaeological Institute of Kashihara, Nara. Salamé-Sarkis, H. 1986, “La Nécropole de Tyr: a propos de publications récentes,” Berytus, vol. 34, pp. 193–205. Salamé-Sarkis, H. 1987, “Heliopolitana Monumenta,” Berytus, vol. 35, pp. 126–139. Saliby, N. 1989, “’Amrit” in Archéologie et histoire de la Syrie. La Syrie de l’époque achéménide à l’avènement de l’Islam, eds. J.-M. Dentzer & W. Orthmann,Saarbrücker Druckerei und Verlag, Saarbrücken, pp. 19–30. Saliby, N. 1992, “L’hypogée de Sassan fils de Malê à Palmyre,” Damaszener Mitteilungen, pp. 267–289. Saliby, N. 1993, “Die Katakomben von Emesa/Homs (Hims)” in Syrien: von den Aposteln zu den Kalifen, ed. E.M. Ruprechtsberg, Stadtmuseum Nordico, Linz, pp. 265–273. Saliou, C. & Dandrau, A. 1997, “Données nouvelles sur les quartiers sud-est de DouraEuropos” in Doura – Europos. Etudes IV 1991–1993, ed. P. Leriche, IFAPO, Beirut, pp. 95–106.

BIBLIOGRAPHY

Saller, R.P. & Shaw, B.D. 1984, “Tombstones and Roman family relations in the principate: civilians, soldiers and slaves,” The Journal of Roman Studies, vol. 74, pp. 124–156. Salles, J. 1980, La Nécropole “K” de Byblos, Editions ADPF, Maison de L’Orient, Lyon. Sartre, M. 2005, The Middle East under Rome, The Belknap Press of Harvard University Press, Cambridge & London. Sartre, M. 2007, “Le cadre historique” in Bosra. Aux portes de l’Arabie, eds. J. Dentzer-Feydy, M. Vallerin, T. Fournet, R. Mukdad, & A. Mukdad, IFPO, Beirut, pp. 9–12. Sartre-Fauriat, A. 2001, Des Tombeaux et des Morts. Monuments funéraires, société et culture en syrie du sud du Ier s. av J.-C. au VIIe S. apr. J.-C. (Volume I–II), IFAPO, Beirut. Sartre-Fauriat, A. 2002–2003, “Architecture et épigraphy funéraires du Hauran du 1er s. av. J.-C. au VIIe s. ap. J.-C.,” A.A.A.S., vol. 45–46, pp. 323–329. Sartre-Fauriat, A. 2007a, “La nécropole de Tell Aswad” in Bosra. Aux portes de l’Arabie, eds. J. Dentzer-Feydy, M. Vallerin, T. Fournet, R. Mukdad, & A. Mukdad, IFPO, Beirut, pp. 208–210. Sartre-Fauriat, A. 2007b, “Les nécropoles et la mort à Bosra” in Bosra. Aux portes de l’Arabie, eds. J. Dentzer-Feydy, M. Vallerin, T. Fournet, R. Mukdad, & A. Mukdad, IFPO, Beirut, pp. 329–333. Scheid, J. 1984, “Contraria facere: renversements et déplacements dans les rites funéraires,” nnali del Seminario di Studi del Mondo Classico. Istituto Universitario Orientale Napoli: Sezione di Archeologia e Storia Antica, vol. 6, pp. 117–139. Scheid, J. 2007, “Sacrifices for gods and ancestors” in A Companion to Roman Religion, ed. J. Rüpke, Blackwell, Malden, MA, pp. 263–271. Schlumberger, D. 1951, La Palmyrène du NordOuest, Librairie Orientaliste Paul Geuthner, Paris. Schlumberger, D. 1960, “Descendants nonméditerranéens de l’art grec,” Syria, vol. 37, no. 3/4, pp. 253–318. Schmidt-Colinet, A. 1992, Das Tempelgrab Nr. 36 in Palmyra. Studien zur Palmyrenischen

357 Grabarchitektur und ihrer Ausstattung, Verlag Philipp von Zabern, Mainz am Rhein. Schmidt-Colinet, A. 1997, “Aspects of ‘Romanization’: the tomb architecture at Palmyra and its decoration” in The Early Roman Empire in the East, ed. S.E. Alcock, Oxbow books, Oxford, pp. 157–177. Schmidt-Colinet, A. 2004, “Palmyrenische Grabkunst als Ausdruck lokaler Identität(en)” in Lokale Identitäten in Randgebieten des Römischen Reiches, ed. A. Schmidt-Colinet, Phoibos Verlag, Vienna, pp. 189–198. Schmidt-Colinet, A. & al-As’ad, K. 2000, “Zur Urbanistik des hellenistischen Palmyra. Ein Vorbericht,” Damaszener Mitteilungen, vol. 12, pp. 61–93. Schmidt-Colinet, A. & al-As’ad, K. 2002, “Archaeological news from Hellenistic Palmyra,” Parthica, vol. 4, pp. 157–166. Schmidt-Colinet, A. & al-As’ad, K. 2003, “Nouvelles données sur Palmyre Hellénistique” in La Syrie Hellénistique Topoi, Lyon, pp. 299–302. Schmidt-Colinet, A., Stauffer, A., & Al-As’ad, K. 2000, Die Textilien aus Palmyra, Philipp von Zabern, Mainz am Rhein. Schmidt-Colinet, A., Al-As’ad, K., & Al-As’ad, W. 2008, “Untersuchungen im Areal der ‘hellenistischen’ Stadt von Palmyra. Zweiter Vorbericht,” Zeitschrift für Orient-Archäologie, vol. 1, pp. 452–478. Schnädelbach, K. 2010, Topographia Palmyrena, DAI, Damascus. Schwartz, G.M., Curvers, H.H., Dunham, S., & Stuart, B. 2003, “A third millennium B.C. elite tomb and other new evidence from Tell Umm el-Marra, Syria,” American Journal of Archaeology, vol. 107, pp. 325–361. Scott, S. & Webster, J. (eds.) 2003, Roman Imperialism and Provincial Art, Cambridge University Press, Cambridge. Seeden, H. 1991, “A tophet in Tyre?,” Berytus, vol. 39, pp. 39–82. Seeden, H. & Thorpe, R. 1997–1998, “Beirut from Ottoman landfills and sea walls to a twelfth century BC burial. Report on the archaeological excavations of the Souks northern area (BEY 007),” Berytus, vol. 43, no. 221–254.

358

BIBLIOGRAPHY

Segal, A. 1997, From function to monument. Urban landscapes of Roman Palestine, Syria and Provincia Arabia, Oxbow, Oxford. Seyrig, H. 1939, “Le cimetière des marins a Séleucie de Pièrie” in Mélanges Syriens offerts à Monsieur René Dussaud, Librairie Orientaliste Paul Geuthner, Paris, pp. 451–459. Seyrig, H. 1940, “Antiquités syriennes,” Syria, vol. 21, no. 2, pp. 113–122. Seyrig, H. 1952a, “Antiquités Syriennes. Antiquités de la nécropole d’Émèse I,” Syria, vol. 29, pp. 204–250. Seyrig, H. 1952b, “Le casque d’Emèse,” Les Annales Archéologiques de Syrie, vol. 2, pp. 101–108. Seyrig, H. 1953, “Antiquités Syriennes. Antiquités de la nécropole d’Émèse II,” Syria, vol. 30, pp. 12–24. Seyrig, H. 1958, “Inscriptions Grecques” in Villages Antiques de la Syrie du Nord. III, ed. G. Tchalenko, Librairie Orientaliste Paul Geuthner, Paris, pp. 2–62. Seyrig,H.1964,“Monnaies hellénistiques,”Revue numismatique, vol. 6, pp. 7–67. Shapiro, A. 1997, “Petrographic analysis of Roman clay sarcophagi from northwestern Israel and Cyprus,” ’Atiqot, vol. 33, pp. 1–5. Shaw, B.D. 1983, “Soldiers and society: the army in Numidia,” Opus, vol. 2, pp. 133–59. Sherwin-White, S. & Kuhrt, A. 1993, From Samarkhand to Sardis: A New Approach to the Seleucid Empire, Duckworth, London. Shibasaki, R., Izumi, T., Tanaka, H., Nakagawa, M., Iwamoto, Y., Fujiwara, H., & Manandhar, D. 2006, “A laser-scanner system for acquiring archaeological data: case of the Tyre remains” in GIS-Based Studies in the Humanities and Social Sciences, ed. A. Okabe, CRC Taylor and Francis, Boca Raton & London & New York, pp. 35–54. Skupi´nska-Løvset, I. 1983, Funerary Portraiture of Roman Palestine: An Analysis of the Production in ˚ its Culture-Historical Context, Aström, Gothenburg. Skupi´nska-Løvset, I. 1999, Portraiture in Roman Syria, Wydawnictwo Uniwersytetu Łódzkiego, Łód´z.

Smith,A.M.I.2013,Roman Palmyra:Identity,Community, and State Formation, Oxford University Press, Oxford. Smith, R.R.R. & Ertu˘g, A. 2001, Sculptured for Eternity. Treasures of Hellenistic, Roman and Byzantine Art from Istanbul Archaeological Museum, Ertu˘g & Kocabiyik, Istanbul. Smith,R.W.1992,“Secondary use of the necropoleis of the Decapolis,” ARAM-periodical, vol. 4, pp. 215–228. Sofaer, J.R. 2006, “Gender, bioarchaeology and human ontogeny” in Social Archaeology of Funerary Remains, eds. R. Gowland & C. Knüsel, Oxbow Books, Oxford, pp. 153–167. Sommer, M. 2005, Roms orientalische Steppengrenze, Franz Steiner Verlag, Stuttgart. Stauffer, A. 2012, “Dressing the dead in Palmyra in the second and third centuries AD” in Dressing the Dead in Classical Antiquity, eds. M. Carroll & J.P. Wild, Amberley, Gloucestershire, pp. 89–98. Steinsapir, A.I. 2005, Rural Sanctuaries in Roman Syria: The Creation of a Sacred Landscape, Hadrian Books, Oxford. Stern, E. 1982, Material Culture of the Land of the Bible in the Persian period, 538–332 B.C. Aris & Phillips; Israel Exploration Society, Warminster & Jerusalem. Stevens, S.T. 1991, “Charon’s obol and other coins in ancient funerary practice,” Phoenix, vol. 45, no. 3, pp. 215–229. Strube, C. 1996, Die ‘Toten Städte’. Stadt und Land in Nordsyrien während der Spätantike, Verlag Philipp von Zabern, Mainz am Rhein. Stuart, B. 2001, “Cemeteries in Beirut,” ARAMperiodical, vol. 13, pp. 87–112. Sweetman, R.J. 2007, “Roman Knossos: the nature of a globalized city,” American Journal of Archaeology, vol. 111, no. 1, pp. 61–81. Tabaczek, M. 2004, “Die Säulenstrassen des Vorderen Orients. Die Enstehung eines Monumenttypus im Spannungsfeld zweier Kulturen” in Lokale Identitäten in Randgebieten des Römischen Reiches, ed. A. Schmidt-Colinet, Phoibos Verlag, Vienna, pp. 209–221. Takase, Y., Sasaki, Y., Nakagawa, M., Shimizu, M., Yamada, O., Izumi, T., & Shibasaki, R. 2003, “Reconstruction with laser scanning

BIBLIOGRAPHY

and 3D visualization of Roman monuments and remains in Tyre, Lebanon,” The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, vol. 34, no. 5/W 12, pp. 325–329. Tarlow, S. 2013, “Belief and the archaeology of death” in The Oxford Handbook of the Archaeology of Death and Burial, eds. L. Nilsson Stutz & S. Tarlow, Oxford University Press, Oxford, pp. 617–630. Tate,G.1992,Les Campagnes de la Syrie du Nord du IIe au VIIe siècle: un exemple d’expansion démographique et économique dans les campagnes à la fin de l’antiquité. Tome 1, Paul Geuthner, Paris. Tate, G. 1997, “The Syrian countryside during the Roman era” in The Early Roman Empire in the East, ed. S.E. Alcock, Oxbow books, Oxford, pp. 55–71. Tchalenko, G. 1953, Villages antiques de la Syrie du Nord; le massif du Bilus à l’époque romaine. Vol. I + II, Paul Geuthner, Paris. Tchalenko, G. 1958, Villages antiques de la Syrie du Nord; le massif du Bilus à l’époque romaine. Vol. III, Paul Geuthner, Paris. Teixidor, J. 1979, The Pantheon of Palmyra, Brill, Leiden. Theodossiev, N. 1998, “The dead with golden faces: Dasaretian, Pelagonian, Mygdonian and Boeotian funerary masks,” Oxford Journal of Archaeology, vol. 17, no. 3, pp. 345–367. Thissen, L. 1988, “The burials” in Hammam et – Turkman I. Report on the University of Amsterdam’s 1981–84 Excavations in Syria, ed. M.N. van Loon, Nederlands Instituut voor het Nabije Oosten/Netherlands Institute for the Near East (NINO), Leiden, pp. 143–179. Thomas, N. 1991, Entangled Objects. Exchange, Material Culture and Colonialism in the Pacific, Harvard University Press, Cambridge. Thorpe, R. 1998–1999, “Bey 045, preliminary report on the excavations,” Bulletin d’Archéologie et d’Architecture Libanaises, vol. 3, pp. 57–83. Toll, N.P. 1937, “The necropolis of HalebieZenobia,” Annales de l’institut Kondakov, vol. 9, pp. 11–22. Toll, N.P. 1946, Excavations at Dura-Europos Preliminary Report of the Ninth Season of Work

359 1935–1936 II. The Necropolis, Yale University Press, New Haven, CT. Toynbee, J.M.C. 1971, Death and Burial in the Roman World, Thames and Hudson, London. Ucko, P.J. 1969, “Ethnography and archaeological interpretation of funerary remains,” World Archaeology, vol. 1, pp. 262–280. Ur, J.A. 2010, Urbanism and Cultural Landscapes in Northeastern Syria: The Tell Hamoukar Survey, 1999–2001, Oriental Institute of the University of Chicago, Chicago, IL. Valtz, E. 1986, “Trench on the east side of the Archive Square (Seleucia, 12th season),” Mesopotamia, vol. 21, pp. 11–20. Valtz, E. 1988, “Trench on the east side of the Archive Square (Seleucia, 13th season),” Mesopotamia, vol. 23, pp. 19–30. Van Andringa, W., Duday, H., & Lepetz, S. 2013, Mourir à Pompéi. Fouille d’un quartier funéraire de la nécropole romaine de Porta Nocera (2003–2007), École Française de Rome, Rome. van Dommelen, P.A.R. 1997, “Colonial constructs: colonialism and archaeology in the Mediterranean,” World Archaeology, vol. 28, pp. 305–323. van Dommelen, P.A.R. 1998, On Colonial Grounds: A Comparative Study of Colonialism and Rural Settlement in First Millennium BC West Central Sardinia, Faculty of Archaeology, University of Leiden, Leiden. van Dommelen, P.A.R. 2005, “Colonial interaction and hybrid practices. Phoenician and Carthaginian settlement in the ancient Mediterranean” in The Archaeology of Colonial Encounters, ed. G.J. Stein, School of American Research Press, Santa Fe, NM, pp. 109–141. van Ess, M. 1998, Heliopolis Baalbek, 1898–1998. Forschungen in Ruinen, DAI/DGA, Beirut. van Ess, M. & Petersen, L. 2003, “Excavation of a Late Roman necropolis in Baalbek – Douris,” Bulletin d’Archéologie et d’Architecture Libanaises, vol. 7, pp. 83–107. van Ess, M. & Rheidt, K. (eds.) 2014, Baalbek – Heliopolis. 10 000 Jahre Stadtgeschichte, Philipp von Zabern, Mainz am Rhein. van Ess, M. & Weber, T. (eds.) 1999, Baalbek. Im Bann Römischer Monumentalarchitektur, Verlag Philipp von Zabern, Mainz am Rhein.

360

BIBLIOGRAPHY

van Ess, M., Gelin, M., Hakimian, S., & Rifai, T. 1999, “Graben, dokumentieren und restaurieren. Deutsche, französische und libanesische Archäologen in Baalbek” in Baalbek. Im Bann Römischer Monumentalarchitektur, eds. M. van Ess & T. Weber, Verlag Philipp von Zabern, Mainz am Rhein, pp. 14–40. van Gennep, A. 1960 [1908], The Rites of Passage, University of Chicago Press, Chicago, IL. van Loon, M.N. (ed.) 2001, Selenkahiye. Final Report on the University of Chicago and University of Amsterdam Excavations in the Tabqa Reservoir,Northern Syria,1967–1975,Nederlands Historisch-Archaeologisch Instituut te Istanbul, Istanbul. van Loon, M.N. & Meijer, D.J.W. 1983, “Hammâm et-Turkmân on the Balikh: first results,” Les Annales Archéologiques Arabes Syriennes, vol. 33, no. 1, pp. 131–152. van Nijf, O.M. 1997, The Civic World of Professional Associations in the Roman East, J.C. Gieben, Amsterdam. van Nijf, O.M. 2010, “Being Termessian: local knowledge and identity politics in a Pisidian city” in Local Knowledge and Microidentities in the Roman East, ed. T. Whitmarsh, Cambridge University Press, Cambridge, pp. 163–188. van Rengen, W. 1969, “Inscriptions Grecques et Latines” in Apamée de Syrie. Bilan des recherches archéologiques 1965–1968, ed. J. Balty, Centre Belge de recherches archéologiques à Apamée de Syrie, Brussels. van Rengen, W. 1972, “Nouvelles Inscriptions Grecques et Latines” in Apamée de Syrie. Bilan des recherches archéologiques 1969–1971, eds. J.C. Balty & J. Balty, Centre Belge de recherches archéologiques à Apamée de Syrie, Brussels, pp. 97–112. Vandenabeele, F. 1972, “Sondages dans la Nécropole Nord d’Apamée, 1969–1970” in Apamée de Syrie. Bilan des recherches archéologiques 1969–1971, eds. J. Balty & J. Balty, Centre Belge de recherches archéologiques à Apamée de Syrie, Brussels, pp. 85–92. Venit, M.S. 2002, Monumental Tombs of Ancient Alexandria. The Theater of the Dead, Cambridge University Press, Cambridge.

Verhoeven, M. 2011, “The many dimensions of ritual” in The Oxford Handbook of the Archaeology of Ritual and Religion, ed. T. Insoll, Oxford University Press, Oxford, pp. 115–132. Viviers, D. & Vokaer, A. 2007, “Travaux de la Mission archéologique belge à Apamée de Syrie XLe campagne (2006),” Revue Belge de philologie et d’histoire, vol. 85, no. 1, pp. 125–156. von Hackwitz, K. & Lindholm, K. 2015, “Landscapes of mortuary practices” in Ancient Death Ways. Proceedings of the Workshop on Archaeology and Mortuary Practices Uppsala, 16–17 May 2013, eds. K. Von Hackwitz & R. Peyroteo-Stjerna, Uppsala University, Uppsala, pp. 143–166. von Hesberg, H. & Zanker, P. (eds.) 1987, Römische Gräberstrassen, Selbstdarstellung, Status, Standard, Verlag der Bayerischen Akademie der Wissenschaften, Munich. Wagner,J.(ed.) 2000,Gottkönige am Euphrat.Neue Ausgrabungen und Forschungen in Kommagene, Philipp von Zabern, Mainz am Rhein. Walbank, M.E.H. 2002, “Unquiet graves: burial practices of the Roman Corinthians” in Urban Religion in Roman Corinth. Interdisciplinary Approaches, eds. D.N. Schowalter & S.J. Friesen, Harvard University Press, Cambridge, pp. 249–280. Walker, S.E.C. 1997, “Mummy portraits in their Roman context” in Portraits and Masks: Burial Customs in Roman Egypt, ed. M.L. Bierbrier, published for the Trustees of the British Museum by British Museum Press, London, pp. 1–6. Warburton, D.A. 1989, “Stratigraphy and structures of the ‘Résidence Achéménide’,” Northern Akkad Project Reports, vol. 4, pp. 9–24. Ward-Perkins, J.B. 1958, “Four Roman garland sarcophagi in America,” Archaeology, vol. 11, pp. 98–104. Ward-Perkins, J.B. 1969, “The imported sarcophagi of Roman Tyre,” Bulletin du Musée de Beyrouth, vol. 22, pp. 109–145. Waterman, L. (ed.) 1931, Preliminary Report upon the Excavations at Tell Umar, Iraq, University of Michigan Press, Ann Arbor, MI. Watzinger, C. 1923, “Das Grabmal des Samsigeramos von Emesa,” Kunsthistoriske Sällskapets Publikation, pp. 18–43.

BIBLIOGRAPHY

Watzinger, C. 1932, “Zur Geschichte des Grabturms” in Palmyra, Ergebnisse der Expeditionen von 1902 und 1917, ed. T. Wiegand, Heinrich Keller Verlag, Berlin, pp. 77–84. Weber, T. 2006, Sculptures from Roman Syria in the Syrian National Museum at Damascus, Wernersche Verlagsgesellschaft, Worms. Webster, J. 1997, “A negotiated syncretism: reading on the development of Romano-Celtic religion” in Dialogues in Roman Imperialism. Power, Discourse, and Discrepant Experience in the Roman Empire, ed. D.J. Mattingly, Journal of Roman Archaeology Supplementary Series, Portsmouth, pp. 165–184. Webster, J. 2001, “Creolizing the Roman provinces,” American Journal of Archaeology, vol. 105.2, pp. 209–225. Weir, R.G.A. 2001, “Antiochene grave stelai in Princeton” in Roman Sculpture in the Art Museum, ed. M. Padgett, Princeton University Press, Princeton, NJ, pp. 274–310. Welcomme, E., Walter, P., Elslande, E.V., & Tsoucaris, G. 2006, “Investigation of white pigments used as make-up during the GrecoRoman period,” Journal Applied Physics A: Materials Science & Processing, vol. 83, no. 4, pp. 551–556. Wenning, R. 2001, “Hauranite sculpture” in Roman Sculpture in the Art Museum, ed. M. Padgett, Princeton University Press, Princeton, NJ, pp. 312–347. Whittaker, C.R. 1997, “Imperialism and culture: the Roman initiative” in Dialogues in Roman Imperialism. Power, Discourse, and Discrepant Experience in the Roman Empire, ed. D.J. Mattingly, Journal of Roman Archaeology Supplementary Series, Portsmouth, pp. 143– 163. Wiegand, T. 1921, Baalbek. Ergebnisse der Ausgrabungen und Untersuchungen in den Jahren 1898 bis 1905. Erster Band, Vereinigung Wissenschaftlicher Verleger. Walter de Gruyter & Co., Berlin & Leipzig. Wiesehöfer,J.(ed.) 1998,Das Partherreich und seine Zeugnisse. The Arsacid Empire; Sources and Documentation, Franz Steiner Verlag, Stuttgart. Wilburn, A.T. 2012, Materia Magica: The Archaeology of Magic in Roman Egypt, Cyprus, and Spain,

361 University of Michigan Press, Ann Arbor, MI. Wilkinson, T.J. 1997–1998, Tell Beydar Survey. 1997–98 Annual Report, http://oi.uchicago .edu/sites/oi.uchicago.edu/files/uploads/ shared/docs/ar/91-00/97-98/97-98_Beydar .pdf . Wilkinson, T.J. 2003, Archaeological Landscapes of the Near East, The University of Arizona Press, Tucson, AZ. Wilkinson, T.J. & Rayne, L. 2010, “Hydraulic landscapes and imperial power in the Near East,” Water History, vol. 2, no. 2, pp. 115–144. Will, E. 1946–1948, “Un sarcophage romain à sujet eschatologique au Musée de Beyrouth,” Bulletin du Musée de Beyrouth, vol. 8, pp. 109–127. Will, E. 1949a, “La tour funéraire de Palmyre,” Syria, vol. 26, pp. 87–116. Will, E. 1949b, “La tour funéraire de la Syrie et les monuments apparentés,” Syria, vol. 26, pp. 258–312. Will, E. 1965, “La Syrie romaine entre l’Occident gréco-romain et l’Orient parthe” in Le rayonnement des civilisation grecque et romaine sur les cultures périphériques Édition E. de Boccard, Rome, pp. 511–526. Williams, H. 2013, “Death, memory, and material culture: catalytic commemoration and the cremated dead” in The Oxford Handbook of the Archaeology of Death and Burial, eds. L. Nilsson Stutz & S.Tarlow,Oxford University Press, Oxford, pp. 1–11. Witecka, A. 1994, “Catalogue of jewellery found in the tower-tomb of Atenatan in Palmyra” in Studia Palmyrenskie IX, eds. M. Gawlikowski & S.P. Kowalski, Polish Center of Mediterranean Archaeology, Warsaw, pp. 71–91. Woolf, G. 1996, “Monumental writing and the expansion of Roman society in the early empire,” Journal of Roman Studies, vol. 86, pp. 22–39. Woolf, G. 1998, Becoming Roman. The Origins of Provincial Civilization in Gaul, Cambridge University Press, Cambridge. Woolf, G. 2000, “Urbanization and its discontents in early Roman Gaul” in Romanization and the City. Creation, Transformation and

362

BIBLIOGRAPHY

Failures, ed. E. Fentress, Journal of Roman Archaeology Supplementary Series, Portsmouth, pp. 115–131. Woolf, G. 2002, “Generations of aristocracy. Continuities and discontinuities in the societies of Interior Gaul,” Archaeological Dialogues, vol. 9.1, pp. 2–15. Yeivin, S. 1933, “The Tombs found at Seleucia (seasons 1929–30 and 1931–32)” in Second Preliminary Report upon the Excavations at Tell Umar, Iraq, ed. L. Waterman, University of Michigan Press, Ann Arbor, MI. Zanker, P. 2000, “The city as symbol: Rome and the creation of an urban image” in Romanization and the City. Creation, Transformation and Failures, ed. E. Fentress, Journal of

Roman Archaeology Supplementary Series, Portsmouth, pp. 25–41. Zanker, P. & Ewald, B.C. 2004, Mit Mythen leben. Die Bilderwelt der römischen Sarkophage, Hirmer, Munich. Zaqzûq, A. 2001, “The Roman and later graves and their contents” in Selenkahiye. Final Report on the University of Chicago and University of Amsterdam Excavations in the Tabqa Reservoir, Northern Syria, 1967–1975, ed. M.N. van Loon, Nederlands Historisch-Archaeologisch Instituut, Istanbul, pp. 13.515–13.565. Zuiderhoek, A. 2009, The Politics of Munificence in the Roman Empire: Citizens, Elites and Benefactors in Asia Minor, Cambridge University Press, Cambridge & New York.

INDEX

age, 112, 124–126, 127–128, 165, see also children Antioch, 7, 131, 156, 333 Apamea burial practices, 225–228 Hellenistic period, 7 history, 225 soldiers, 30 tomb types, 57 army commemoration, 129, 132–138, 141 funerary ritual, 156 soldiers, 30, 167, 214 veterans, 137 Baalbek burial practices, 229–232 gender, 121 history, 10, 228 Beirut burial practices, 233–237 communal burial, 116 grave goods, 92 Hellenistic period, 7, 61 history, 10, 233 redevelopment, 35 body treatment, 110–111, 149–150, 157–158 borders, 10, 11–13 Bosra burial practices, 238–241 history, 7, 12, 237 location cemeteries, 21 Brad (Limestone Plateau), 25, see also Limestone Plateau burial differentiation, 163–168 burial rights, 143 Byzantine period. See Late Roman period cenotaphs, 158–160 children, 166–167 cist-graves, 41, 314 cities civic identity, 190–191 location cemeteries, 28, 176–177 redevelopment, 9, 35, 176–177, 188–190 urbanization, 10, 189–190 coffins, 52–55, 326–329, see also sarcophagi

coins, 89 commemoration, 14–15, 171–174 communal burial commemoration, 112–119 differentiation, 167 funerary ritual, 101 grave goods, 94–95, 116 layout, 60 consolation, 144, 160, 197 countryside. See also East Syria, Limestone Plateau, Hauran intramural burial, 148 Late Roman period, 214–215 location cemeteries, 28, 35, 191–192 redevelopment, 9, 207–208 tomb types, 57 cremation, 46, 110, 318, see also Palmyra, Dura Europos, Apamea curses, 143, see also magic Deb’aal burial practices, 241–242 grave goods, 77 Dura Europos burial practices, 246–248 Hellenistic cemeteries, 243 Hellenistic period, 7 history, 9, 10, 242 Parthian cemeteries, 245 soldiers, 30 East Syria. See also Selenkahiye, Dura Europos burial practices, 223 history, 7, 8, 12 location cemeteries, 191 elite benefactions, 189 commemoration, 206–207 culture, 200 Emesa. See Homs extramural burial, 28, 33 family, 128–130, 141, 173, 196–197, 209–210 Frikya (Limestone Plateau) burial practices, 279 Tomb of Abedrapsas, 113, 119, 162

363

364

INDEX

funerary banquets, 87, 131–132, 152 funerary enclosures, 45, 315–316 funerary ritual. See also funerary banquets, offerings, pollution grave goods, 98–101 rites of passage, 14, 157 theoretical approaches, 13–16 tomb-side activities, 150–154, 170–171 gender, 112, 120–124, 128, 142 glass. See vessels globalization concept, 182–183 glocalization, 186 localization, 202 Roman period, 182, 183–186 taste, 178, 198–202 Hama burial practices, 249–251 communal burial, 117 history, 248 Hauran. See also countryside burial practices, 253–263 history, 7, 9, 10, 12, 251 intramural burial, 148 Hellenistic period commemoration, 139, 140 funerary architecture, 61 funerary ritual, 169–170 grave goods, 95–97 history, 7–9 human remains, 139 layout cemeteries, 32 location cemeteries, 29, 35 outside Syria, 75 sample, 17 tomb markers, 66 hellenization, 199–200 Hermel, 63 Homs burial practices, 264–269 history, 263 Tomb of Sampsigeramos, 2 hypogea, 41, 316–318 iconography, 124, 160–163 incense, 152–156 individual identity, 130–131, 142, 172, 177–178, 192–197 inhumation, 110, 149 intramural burial, 24–25, 29, 31, 148 jar-burials, 46, 318 Jebel Khalid, 17 Jebleh burial practices, 269–270 history, 269 jewelry. See personal adornment

lamps, 87–89 Late Roman period, 11, 30, 174, 212–215 Limestone Plateau. See also countryside burial practices, 272–284 history, 7, 10, 271 intramural burial, 148 magic, 83, 90, 99 mausolea, 42, 318–320 migration, 207–208 monumentality, 65–71 mourning. See consolation mummification, 111 nefesh (nfs), 159 North Mesopotamia. See East Syria offerings, 86, 100, 152–156 ownership, 142–143, 196 Palmyra abandonment, 30 age, 126 burial practices, 210–212, 286–301 commemoration, 112 communal burial, 117 funerary ritual, 152 gender, 122 history, 7, 10, 11, 284 North Cemetery, 31, 300 Northeast Cemetery, 300 pre-Roman period, 169 pre-Roman tombs, 285–286 soldiers, 135 Southeast Cemetery, 301 Southwest Cemetery, 31, 301 Tomb C, 117, 152 West Cemetery (Valley of the Tombs), 22, 24, 31, 298–300 Parthian period borders, 11–12 commemoration, 139, 140 funerary architecture, 61 funerary ritual, 169–170 grave goods, 95–97 history, 7–9 human remains, 139 layout cemeteries, 32 location cemeteries, 29 sample, 17 tomb markers, 66 personal adornment, 81–83 pit-graves, 41, 320–321 pollution, 33–34, 99, 148–149 portraiture, 107–110, 192–197, 293–295 pottery. See vessels profession, 131–132

INDEX

Qanawat. See Hauran Qatura. See Limestone Plateau regional identity, 202–204 regional trends commemoration, 138–139, 172 decoration, 162 funerary architecture, 58–60 grave goods, 93 inscriptions, 107 sculpture, 109 rock-reliefs, 55, 329–331 Romanization, 179–182, 183–186 sarcophagi, 46, 53, 321, 327–328, see also coffins sculpture. See portraiture Selenkahiye burial practices, 302–303 sex, 119–120, 128 Shahb¯a (Hauran) burial practices, 260–261 history, 11 intramural burial, 148 Philippeion, 25 Si’ (Hauran). See also Hauran burial practices, 261–262 sanctuary, 9 single burial commemoration, 114

365 differentiation, 167 funerary ritual, 101 grave goods, 94, 116 layout, 60 soldiers. See army status, 15, 165, 168, see also elite stelae, 48–52, 331–334 Suweida. See Hauran Tell Sheikh Hamad, 9, 17, 61, 139 tomb markers, 65–66, see also stelae tower-tombs, 46, 322–324 tumuli, 46, 324–325 Tyre al-Awatin Tomb, 152, 307 al-Bass Cemetery, 21–22, 24, 31, 309–313 burial practices, 306–313 elevated coffins, 201 grave goods, 80 Hellenistic period, 7 history, 304 Tomb of Hiram, 71 urbanization. See cities uses of the past, 201, 203–204 vessels, 83–87 veterans. See army