Once upon a Time in the East: The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East 1784911208, 9781784911201

Citation preview

Archaeopress

Chapter Title: Front Matter Book Title: Once upon a Time in the East Book Subtitle: The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East Book Author(s): Philip Bes Published by: Archaeopress. (2015) Stable URL: https://www.jstor.org/stable/j.ctvr43kch.1 JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected]. Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at https://about.jstor.org/terms

Archaeopress is collaborating with JSTOR to digitize, preserve and extend access to Once upon a Time in the East

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:24 UTC All use subject to https://about.jstor.org/terms

Once upon a Time in the East The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East Philip Bes

Roman and Late Antique Mediterranean Pottery 6 2015 This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:24 UTC All use subject to https://about.jstor.org/terms

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:24 UTC All use subject to https://about.jstor.org/terms

Roman and Late Antique Mediterranean Pottery Archaeopress Series EDITORIAL BOARD (in alphabetical order)

Series Editors Michel BONIFAY, Centre Camille Jullian, Maison Méditerranéenne des Sciences de l’Homme, (Aix Marseille Université/ CNRS) Miguel Ángel CAU, Institució Catalana de Recerca i Estudis Avançats (ICREA)/Equip de Recerca Arqueològica i Arqueomètrica, Universitat de Barcelona (ERAAUB) Paul REYNOLDS, Institució Catalana de Recerca i Estudis Avançats (ICREA)/Equip de Recerca Arqueològica i Arqueomètrica, Universitat de Barcelona (ERAAUB) Honorary editor John HAYES, Institute of Archaeology, University of Oxford Associate editors Philip KENRICK, Institute of Archaeology, University of Oxford John LUND, The National Museum of Denmark, Denmark Scientific Committee for Pottery Xavier AQUILUÉ, Paul ARTHUR, Cécile BATIGNE, Moncef BEN MOUSSA, Darío BERNAL, Raymond BRULET, Claudio CAPELLI, Armand DESBAT, Nalan FIRAT, Michael G. FULFORD, Ioannis ILIOPOULOS, Sabine LADSTÄTTER, Fanette LAUBENHEIMER, Mark LAWALL, Sévérine LEMAÎTRE, Hassan LIMANE, Daniele MALFITANA, Archer MARTIN, Thierry MARTIN, Simonetta MENCHELLI, Henryk MEYZA, Giuseppe MONTANA, Rui MORAIS, Gloria OLCESE, Carlo PAVOLINI, Theodore PEÑA, Verena PERKO, Platon PETRIDIS, Dominique PIERI, Jeroen POBLOME, Natalia POULOU, Albert RIBERA, Lucien RIVET, Lucia SAGUI, Sara SANTORO, Anne SCHMITT, Gerwulf SCHNEIDER, Kathleen SLANE, Roberta TOMBER, Inês VAZ PINTO, Caterina VIEGAS, Yona WAKSMAN General advisors Richard HODGES, Richard REECE, Gisela RIPOLL, Bryan WARD-PERKINS, Chris WICKHAM, Enrico ZANINI

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:24 UTC All use subject to https://about.jstor.org/terms

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:24 UTC All use subject to https://about.jstor.org/terms

Once upon a Time in the East The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East

Philip Bes

Roman and Late Antique Mediterranean Pottery 6

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:24 UTC All use subject to https://about.jstor.org/terms

Archaeopress Publishing Ltd Gordon House 276 Banbury Road Oxford OX2 7ED

www.archaeopress.com

ISBN 978 1 78491 120 1 ISBN 978 1 78491 121 8(e-Pdf)

© Archaeopress and P Bes 2015 Cover illustrations, from left to right: Eastern Sigillata A from the intensive field survey at urban Tanagra (© Philip Bes/Ancient Cities of Boeotia Project); Eastern Sigillata C from the intensive field survey at urban Tanagra (© Ancient Cities of Boeotia Project); Eastern Sigillata D/Cypriot Sigillata (© K. Woszczynska/Polish Mission of the University of Warsaw at Kato Paphos, reproduced with kind permission); Italian Sigillata from the intensive field survey at urban Thespiae (© Ancient Cities of Boeotia Project); Eastern Sigillata B from the intensive field survey at urban Tanagra (© Ancient Cities of Boeotia Project); African Red Slip Ware from the intensive field survey at urban Tanagra (© Ancient Cities of Boeotia Project); Phocaean Late Roman C from the intensive field survey at urban Tanagra (© Ancient Cities of Boeotia Project); Cypriot Red Slip Ware/Late Roman D from Sagalassos (© Sagalassos Archaeological Research Project).

All rights reserved. No part of this book may be reproduced, stored in retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying or otherwise, without the prior written permission of the copyright owners.

Printed in England by Oxuniprint Ltd, Oxford

This book is available direct from Archaeopress or from our website www.archaeopress.com

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:24 UTC All use subject to https://about.jstor.org/terms

‘and I half-closed my eyes and imagined this was the spot where everything I’d ever lost since my childhood had washed up, and I was now standing here in front of it, and if I waited long enough, a tiny figure would appear on the horizon across the field, and gradually get larger’ Kazuo Ishiguro, Never let me go (2005)

In Memoriam

Virginia Ludovica Van Praet 29 June 1920, Niel (Belgium) 7 August 2012, Den Haag (The Netherlands)

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:24 UTC All use subject to https://about.jstor.org/terms

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:24 UTC All use subject to https://about.jstor.org/terms

Archaeopress

Chapter Title: Table of Contents Book Title: Once upon a Time in the East Book Subtitle: The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East Book Author(s): Philip Bes Published by: Archaeopress. (2015) Stable URL: https://www.jstor.org/stable/j.ctvr43kch.2 JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected]. Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at https://about.jstor.org/terms

Archaeopress is collaborating with JSTOR to digitize, preserve and extend access to Once upon a Time in the East

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:25 UTC All use subject to https://about.jstor.org/terms

Contents List of Figures����������������������������������������������������������������������������������������������������������������������������������������������������������� iii List of Abbreviations�������������������������������������������������������������������������������������������������������������������������������������������������� v Acknowledgements�������������������������������������������������������������������������������������������������������������������������������������������������� vi Chapter 1: Aims and Methodology Introduction�������������������������������������������������������������������������������������������������������������������������������������������������������������������� 1 1.1 Definition of Terms�������������������������������������������������������������������������������������������������������������������������������������������������� 1 1.2 Aims������������������������������������������������������������������������������������������������������������������������������������������������������������������������� 2 1.3 Methodology������������������������������������������������������������������������������������������������������������������������������������������������������������ 3 1.3.1 The Published Record and Data Collection����������������������������������������������������������������������������������������������������� 3 1.3.2 Typo-Chronologies������������������������������������������������������������������������������������������������������������������������������������������ 5 1.3.3 Descriptive Statistics��������������������������������������������������������������������������������������������������������������������������������������� 5 1.3.4 Data: Problems and Limitations���������������������������������������������������������������������������������������������������������������������� 7 1.3.5 The Chronological Setting������������������������������������������������������������������������������������������������������������������������������� 8 1.3.6 The Geographical Setting�������������������������������������������������������������������������������������������������������������������������������� 9 Chapter 2: Definition of the Topic���������������������������������������������������������������������������������������������������������������������������� 11 2.1 History of Research������������������������������������������������������������������������������������������������������������������������������������������������ 11 2.1.1 Pioneering Explorations�������������������������������������������������������������������������������������������������������������������������������� 11 2.1.2 The Current Framework�������������������������������������������������������������������������������������������������������������������������������� 11 2.2 Definition of the Eight Wares��������������������������������������������������������������������������������������������������������������������������������� 12 2.2.1 Eastern Sigillata A (ESA)������������������������������������������������������������������������������������������������������������������������������ 12 2.2.2 Eastern Sigillata B (ESB)������������������������������������������������������������������������������������������������������������������������������ 16 2.2.3 Eastern Sigillata C, Çandarlı Ware (ESC)����������������������������������������������������������������������������������������������������� 17 2.2.4 Eastern Sigillata D/Cypriot Sigillata (ESD)�������������������������������������������������������������������������������������������������� 19 2.2.5 Italian Sigillata (ITS)������������������������������������������������������������������������������������������������������������������������������������� 20 2.2.6 African Red Slip Ware (ARSW)�������������������������������������������������������������������������������������������������������������������� 22 2.2.7 Cypriot Red Slip Ware/Late Roman D (LRD)���������������������������������������������������������������������������������������������� 24 2.2.8 Phocaean Red Slip Ware/Late Roman C (LRC)�������������������������������������������������������������������������������������������� 24 Chapter 3: The Late Hellenistic and Early Roman Periods: Discussion and Chronological Phasing����������������������������� 27 3.1 Presentation of the Collected Published Evidence������������������������������������������������������������������������������������������������� 27 3.2 Regional Developments����������������������������������������������������������������������������������������������������������������������������������������� 28 3.2.1 Africa-Cyrenaica�������������������������������������������������������������������������������������������������������������������������������������������� 28 3.2.2 Crete�������������������������������������������������������������������������������������������������������������������������������������������������������������� 30 3.2.3 Aegean-Southwest����������������������������������������������������������������������������������������������������������������������������������������� 33 3.2.4 Aegean-North������������������������������������������������������������������������������������������������������������������������������������������������� 36 3.2.5 Aegean-East��������������������������������������������������������������������������������������������������������������������������������������������������� 38 3.2.6 Asia Minor-West Central������������������������������������������������������������������������������������������������������������������������������� 41 3.2.7 Asia Minor-South Coast�������������������������������������������������������������������������������������������������������������������������������� 42 3.2.8 Asia Minor-East Central�������������������������������������������������������������������������������������������������������������������������������� 43 3.2.9 Asia Minor-Cilicia����������������������������������������������������������������������������������������������������������������������������������������� 43 3.2.10 Levant-Coastal North���������������������������������������������������������������������������������������������������������������������������������� 45 3.2.11 Levant-Interior North����������������������������������������������������������������������������������������������������������������������������������� 46 3.2.12 Levant-Coastal South���������������������������������������������������������������������������������������������������������������������������������� 50 3.2.13 Levant-Interior South���������������������������������������������������������������������������������������������������������������������������������� 53 3.2.14 Africa-Egypt������������������������������������������������������������������������������������������������������������������������������������������������ 55 3.2.15 Cyprus���������������������������������������������������������������������������������������������������������������������������������������������������������� 57 3.3 The Distribution of Late Hellenistic and Early Roman Terra Sigillata in the Eastern Mediterranean������������������� 61 3.3.1 Phase 1 (c. 150-30 BC)���������������������������������������������������������������������������������������������������������������������������������� 61 3.3.2 Phase 2 (c. 30 BC-25/30)������������������������������������������������������������������������������������������������������������������������������� 64 3.3.3 Phase 3 (c. 30-60/70)������������������������������������������������������������������������������������������������������������������������������������� 68 3.3.4 Phase 4 (c. 70/75-200)����������������������������������������������������������������������������������������������������������������������������������� 72 Chapter 4: Framing Late Hellenistic and Early Roman Tablewares���������������������������������������������������������������������������� 77 Introduction������������������������������������������������������������������������������������������������������������������������������������������������������������������ 77 4.1 Difficulties of the Evidence������������������������������������������������������������������������������������������������������������������������������������ 77 4.2 Contextualising Tablewares����������������������������������������������������������������������������������������������������������������������������������� 77

i

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:25 UTC All use subject to https://about.jstor.org/terms

4.2.1 Modes of Exchange��������������������������������������������������������������������������������������������������������������������������������������� 77 4.2.2 Human and Institutional Factors�������������������������������������������������������������������������������������������������������������������� 78 4.2.3 Goods and Commodities������������������������������������������������������������������������������������������������������������������������������� 78 4.3 Interpretation of the Collected Evidence���������������������������������������������������������������������������������������������������������������� 78 4.3.1 Phase 1: The Late Hellenistic Period till the Reign of Augustus������������������������������������������������������������������� 78 4.3.2 Phase 2: The Augustan-Tiberian Period�������������������������������������������������������������������������������������������������������� 82 4.3.3 Phase 3: The Claudian-Neronian Period������������������������������������������������������������������������������������������������������� 85 4.3.4 Phase 4: The Late 1st and 2nd Century��������������������������������������������������������������������������������������������������������� 87 Chapter 5: The Mid- and Late Roman Periods: Discussion and Chronological Phasing���������������������������������������������� 90 Introduction������������������������������������������������������������������������������������������������������������������������������������������������������������������ 90 5.1 The Collected Published Evidence������������������������������������������������������������������������������������������������������������������������� 90 5.2 Comparing the Collected Evidence with Key Literature��������������������������������������������������������������������������������������� 90 5.3 Regional Developments����������������������������������������������������������������������������������������������������������������������������������������� 93 5.3.1 Africa-Cyrenaica�������������������������������������������������������������������������������������������������������������������������������������������� 93 5.3.3 Aegean-Southwest����������������������������������������������������������������������������������������������������������������������������������������� 97 5.3.4 Aegean-North����������������������������������������������������������������������������������������������������������������������������������������������� 100 5.3.5 Aegean-East������������������������������������������������������������������������������������������������������������������������������������������������� 103 5.3.6 Asia Minor-West Central����������������������������������������������������������������������������������������������������������������������������� 105 5.3.7 Asia Minor-South Coast������������������������������������������������������������������������������������������������������������������������������ 106 5.3.8 Asia Minor-East Central������������������������������������������������������������������������������������������������������������������������������ 106 5.3.9 Asia Minor-Cilicia��������������������������������������������������������������������������������������������������������������������������������������� 107 5.3.10 Levant-Coastal North�������������������������������������������������������������������������������������������������������������������������������� 108 5.3.11 Levant-Interior North����������������������������������������������������������������������������������������������������������������������������������111 5.3.12 Levant-Coastal South�������������������������������������������������������������������������������������������������������������������������������� 113 5.3.13 Levant-Interior South�������������������������������������������������������������������������������������������������������������������������������� 115 5.3.14 Africa-Egypt���������������������������������������������������������������������������������������������������������������������������������������������� 117 5.3.15 Cyprus�������������������������������������������������������������������������������������������������������������������������������������������������������� 119 5.4 Tableware Distribution in the Mid- to Late Roman East������������������������������������������������������������������������������������� 122 5.4.1 Phase 5 (c. 200/225-325)����������������������������������������������������������������������������������������������������������������������������� 122 5.4.2 Phase 6 (c. 325-400/425)����������������������������������������������������������������������������������������������������������������������������� 125 5.4.3 Phase 7 (c. 425/450-500/525)���������������������������������������������������������������������������������������������������������������������� 127 5.4.4 Phase 8 (c. 500/525-575/600)���������������������������������������������������������������������������������������������������������������������� 128 5.4.5 Phase 9 (c. 575/600-700)����������������������������������������������������������������������������������������������������������������������������� 130 Chapter 6: Interpretation of the Quantitative, Chronological, and Geographical Evidence for the Mid- to Late Roman Period������������������������������������������������������������������������������������������������������������������������������������������������������������������� 133 Introduction���������������������������������������������������������������������������������������������������������������������������������������������������������������� 133 6.1 Interpretation of the Collected Evidence�������������������������������������������������������������������������������������������������������������� 133 6.1.1 The Distribution of ARSW in the 3rd Century�������������������������������������������������������������������������������������������� 134 6.1.2 After the early 4th Century�������������������������������������������������������������������������������������������������������������������������� 135 6.1.3 The Vandal Period���������������������������������������������������������������������������������������������������������������������������������������� 137 6.1.4 The Later 5th to Mid-6th Century��������������������������������������������������������������������������������������������������������������� 138 6.1.5 The Later 6th Century to the ‘End’ of Antiquity����������������������������������������������������������������������������������������� 139 6.2 Summary�������������������������������������������������������������������������������������������������������������������������������������������������������������� 140 Chapter 7: Thoughts on the Socio-Economic Significance of Red Slip Tablewares��������������������������������������������������� 142 7.1 Socio-Economic Observations����������������������������������������������������������������������������������������������������������������������������� 142 7.2 Socio-Cultural Observations�������������������������������������������������������������������������������������������������������������������������������� 147 7.3 Concluding Thoughts������������������������������������������������������������������������������������������������������������������������������������������� 149 7.3.1 Typology and Chronology��������������������������������������������������������������������������������������������������������������������������� 149 7.3.2 Quantity and Quantification������������������������������������������������������������������������������������������������������������������������ 149 7.3.3 Local, Close-Regional and Supraregional��������������������������������������������������������������������������������������������������� 150 7.3.4 Scope and Category������������������������������������������������������������������������������������������������������������������������������������� 150 Bibliography���������������������������������������������������������������������������������������������������������������������������������������������������������� 153 Classical Authors�������������������������������������������������������������������������������������������������������������������������������������������������������� 153 Modern Authors��������������������������������������������������������������������������������������������������������������������������������������������������������� 153 Appendices����������������������������������������������������������������������������������������������������������������������������������������������������������� 182

ii

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:25 UTC All use subject to https://about.jstor.org/terms

Archaeopress

Chapter Title: List of Figures Book Title: Once upon a Time in the East Book Subtitle: The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East Book Author(s): Philip Bes Published by: Archaeopress. (2015) Stable URL: https://www.jstor.org/stable/j.ctvr43kch.3 JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected]. Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at https://about.jstor.org/terms

Archaeopress is collaborating with JSTOR to digitize, preserve and extend access to Once upon a Time in the East

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:26 UTC All use subject to https://about.jstor.org/terms

List of Figures Figure 1. An example of the method of distribution by Fentress and Perkins 1988 ��������������������������������������������������������������������������������6 Figure 2. Line chart showing the development, in absolute numbers, of the collected evidence per 15-year interval (n=15,665) �������6 Figure 3. The nine-phase chronological framework used in this study ���������������������������������������������������������������������������������������������������8 Figure 4. Map of the eastern Mediterranean showing the regional framework used throughout the text; �������������������������������������������9 Figure 5. Late Hellenistic ESA (predating c. 60-50 BC) from the excavations at Kinet Höyük (Hatay, Turkey), ���������������������������������������13 Figure 6. Three fragments of ITS ������������������������������������������������������������������������������������������������������������������������������������������������������������21 Figure 7. Sites and areas of production of ARSW in Tunisia ������������������������������������������������������������������������������������������������������������������23 Figure 8. Several types of Phocaean LRC Hayes form 3, collected during intensive field survey at urban Tanagra �������������������������������25 Figure 9. Line chart showing the development, in absolute numbers, of ITS, ESA, ESB, ESC, ESD and ARSW ���������������������������������������27 Figure 10. Africa-Cyrenaica: pie chart showing the proportions of the different wares �����������������������������������������������������������������������28 Figure 11. Africa-Cyrenaica: map showing the ware proportions of the different sites ������������������������������������������������������������������������29 Figure 12. Africa-Cyrenaica: line chart showing the development, in absolute numbers, of the different wares ��������������������������������29 Figure 13. Crete: pie chart showing the proportions of the different wares �����������������������������������������������������������������������������������������30 Figure 14. Crete: map showing the ware proportions of the different sites and surveys ���������������������������������������������������������������������31 Figure 15. Crete: line chart showing the development, in absolute numbers, of the different wares ��������������������������������������������������32 Figure 16. Aegean-Southwest: pie chart showing the proportions of the different wares �������������������������������������������������������������������33 Figure 17. Aegean-Southwest: map showing the ware proportions of the different sites and surveys ������������������������������������������������34 Figure 18. Aegean-Southwest: line chart showing the development, in absolute numbers, of the different wares ����������������������������34 Figure 19. Aegean-North: pie chart showing the proportions of the different wares ���������������������������������������������������������������������������36 Figure 20. Aegean-North: map showing the ware proportions of the different sites ���������������������������������������������������������������������������37 Figure 21. Aegean-North: line chart showing the development, in absolute numbers, of the different wares ������������������������������������37 Figure 22. Aegean-East: pie chart showing the proportions of the different wares ������������������������������������������������������������������������������39 Figure 23. Aegean-East: map showing the ware proportions of the different sites ������������������������������������������������������������������������������39 Figure 24. Aegean-East: line chart showing the development, in absolute numbers, of the collected evidence ���������������������������������40 Figure 25. Asia Minor-West Central: pie chart showing the proportions of the different wares ����������������������������������������������������������41 Figure 26. Asia Minor-West Central, -East Central and -South Coast: map showing the ware proportions ������������������������������������������42 Figure 27. Asia Minor-East Central: pie chart showing the proportions of the different wares ������������������������������������������������������������43 Figure 28. Asia Minor-Cilicia: pie chart showing the proportions of the different wares ����������������������������������������������������������������������43 Figure 29. Asia Minor-Cilicia: map showing the ware proportions of the different sites and surveys ��������������������������������������������������44 Figure 30. Asia Minor-Cilicia: line chart showing the development, in absolute numbers, of the different wares �������������������������������45 Figure 31. Levant-Coastal North: pie chart showing the proportions of the different wares ����������������������������������������������������������������46 Figure 32. Levant-Coastal North: map showing the ware proportions of the different sites ����������������������������������������������������������������47 Figure 33. Levant-Coastal North: line chart showing the development, in absolute numbers, of the different wares �������������������������48 Figure 34. Levant-Interior North: pie chart showing the proportions of the different wares ���������������������������������������������������������������48 Figure 35. Levant-Interior North: map showing the ware proportions of the different sites ����������������������������������������������������������������49 Figure 36. Levant-Interior North: line chart showing the development, in absolute numbers, of the different wares ������������������������49 Figure 37. Levant-Coastal South: pie chart showing the proportions of the different wares ����������������������������������������������������������������50 Figure 38. Levant-Coastal South: map showing the ware proportions of the different sites ����������������������������������������������������������������51 Figure 39. Levant-Coastal South: line chart showing the development, in absolute numbers, of the different wares �������������������������51 Figure 40. Levant-Interior South: pie chart showing the proportions of the different wares ���������������������������������������������������������������53 Figure 41. Levant-Interior South: map showing the ware proportions of the different sites ����������������������������������������������������������������54 Figure 42. Levant-Interior South: line chart showing the development, in absolute numbers, of the different wares ������������������������54 Figure 43. Africa-Egypt: pie chart showing the proportions of the different wares ������������������������������������������������������������������������������55 Figure 44. Africa-Egypt: map showing the ware proportions of the different sites ������������������������������������������������������������������������������56 Figure 45. Africa-Egypt: line chart showing the development, in absolute numbers, of the different wares ���������������������������������������56 Figure 46. Cyprus: pie chart showing the proportions of the different wares ��������������������������������������������������������������������������������������57 Figure 47. Cyprus: map showing the ware proportions of the different sites and surveys �������������������������������������������������������������������58 Figure 48. Cyprus: line chart showing the development, in absolute numbers, of the different wares �����������������������������������������������58 Figure 49. Line chart showing the development, in absolute numbers, of ESA and ESD on western and eastern Cyprus ��������������������60 Figure 50. Map for phase 1a, showing the distribution, in absolute numbers, of ESA, ESC and ESD forms that predate c. 75 BC �������62 Figure 51. Map showing the distribution, in absolute numbers, of ESA, ESC and ESD during phase 1b (c. 150-30 BC) ������������������������63 Figure 52. Map showing the distribution, in absolute numbers, of ITS, ESA, ESB, ESC and ESD during phase 2(a) (c. 30 BC-25/30) ����65 Figure 53. The distribution, in absolute numbers, of ITS form and stamp entries that predate 1 ��������������������������������������������������������67 Figure 54. Map showing the distribution, in absolute numbers, of ITS, ESA, ESB, ESC and ESD during phase 3 (c. 30-60/70) �������������69 Figure 55. Map showing the distribution, in absolute numbers, of ITS, ESA, ESB, ESC, ESD and ARSW during phase 4 �����������������������73 Figure 56. Line chart showing the development, in absolute numbers, of ESB in the Aegean-East ������������������������������������������������������75 Figure 57. Column chart comparing the distribution (in percentages) of ESA forms of phase 1 �����������������������������������������������������������80 Figure 58. Line chart showing the development, in absolute numbers, of ESC, ESD, ARSW, LRD and LRC ��������������������������������������������91 Figure 59. Africa-Cyrenaica: pie chart showing the proportions of the different wares �����������������������������������������������������������������������93 Figure 60. Africa-Cyrenaica: map showing the ware proportions of the different sites ������������������������������������������������������������������������94 Figure 61. Africa-Cyrenaica: line chart showing the development, in absolute numbers, of the different wares ��������������������������������94 Figure 62. Crete: pie chart showing the proportions of the different wares �����������������������������������������������������������������������������������������95 Figure 63. Crete: map showing the ware proportions of the different sites and surveys ���������������������������������������������������������������������96 Figure 64. Crete: line chart showing the development, in absolute numbers, of the different wares ��������������������������������������������������96 Figure 65. Aegean-Southwest: pie chart showing the proportions of the different wares �������������������������������������������������������������������98 Figure 66. Aegean-Southwest: map showing the ware proportions of the different sites and surveys ������������������������������������������������99 Figure 67. Aegean-Southwest: line chart showing the development, in absolute numbers, of the different wares ����������������������������99

iii

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:26 UTC All use subject to https://about.jstor.org/terms

Figure 68. Aegean-North: pie chart showing the proportions of the different wares �������������������������������������������������������������������������100 Figure 69. Aegean-North: map showing the ware proportions of the different sites �������������������������������������������������������������������������101 Figure 70. Aegean-North: line chart showing the development, in absolute numbers, of the different wares ����������������������������������101 Figure 71. Aegean-East: pie chart showing the proportions of the different wares ����������������������������������������������������������������������������103 Figure 72. Aegean-East: map showing the ware proportions of the different sites ����������������������������������������������������������������������������104 Figure 73. Aegean-East: line chart showing the development, in absolute numbers, of the different wares �������������������������������������104 Figure 74. Asia Minor-West Central, -South Coast and -East Central: map showing the ware proportions of the different sites ������105 Figure 75. Asia Minor-South Coast: pie chart showing the proportions of the different wares ����������������������������������������������������������106 Figure 76. Asia Minor-Cilicia: pie chart showing the proportions of the different wares ��������������������������������������������������������������������107 Figure 77. Asia Minor-Cilicia: map showing the ware proportions of the different sites and surveys ������������������������������������������������108 Figure 78. Column chart, capturing the differences (in percentages) of ARSW, LRD and LRC �������������������������������������������������������������108 Figure 79. Asia Minor-Cilicia: line chart showing the development, in absolute numbers, of the different wares �����������������������������109 Figure 80. Levant-Coastal North: pie chart showing the proportions of the different wares ��������������������������������������������������������������109 Figure 81. Levant-Coastal North: map showing the ware proportions of the different sites ��������������������������������������������������������������110 Figure 82. Levant-Coastal North: line chart showing the development, in absolute numbers, of the different wares �����������������������110 Figure 83. Levant-Interior North: pie chart showing the proportions of the different wares �������������������������������������������������������������111 Figure 84. Levant-Interior North: map showing the ware proportions of the different sites ��������������������������������������������������������������112 Figure 85. Levant-Interior North: line chart showing the development, in absolute numbers, of the different wares ����������������������112 Figure 86. Levant-Coastal South: pie chart showing the proportions of the different wares ��������������������������������������������������������������113 Figure 87. Levant-Coastal South: map showing the ware proportions of the different sites ��������������������������������������������������������������114 Figure 88. Levant-Coastal South: line chart showing the development, in absolute numbers, of the different wares �����������������������114 Figure 89. Levant-Interior South: pie chart showing the proportions of the different wares �������������������������������������������������������������115 Figure 90. Levant-Interior South: map showing the ware proportions of the different sites and surveys ������������������������������������������116 Figure 91. Levant-Interior South: line chart showing the development, in absolute numbers, of the different wares ����������������������116 Figure 92. Africa-Egypt: pie chart showing the proportions of the different wares ����������������������������������������������������������������������������117 Figure 93. Africa-Egypt: map showing the ware proportions of the different sites�����������������������������������������������������������������������������118 Figure 94. Africa-Egypt: line chart showing the development, in absolute numbers, of the different wares �������������������������������������118 Figure 95. Cyprus: pie chart showing the proportions of the different wares ������������������������������������������������������������������������������������120 Figure 96. Cyprus: map showing the ware proportions of the different sites and surveys �����������������������������������������������������������������121 Figure 97. Cyprus: line chart showing the development, in absolute numbers, of the different wares ���������������������������������������������121 Figure 98. Map showing the distribution, in absolute numbers, of ESC and ARSW during phase 5b �������������������������������������������������122 Figure 99. Map showing the distribution, in absolute numbers, of ARSW during phase 5a ���������������������������������������������������������������123 Figure 100. Map showing the distribution, in absolute numbers, of ARSW, LRD and LRC during phase 6 ������������������������������������������125 Figure 101. Map showing the distribution, in absolute numbers, of ARSW, LRD and LRC during phase 7 ������������������������������������������128 Figure 102. Map showing the distribution, in absolute numbers, of ARSW, LRD and LRC during phase 8 ������������������������������������������129 Figure 103. Map showing the distribution, in absolute numbers, of ARSW, LRD and LRC during phase 9 ������������������������������������������131 Figure 104. The collected data concerning the late Hellenistic and early Roman eastern Mediterranean������������������������������������������142 Figure 105. The collected data concerning the mid- and late Roman eastern Mediterranean, showing core zones of distribution ��143 Figure 106. Table showing the number of different ESA forms in its core zone (Levant and Cyprus)��������������������������������������������������144 Figure 107. Stacked column chart showing the number of different forms per ware, per fifteen-year interval ���������������������������������147 Figure 108. Morphological correspondence between some of the more common ITS forms ������������������������������������������������������������148 Figure 109. Column chart showing counts of the identified ITS forms, in percentages, in the collected data ������������������������������������149 Appendix 1a. Map showing all sites and surveys from where data was collected ������������������������������������������������������������������������������182 Appendix 1b. Map showing sites and surveys otherwise mentioned in the text ��������������������������������������������������������������������������������183 Appendix 2. Table showing, per region, the sites and surveys from where data was collected ����������������������������������������������������������184 Appendix 3a. Table showing the absolute and relative quantities, per region, �����������������������������������������������������������������������������������185 Appendix 3b. Table showing the absolute and relative quantities, per region, ����������������������������������������������������������������������������������185 Appendix 4a. Table showing the absolute quantities, per phase and per site, of ITS, ESA, ESB, ESC, ESD and ARSW �������������������������190 Appendix 4b. Table showing the absolute quantities, per phase and per site, of ESC, ESD, ARSW, LRD and LRC �������������������������������194

iv

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:26 UTC All use subject to https://about.jstor.org/terms

Archaeopress

Chapter Title: List of Abbreviations Book Title: Once upon a Time in the East Book Subtitle: The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East Book Author(s): Philip Bes Published by: Archaeopress. (2015) Stable URL: https://www.jstor.org/stable/j.ctvr43kch.4 JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected]. Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at https://about.jstor.org/terms

Archaeopress is collaborating with JSTOR to digitize, preserve and extend access to Once upon a Time in the East

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:27 UTC All use subject to https://about.jstor.org/terms

List of Abbreviations

BC Before Christ AD Anno Domini AAS Atomic Absorption Spectrometry ARSW African Red Slip Ware Atlante Enciclopedia dell’Arte. Classica e Orientale. Atlante della forme ceramiche II BSP Black-Slipped Predecessor Consp Conspectus Formarum CRSW Cypriot Red Slip Ware CVA Corpus Vasorum Arretinorum ERSW1 Egyptian Red Slip Ware ESA Eastern Sigillata A ESB Eastern Sigillata B ESC Eastern Sigillata C ESD Eastern Sigillata D (also known as Cypriot Sigillata) ETS Eastern Terra Sigillata ITS Italian Terra Sigillata LRC Late Roman C LRD Late Roman D LRP Late Roman Pottery NAA Neutron Activation Analysis OCK Oxé, Comfort and Kenrick 2000 PRSW Phocaean Red Slip Ware SRSW Sagalassos Red Slip Ware XRF X-Ray Fluoresence

1

Not to be confused with Ephesian Red Slip Ware, for which see Iro et al. 2009, 55, ns 11-12.

v

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:27 UTC All use subject to https://about.jstor.org/terms

Archaeopress

Chapter Title: Acknowledgements Book Title: Once upon a Time in the East Book Subtitle: The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East Book Author(s): Philip Bes Published by: Archaeopress. (2015) Stable URL: https://www.jstor.org/stable/j.ctvr43kch.5 JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected]. Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at https://about.jstor.org/terms

Archaeopress is collaborating with JSTOR to digitize, preserve and extend access to Once upon a Time in the East

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:27 UTC All use subject to https://about.jstor.org/terms

Acknowledgements

Early on a cold and cloudy Friday, 2 January 2004, I arrived in Leuven as a newly qualified archaeologist. On a warm and sunny Thursday, 17 July 2014, I left Leuven, a city that has further shaped, turned and moulded me, both as a person and as an archaeologist. Those 10.5 years (over 2.5 of which were spent abroad) have not left me unaffected, and with both great joy and the occasional and always unexpected flash of melancholy I look back on my time in Leuven. First and foremost I wish to express my deepest gratitude to Jeroen Poblome. Throughout my time in Leuven, particularly over the course of my doctoral research at the University of Leuven between 2004 and 2007, but also afterwards, he has been a patient, supportive and crucial guide, who quickly realised what would and would not work. It is not easy for me to capture in words the opportunity he has provided to let me discover and develop my love for all those pottery fragments. But perhaps in reminiscing about the snow-covered hills of Boeotia, the cruising across the vast plains of Cilicia Pedias in search for that elusive ESA, and of course discussing Sagalassos Red Slip Ware in front of your little corner of the world (regularly interspersed with telephone calls), Sagalassos Kazı Evi, depo yedi, you can see the combination of archaeology, the search for and pleasure of good food, and the enjoyment of that often wonderful luxury of being an archaeologist working in the Mediterranean that these opportunities have afforded me. Thank you so much Jeroen. Marc Waelkens, former director of the Sagalassos Archaeological Research Project, deserves a special mention for creating the framework in which I, one of many, was also able to learn and work. Sagalassos, as you know, is very dear to me. Ever since my very first participation, in 2002, I have come to deeply appreciate that place and its surroundings and the project it accommodates. It is, truly, one of the most beautiful places that I know, and I am most grateful that I was able to spend so many months there. Leuven/Sagalassos — at some point the two began to gradually merge — was inhabitated by a lot of colleagues, many of whom have contributed to my time in Leuven in one way or the other. Some deserve special mention: my first compatriots on the (then still) isolated third floor of the Sagalassoshuis, Julian Richard and Ine Jacobs; Ine, thank you for some fine years. Rinse Willet, for long talks and laughs, as well as our subtle attempt at colonising Flanders. A warm thank you to: Isabelle Aerts, Jan Baeten, Johan Bakker, Roel van Beeumen, Rick Bonnie, Dennis Braekmans, Tom Brughmans, Frank Carpentier, Johan Claeys, Bart Clijsner, Markku Corremans, Ralf van Dam, Jonas Danckers, Frederik Daniëls, Patrick Degryse, Koen Demarsin, Sevgi Gerçek, Vera Goethals, Bart De Graeve, Ann Hasendonckx, Johan van Heesch, Eva Kaptijn, Veerle Lauwers, Eva Leplat, Semra Mägele, Eliane Mahy, Elena Marinova, Femke Martens, Johan Van Neck, Toon Putzeys, Rob Rens, Kerlijne Romanus, Ilse Schoep, Peter Talloen, Shuki Tang, Marianne Toonen, Ebru Torun, Inge Uytterhoeven, Hannelore Vanhaverbeke, Tom Verbist, Tijl Vereenooghe and Kim Vyncke. Joeri Theelen deserves special mention for working his magic on all distribution maps, and helping out on various technical and software matters. In Sagalassos proper, where life in the excavation house mostly concerns the depots (before 18.00 o’clock) and the pergola (after 18.00 o’clock), I particularly would like to thank Yaprak(ske) Özkönü and Senem(meke) Özden for some wonderful years working together, and your hospitality in Istanbul. Nalan Fırat, for your kindness and showing me around Kaş; the bunker — aka depot 13 — will simply never be the same without you. Further thanks to: Kerim Altuğ, Söner Bellibas, Mike Carremans, Bea De Cupere, Pieterjan Deckers, In Aé Delhaye, Bernard Van Daele, Ertuğ Ergürer, Uğur Gürsoy, Paul Hostyn, Geroen Joris, Nathalie Kellens, Emine Kocak, Emrah Kosgeroğlu, Véronique de Laet, Marie Lefere, Igor Medarič, Dirk Menten, Elizabeth Murphy, Branko Mušič, Wim Van Neer, Karel Paul, Ben Rubin, Nerina de Silva, Ilse de Smedt, Esin Tekin, Sofie Thys, Hendrik Uleners, Erdal Ünal, Serpil Uyar, Sarah Vaelen, Leo Vanhecke, Jeroen Vermeersch and Neslihan Yilmaz. Doctors and other volunteers have further helped to bring each campaign to a good end. Bruno Vandermeulen and Danny Veys, the two towers, thank you for transforming so many sherds and pots into wonderful bits and bytes. The Ağlasun men that work(ed) on the Sagalassos excavation created so much atmosphere and were also never short of coming up with new and interesting ways to improve it. Mustafa and Kamille Kantekin, thank you for your care for the team each summer. Prior to moving to Leuven I had already mustered a great fondness for Boeotia. I am most grateful to John Bintliff and Anthony Snodgrass, directors of the Ancient Cities of Boeotia Project, for appreciating, year after year, my taking care (often in collaboration) of the Roman-period pottery amassed since the late 1970s. The scope of your project, and the ‘Mediterraneanness’ of the pottery, has taught me much. Vladimir Stissi, Athanasios Vionis, Emeri Farinetti, Kalliopi Sarri, Bart Noordervliet, Chiara Piccoli and Janneke van Zwienen are long-time colleagues and have helped to bring each productive and pleasurable campaign to a good end. In more recent years, a well-working labteam — also freely volunteering to help bring word jokes to the next level; post-modern sherds can indeed be found, when properly looked for — emerged, and I particularly thank Yannick Boswinkel, Dean Peeters and Anna Meens (also for cooking dinner

vi

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:27 UTC All use subject to https://about.jstor.org/terms

for the team together, on occasion). Coen Bernoster, Elisabeth de Campenhout, Damjan Donev, Roderick Geerts, Peter de Graaf, Nikè Haverkamp, Peter Hazen, Xerena Jué, Chrystalla Loizou, Fabienne Marchand, Esther Mulder, Yasemin Özarslan, Leon Theelen, Marjolein Verschuur, Keith and Myra Wilkinson-Van Hoek and all other participants that have in one way or the other contributed to the Boeotia Project, I thank you. My dear friend Mark van der Enden deserves special mention: your delightful sense of humour and our frequent ‘endless’ talks continue to offer an extremely enjoyable working atmosphere. In more recent years I have been working as a freelance archaeologist, which gave me the opportunity to participate in other projects, thus broadening my scope and knowledge, and which led to pleasant and instructive collaborations. In (central) Greece, I warmly thank Kostas Papagiannopoulos and Eleni Simoni (Achaïa, Sageïka and Kamenitsa surveys), for their kindness and hospitality, together with Vassilis Argyropoulos and Niki Ralli; Kris Tytgat, former director of the Nederlands Instituut Athene (NIA), for allowing me to study the pottery from her excavations at Titanè; Nienke Pieters, Gert-Jan van Wijngaarden and Vladimir Stissi for welcoming me to study the pottery finds from their Zakynthos Archaeological Project. Finally, during three campaigns on Skyros I closely collaborated with Michalis Karambinis on the late Roman ceramic finds from his survey of the island. From windswept valleys to the Calypso Bar, each time was a great pleasure — also thanks to your family and Marianna Gkeka — and I am most pleased that we have found a way to continue our collaboration. In Greece, I am further grateful to Katerina Chamilaki for making things work, Evgenia Gerousi, Panagiotis Kontolaimos, Maria Papoulia and her family, and Emmy Makri of the Nederlands Instituut Athene. In Turkey, within the Cide Archaeological Project, that carried out field survey in northern Turkey, I am very grateful to Bleda Düring, Claudia Glatz and Toby Wilkinson. Involvement with the — now finished — excavations at Kinet Höyük, offered us the opportunity to familiarise ourselves with Hellenistic pottery, specifically with ESA, BSP and related classes. I am indebted to Marie-Henriette and Charles Gates; Gunnar Lehmann, Asa Eger, Yağmur Heffron and Salima Ikram, among others, contributed to an unforgettable campaign in July 2012. Hatay in July… The Austrian excavations at Limyra have offered me a quiet and enjoyable working atmosphere for two years: I thankfully name Martin Seyer, director of the Austrian excavations at Limyra. Zeynep Kuban, Banu Yener-Marksteiner, Sandra Mayer, Ulrike Schuh, Regina Hügli, Katinka Sewing, Kutay Yüncüler, Christian Kurtze, Jan Novacek, Kristina Scheelen and Helmut Lotz have all added to an enjoyable and productive climate. Finally, in 2012 I was fortunate enough to join the Kinneret Regional Project (Galilee, northern Israel), specifically for the study of the mid- to late Roman pottery from the synagogue and domestic area excavations at Horvat Kur: I express many thanks to the directors, Jürgen Zangenberg, Raimo Hakola, Stefan Münger and Byron McCane, and all others that I met and/or collaborated with: Anneke Berkheij-Dol, Rick Bonnie, Benjamin de Groot, Irena Guttmann, Lise den Hertog, Saimi Kautonen, Damian Kessi, Christa Lennert, Eric Ottenheijm, Tine Rassalle, Annalize Rheeder and Ulla Tervahauta. Sirpa Alaranta deserves to be singled out for being a wonderful project manager and fine person. Michel Bonifay, John Lund and Ilse Schoep kindly accepted the invitation to be members on my doctoral jury back in 2007, and ploughed their way through the repetitive descriptions. You have all continued to be helpful, for which I warmly thank you. Advice, suggestions and comments from other colleagues have without exception been helpful, and I wish to express my gratitude to: Krzysztof Domżalski, John Hayes, Fabienne Marchand, Carlo de Mitri, Patrick Monsieur, Pláton Petrídis, Kathleen Warner Slane, Roberta Tomber, Joanita Vroom, David Williams and Kristina Winther-Jacobsen. The text was carefully read and corrected by Patricia Francis, who I am most grateful for this task. Mark van der Enden, Jeroen Poblome and Paul Reynolds read and commented on earlier drafts. Paul Reynolds, series editor with Michel Bonifay and Miguel Ángel Cau Ontiveros of the ‘Roman and Late Antique Mediterranean Pottery’, kindly suggested that my doctoral research could become a member of that family: many, many thanks. At Archaeopress, David Davison and Rajka Makjanic kindly guided this process and patiently answered a variety of questions. The plain fact is that I simply could not have been the person in Leuven without my 5.5 years at the University of Leiden: an inspiring Classical Archaeology program, from where I well remember classes by Paul Meyboom, Miguel John Versluys and John Bintliff. For this, dare I say, formative time: thank you Luc Amkreutz, Dienke van Baardewijk, Dennis Bruggink, Mirjam Bruineberg, Pieter de Man, Gerrit Dusseldorp, Emile Eimermann, Rutger Elsma, Ruben Ettes, Carmen Harmsen, Yannick Henk, Leon van Hoof, Hylke de Jong, Niels Kraal, Eva van der Laan, Jolanda Lee, Geeske Langejans, Leendert van der Meij, Janusz Moskal, Huib-Jan van Oort, Roos van Oosten, Job Pantjes, Léonie Postma, Pjotter Sannen, Karin Schuitema, Liesbeth Schuurman, Peter Stokkel, Margriet Stronkhorst, Agnes Timmers, Floor Timmermans, Femke Tomas, Rutger Warrink and Dianne van de Zande. Without the family Struycken-De Block, in particular Jan-Pieter Struycken, I might have remained ignorant about the fact that ESA not only stands for European Space Agency, which hopefully captures well my gratitude. Introducing me to Hans Koot, urban archaeologist of the Rijswijk municipality, and his team, in early 1998 offered me that tiny, necessary spark. My good friend Oscar Struycken: ever since we met in 1989 you have been there to share numerous long talks, Marillion and trips to Irish pubs, among so many other things. Soon we will meet here again; slàinte mhath!

vii

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:27 UTC All use subject to https://about.jstor.org/terms

The Central Library and the library of the Faculty of Arts, both of the University of Leuven, as well as the Koninklijke Bibliotheek/National Library of the Netherlands have often proven their worth, and I am appreciative to these institutions for their services and facilities. Other institutions I would like to thank are, among others, Gülkent Pide and Το Αθηναϊκόν. Sadly, my grandmother could not witness the completion of my time in Belgium, her country of birth. She may have had only an inkling of what I was doing, and I cherish and remember her continued interest and curiosity about my life in Belgium and in, for her, very distant countries. My family remains curious about the life of an archaeologist, especially when he is abroad. Sherds, sand and spoons are, jokingly, keywords. On a more serious level, though, I continue to appreciate their enthusiasm about what befalls me abroad. My brother Elwin knows all too well what it is like to pursue one’s passion, sometimes seemingly against the odds. Besides being that dear brother, your longtime and enriching love for film — angle, colour, sound — makes one look at things differently and it continues to be catching, and has also further shown me to be (more) selective; life’s simply too short to waste: I wish you and Liesje continued happiness. My parents have unconditionally supported me in the less-than-obvious choice of becoming and especially being an archaeologist. Their interest as well as concern, especially while I was on fieldwork for often lengthy periods of time, meant that we could not spend as much time together as would have been possible under different circumstances. My return to native soil, The Hague, completes a wonderful process of discovery and acceptance. Their unwavering support I cannot remain but be amazed at, which perhaps one day I will come to understand more fully. Needless to say that, even if this text has been explored and gone through almost more than I cared for, all faults remaining, omissions, etc., are my own. The Hague, 23 January 2015

viii

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:27 UTC All use subject to https://about.jstor.org/terms

Archaeopress

Chapter Title: Aims and Methodology Book Title: Once upon a Time in the East Book Subtitle: The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East Book Author(s): Philip Bes Published by: Archaeopress. (2015) Stable URL: https://www.jstor.org/stable/j.ctvr43kch.6 JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected]. Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at https://about.jstor.org/terms

Archaeopress is collaborating with JSTOR to digitize, preserve and extend access to Once upon a Time in the East

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:30 UTC All use subject to https://about.jstor.org/terms

Chapter 12

Aims and Methodology Introduction3

Finally, chapter 7 abandons the chronological and regional frameworks, and views and discusses the exchange of tablewares in the eastern Mediterranean diachronically. The idea behind this study is based on the belief that the collected tableware evidence can contribute to an understanding of socio-economic and -cultural aspects of the Roman (eastern) Mediterranean, and make these more visible. Its innovative character lies in its deconstructing approach, with a prime focus on published catalogues of vessels/fragments and seeing as partially subjective entities each with its own history, yet entities that can thus be broken down into the many (objective) elements that catalogues exist of, certain elements perhaps being more significant to this study than others. By subsequently bringing a selection of these individual elements back together within certain geographical and chronological boundaries, it is thought that the subjective cloak is shaken off, thus offering the basis for a more neutral analysis and reconstruction.

With obvious consideration for its methodological and practical pros and cons, as (ancient) pottery was an integral part of many aspects of daily life, it is considered to be a valuable tool to throw light on ancient society-/ ies. This can therefore be done from a number of angles. For one, studying the manufacture of pottery offers us insights into technological aspects but also people’s choices with regard to shape and decoration, which in turn could tell us something about their actual use during eating and drinking practices. Also, both archaeometrical and archaeological studies make it quite clear that, with varying intensity and length of time, pottery was traded on a variety of geographical and contextual levels. This, then, prompts us to investigate the reasons why, but also the mechanisms through which, this pottery (yet not only pottery) was traded. It is, in general terms, that ecomonic viewpoint which is the thematic focus of this study, more particularly, the late Republican/late Hellenistic and Roman-period eastern Mediterranean.

Despite the fact that pottery lies at the very core of this study, it is hoped and thought that some parts will be of certain interest and relevance to those that are less of an insider. In particular this goes for chapters 4 and 6, wherein the regional data is discussed and contextualised, as well as chapter 7, which offers a synthetic discussion.

This study, then, focuses on explaining the distribution of red slip tablewares (hereafter conveniently referred to as tablewares) between the late Hellenistic and late Roman periods,4 and is structured as follows. This chapter introduces the aims and methodology, followed by chapter 2 in which the material evidence, on which this study is based, is discussed. The evidence for the late Hellenistic and early Roman periods is presented and discussed in chapters 3 and 4, spanning the mid-2nd century BC to the late 2nd century. Chapter 3 presents the collected evidence, in line with the chronological and regional frameworks presented in sections 1.3.5 and 1.3.6 respectively. Chapter 4 seeks to explain this evidence in relation to the underlying factors affecting the developing and changing exchange patterns within the Roman (eastern) Mediterranean. The evidence pertaining to the mid- to late Roman periods, the early 3rd to the late 7th century, is presented and discussed in chapter 5, followed by the interpretation in chapter 6.

1.1 Definition of Terms To avoid confusion, several standardised terms are used in this study: form refers to a vessel that is included in a (published) typo-chronological framework and has, or should have, defined morphological and/or decorative features that make it distinct from other forms, whilst shape refers in more general terms to a certain kind of vessel, for example a bowl or dish. W-/ware, then, refers to a coined and acknowledged class of terra sigillata or red slip tableware with recognised properties of fabric composition, typo-chronology and/or provenance.5 What is meant by tablewares in the context of this study are all vessels assumed to have been used on or in close proximity to the table, though table can, or perhaps even should be applied in a less narrow way to other occasions of eating and drinking. This basically denotes vessels used for the serving and consumption of food and beverages, and includes shapes such as bowls, dishes, serving trays, cups, jugs, as a matter of fact any vessel that is perceived as being fit for use on or in relation to the table. Most of the distinguishing morphological criteria

All dates are AD, unless otherwise noted. The original doctoral research, completed in November 2007 (Bes 2007) and of which this is a reworked version, was carried out within the framework of the ICRATES Project (Inventory of Crafts and Trade in the Roman East), supported by the Fund for Scientific Research-Flanders (FWO research projects G.0152.04 and G.0245.02). It also results from the Belgian Programme on Interuniversity Poles of Attraction (IUAP/V) initiated by the Belgian Federal Science Policy Office, and the Concerted Action of the Flemish Government (GOA 02/02). 4  The nomenclature used here is strongly dependent on tableware chronology. Thus, late Hellenistic starts when the first major red slip tableware, ESA, emerges in the northern Levant c. 150-140 BC. Consequently, late Roman ends when the tradition peters out in the second half of the 7th century, although reduced production and distribution may have continued at places into the early 8th. 2

3 

This partly parallels Schneider’s definition of ‘Referenzgruppe’ (1996a, 129).

5 

1

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:30 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East are not necessarily much different from those we apply nowadays, though certain shapes are peculiar to the period in question. Hellenistic and Roman tablewares to a large extent comprise slipped vessels: before firing, the clay body was dipped into a refined and liquid clay solution that covered the vessel’s surface. The colour of the slip of tablewares in the period under study was mostly red or reddish, indicative of an oxidising firing process, yet tableware vessels with a black- or grey-coloured slip, more common in pre-Roman societies, continued to be manufactured in certain parts of the Roman World. Red slip tablewares of late Hellenistic to late Roman date are generally referred to as terra sigillata or (Late) Roman Red Wares. The former is commonly applied to the late Hellenistic and early Roman period, the latter spanning the early to late Roman, although the distinction is largely artificial. Both terms describe tableware vessels covered by a red or reddish slip. In fact, terra sigillata is mostly an incorrect term ― for one, it is not an ancient term ― yet has grown to represent the glossy, early Roman vessels, in particular the Arretine products. On the other hand, the use and meaning of both terms have become firmly embedded in Roman pottery studies, making it difficult, perhaps even unnecessary, to seek alternatives, even though these may be more correct.

1.2 Aims Before describing the aims in more detail, four key points which form the basis of this study deserve to be introduced briefly. First, the distribution of tablewares, and for that matter, most, if not all other goods, was not driven by a single mechanism. In fact, complex networks of intentions, social relationships, individual-collective and/or private-public interests, opportunism, and so forth were the driving forces behind exchange patterns. At the same time, geographical factors also had their share in determining the direction and intensity of such patterns. A commodity formed part of one particular mechanism, for example, in order for those involved to gain a profit. Yet in many cases, at virtually any point in the distribution chain, an object could have been transferred from one exchange mechanism to another, thereby gaining a new or added function or significance. Secondly, disentangling distribution patterns of tablewares is further complicated when one takes into account the hypothesis that tablewares (as well as comparable items, such as oil lamps and glass vessels) were taken on board as a secondary cargo, riding piggy back, stowed away in the hull, between or on top of the main cargo. The evidence from shipwrecks, however, clearly shows that the secondary cargo-hypothesis does not always apply.7 Although this could imply that the distribution of tablewares may have been less firmly organised, moving along with bulk goods more haphazardly, the patterns emerging from the collected evidence largely argue against this. Thirdly, the data presented in chapters 3 and 5 is presented using several descriptive statistical methods. These views should not be perceived as representing detailed historical precision, but rather serve as an index, providing clues to understand the quantitative, chronological and geographical distribution of tablewares; their interpretation needs to take shape accordingly. Finally, a direct cause-and-effect model between events or developments in the more general historical and socioeconomic frameworks and the quantitative, chronological and geographical distribution of tablewares is treading upon thin ice, and not advocated here. However, a political and administrative entity the size of the Roman Empire was dynamic. As such, it touched on, and interacted with, the different communities and different areas that it controlled. Since material culture formed an integral part of these communities, in theory, then, the archaeological evidence potentially reflects socio-economic change and interaction.

Tableware services were supplemented by so-called thinwalled vessels, with walls of sometimes not more then 1-2mm, generally comprising drinking and small serving/ pouring vessels. Performing the same function are leadglazed vessels, drinking or serving vessels covered with a lead glaze. These were produced in a number of places in the Roman Mediterranean, yet possibly due to their intricate manufacturing process such products never became very common until mid-Byzantine times.6 Despite the desire and necessity to categorise the different classes of tablewares, partly in an attempt to identify services (morphologicallyand chronologically-associated vessels), we may picture a more flexible and ad hoc approach to tablewares in the past, with the composition of a household’s tableware collection being dependent on personal choice or taste, social context, distance to the production centre(s), distance to points of distribution and redistribution, and so forth. And let us not forget tablewares made in materials other than pottery. Tableware assemblages thus must have showed a considerable variety in their composition, and may have grown over time. Fine ware is a term often employed in Roman pottery studies, yet is considered too subjective for proper use here, evoking an image of only the finer red slip vessels of the best quality. This study therefore opts to use the more neutral term tableware(s) throughout, as it allows us to include thin-walled wares and other categories in the discussion which were used for eating and drinking but are not necessarily fine in all respects.

The distribution and consumption of tablewares in the eastern Mediterranean between c. 150 BC-700, as investigated in this study, concerns six classes of tableware manufactured in eastern pottery workshops. In addition, two western classes attained a wide and significant distribution, and are included. The emergence and development of a class of tableware points to considerable investment in its production infrastructure, taking advantage of existing and/or newly developing

6  Thin-walled wares: e.g. Marabini Moevs 1973, Mayet 1975; lead-glazed wares: e.g. Hochuli-Geysel 1977, 2002.

7 

2

Jurišić 2000.

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:30 UTC All use subject to https://about.jstor.org/terms

Introduction exchange mechanisms, in order to market the product. Its disappearance may result from the demise or change of such exchange patterns, whatever the cause or causes, or the increasing importance of substitute products.

gradually and in dispersed ways, from a tradition of cataloguing and describing, even though such information remains essential and needs to be made available through publication.

Following an evaluation of the collected evidence, the geographical, chronological and quantitative dimensions of the eight classes of tablewares are mapped using a regional framework.8 These dimensions assist in drawing the outlines of their regional and/or supraregional distribution and consumption, shifts in which can be charted using a diachronic focus. Regarding their distribution, it will be shown that not all wares behaved similarly. It is especially those tablewares whose distribution spanned the (eastern) Mediterranean that suggest that specific factors were working favourably, illustrating the dynamic nature of ‘the’ economy. This study aims to explain this dynamism against the general political and socio-economic background, by considering the different and developing interests of the imperial government, of military and religious centres (pagan and Christian) as well as of private persons (elite and non-elite). The regional and supraregional distribution of tablewares can also be used to evaluate the interaction between the developing political, socio-economic and socio-cultural landscape and the consumption and use of tablewares. The typological repertoire of late Hellenistic to late Roman tablewares is characterised by quite a number of developments, some profound, some subtle. The collected evidence has the potential to explain aspects of consumer preferences against this developing background. Since the distribution patterns of the eight tablewares under study show considerable geographical and quantitative differences, both internal (manufacturers’ and consumers’ choices and responses) and external (the role of pivotal centres such as Delos and Alexandria) factors are considered, which reflect the debate about the integration of, and interaction between, people and their material culture within the Roman world.

Despite the steady progress in the field of Roman pottery studies, some topics are touched upon only superficially, if at all. Our knowledge of the diversity of fabrics and wares increases quite rapidly yet remains still rather basic for many pottery classes, and with the discovery, excavation and/or publication of known and new sites long-term views are required. Like other categories, such as amphorae, cooking wares and oil lamps, some Roman-period tablewares were produced in massive numbers. Lund recently estimated that the entire output of ESA, one of the most prolific tablewares in the east, approximated 24 million vessels, and the south Gaulish potter Castus (i.e. his workshop) is estimated to have manufactured some 300,000 vessels each year.10 In spite once more of the tentative character of such figures, Lund was at least able to illustrate the magnitude we theoretically have to deal with. More importantly, one could start to comprehend the role tablewares played in the economy and daily life of the Roman Mediterranean. Only a fraction of such (tentative) figures has been excavated and published. This fraction nevertheless represents a formidable amount of published data, which from the outset of this research had to be stored and controlled in a uniform way, to allow careful analysis and interpretation. To this end, a relational database was developed wherein the variety of available information was grouped into logically defined groups. The original ACCESS-database consisted of six interrelated tables, into which each published fragment was individually recorded:11 (1) the PUBLICATION table stores information particular to the publication, such as author, full bibliographic reference, whether it contains a catalogue and/or quantification, and the year(s) wherein the excavation and/or survey was conducted; (2) the LOCATION table stores data pertaining to the site(s) from which material was published. It comprises a wide range of parameters: the ancient name (if known), geographical and topographic information (for example in what provincia or theme the site was located), and connections to ancient land and sea routes; (3) the DEPOSIT table contains information that concerns the archaeological and architectural context. This includes the proposed date range of the deposit, the nature of the site or archaeological context where the material was found, the relative quality of the deposit (open-closed, primarysecondary), the dating criteria other than tablewares

1.3 Methodology 1.3.1 The Published Record and Data Collection The quantity of excavated pottery sherds from the study area is staggering, and one may only shiver thinking about what is yet to be unearthed. Hayes, for one, estimated that some 30 to 40 million pottery sherds are excavated around the Mediterranean each year.9 The accuracy of this figure is irrelevant: it does, however, give an impression of the numbers Roman pottery experts are faced with. The question what to do with such quantities — practically, interpretatively, etc. — increasingly suggests itself. Roman pottery studies appear to be moving away, albeit

10  Lund 2005, 233-234; Hartley and Dickinson, quoted in Lewit 2011, 315. 11  Since the conclusion of this research, the original database has been transformed into a web-based application which other scholars may contribute to and consult, and new data is being added. The research potential is greatly enhanced by building in, along similar lines, the possibility to enter (published) data on amphorae, which will allow us to compare and interpret distribution patterns of both categories.

Regional or supraregional studies figure prominently in the published record, with greatly varying thematic scopes. For pottery-based examples, see Reynolds 1995, 2005a, b, 2010; Bonifay 2002, 2003, 2004, 2005; Pieri 2005; Sodini 2000; Tomber 1993a, 2004; Lewit 2011 for a recent attempt at contextualisation. For other examples, see Safrai 1994; Dar 1999; Drexhage 2007; Wickham 1988, 2005; Morrisson and Sodini 2002. 9  Hayes 2000b, 285. 8 

3

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:30 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East (stamped amphorae handles, coins, lamps, glass, historical information, epigraphy); (4) the CHRONOLOGY table12 recorded the (available) information on the chronology of the deposit and the typo-chronological information of the tableware found in that deposit; (5) finally, the CATALOGUE table stores information relating to the individual tableware fragment or vessel, and includes form, measurements, stamps, decoration, etc.; and (6) the ICRATES table stores the main parameters from these five tables, and thus serves as the database’s nerve system.

and consumption patterns, and their interpretation, become more refined. Through the published evidence ― which encompasses the late Hellenistic to late Roman eastern Mediterranean ― it will be noted that certain wares occur more prolifically in one or more regions or during a certain period than others, or that certain forms are more common. In order to evaluate and determine the possible strengths and/or weaknesses of the published data on which this study is so strongly dependent, ceramic data from three archaeological projects, Sagalassos, Boeotia, and Kinet Höyük, is used.15

The published material is critically approached on an individual basis: each catalogued fragment that belongs to one of the eight major classes is collected in the database and receives a unique entry number. The study here works with the data originally collected, comprising about 23,272 pieces, 15,665 of which are typo-chronologically useable. By the end of 2012 the database held nearly 26,000 entries. It was not possible to include this additional information, yet it is referred to where appropriate. It is safe to say that the information that has emerged or been published since, refines, yet does not negate the observations made here. The database is a tool not merely to store data.13 It is a valuable instrument that can answer queries, or carry out searches, customised on the basis of accurately defined research questions. In any case, it is our conviction that the collected evidence is sufficient to investigate the stated aims.

In so far as quantification is concerned, the database unavoidably contains only part of the published evidence, a situation that does little justice to the importance of, as well as concern and valuable work done on this matter.16 The choice by research projects to present material as a catalogue was preceded by selection criteria that are rarely made explicit. Nonetheless, it was decided to refrain from entering publications into the database in their entirety, which would seriously misrepresent the proportions for certain sites and/or regions. Tel Anafa serves as a good example, where the quantification comprised nearly 24,000 sherds.17 Incorporating this entire quantification would simply result in a greatly unbalanced picture, not for the site, but for the region in question. The regions as defined in section 1.3.6 are represented, as far as was possible, by publications that would ideally reflect the quantity and variety to be used for further interpretation. Malfitana used a somewhat similar methodology in his overview of tablewares in the eastern Mediterranean.18

How does a/the collected total relate to the geographical and chronological scope?14 When one considers Lund’s proposed output of ESA, regardless of its tentative character, clearly this figure seems wildly inadequate to answer any such question. However, one has to bear in mind that this approach offers a starting point, yet does not pretend to solve (all major) problems. Rather, it focuses on offering indications on the way tablewares circulated in economic and social spheres of interaction. This is echoed, in our opinion, in the quantified patterns that are discussed in chapters 4, 6 and 7. There, the movements observed in the data should not be seen as reflecting the finest historical detail. Rather, diachronic, geographic and quantitative shifts in the distribution patterns of tablewares should be seen as indicating change. Also, this study is aimed at a regional comparative character concerning the circulation of tablewares, not on the volumes that circulated. What then was the nature of these changes is of great interest and importance, and forms part of the interpretative part of this study. Finally, it is our conviction that in an ongoing process of collecting (new sites and) data, the distribution

This innovative approach is in accord with new directions which the study of Roman pottery is taking. Since the 1970s an important broadening of the discipline has been taking place, characterised by [1] a growing number of, and diversity in, publications, not only contributing to a process of densification of distribution patterns, but also allowing a detailed study of the typo-chronological framework and, where necessary, the making of adjustments;19 [2] quantified pottery studies, providing a better methodological basis for the reconstruction of exchange patterns within the Roman World;20 [3] a steady growth 15  Sagalassos (Turkey): Sagalassos Archaeological Research Project, directed by Jeroen Poblome as of 2014 (University of Leuven, Belgium); Hyettos, Koroneia, Tanagra, Thespiae (Boeotia, Greece): The Ancient Cities of Boeotia Project, directed by Anthony Snodgrass (University of Cambridge) and John Bintliff (University of Edinburgh, Scotland); Kinet Höyük (Turkey): directed by Marie-Henriette Gates (Bilkent University, Ankara, Turkey). Since 2010, the author has become involved in other projects, notably surveys on the islands of Skyros and Zakynthos, and the regions around and west of Patras, in the northwest Peloponnese, as well as excavations in Limyra (Turkey) and Horvat Kur (Israel). Observations made in the course of these projects are included where appropriate. 16  Berlin 2006, 4-11, 21-23; Jackson and Tidmarsh 2011. 17  Slane 1997. 18  Malfitana 2002. 19  Fulford 1984; Bonifay 2004. 20  Riley 1975, 1979; Hayes 1976a, 1977; Tomber 1993a; Kingsley 1999. See Peña 2007b for methodological considerations on quantification, and

12  This table was regarded as redundant and therefore not included in the new web-based application. 13  A preliminary catalogue is available for the pottery found at Troia: http://classics.uc.edu/troy/grbpottery/, and the Pylos Regional Archaeological Project (PRAP) also made the pottery database available on the internet: http://docs.classics.uc.edu/fmi/xsl/prap/pottery_list. xsl?-db=PRAPPottery&-lay=Single&-skip=21930&-max=25&-findall. Other projects have also in one way or the other made information about their pottery available on the internet, for example the excavations at Aqaba. 14  Willet and Poblome 2011.

4

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:30 UTC All use subject to https://about.jstor.org/terms

Introduction in regional syntheses;21 [4] a growing attention for, and understanding of, the distribution into and use of pottery in non-urban and -coastal contexts;22 and [5] archaeometrical analyses, comprising a number of methods with which not only the mineralogical and chemical composition of a fabric can be determined, and consequently a possible provenance proposed, but also serving as an approach to technological matters.23

built into the database. To this end, the following works are used: for ITS, the Conspectus Formarum; for stamps on ITS, the OCK; for ESA, ESB, and ESD, the Atlante; for ESC, Meyer-Schlichtmann’s pergamenischen Sigillata and, occasionally, the Atlante; and Hayes’ LRP and A Supplement to LRP for Çandarlı Ware (ESC hereafter, cf. infra, 2.2.3), ARSW (combined with Bonifay’s Études), CRSW/LRD and PRSW/LRC.30

All these developments are part of a broader trend of conceptual expansion.24 More and more scholars employ the broad palette of research topics and methodologies in order to investigate the role of Roman pottery in local, regional and supraregional patterns of distribution, and find explanations on whether and how pottery (including tablewares) relates to the contemporary political, economic and socio-cultural landscape.25

1.3.3 Descriptive Statistics Archaeological data can be stored and studied in many different ways. One particular method for the quantification of tablewares was developed specifically for ARSW,31 but can be applied to other categories of tablewares, or better still also, to a wide range of material culture. It can be used to assess the quantitative development of a ware through time, or how that ware relates quantitatively to other wares. The methodology requires two parameters: a form is defined (or a range of forms), and each form has a chronological range with a lower and upper limit, the period during which the form/shape is assumed to have been produced and/or distributed.32 A specimen of a specific form has, theoretically speaking, an equal chance of being produced in any of the years of its chronological range. Fentress and Perkins thus took the year as their basic chronological unit. They illustrated this by using a form with a date range of 100 years, thus the vessel had a theoretical chance of 1/100 (or 1%) of being produced in either one of the years of its date range. By multiplying the yearly ‘chance’ by the number of specimens of that particular form found or identified, this results in a diachronic ‘volume’ of that form, and can lead to a diachronic volume for each form by replicating the exercise for all other forms identified. This can then also be replicated for other wares. These volumes subsequently support further study and interpretation. This enabled the authors to ‘count African Red Slip Ware’ by quantifying the forms of ARSW from several archaeological projects.33 Considering the nature of the collected evidence and the objectives of this study, this methodology is suitable for application to the collected data.34

1.3.2 Typo-Chronologies To an excavator, tablewares are first and foremost seen as chronological markers, secondly as objects that reflect ancient socio-cultural and -economic activity. This notion should obviously include other categories of (ceramic) products besides tablewares.26 Fortunately, the eight wares that are discussed in this study can be used because for each one or more typo-chronologies have been published. Since tablewares are important for the aforesaid reasons, the accuracy of a typo-chronology is essential, albeit liable to change due to ongoing research. For a more detailed discussion of these eight wares, see chapter 2. Over the years, a number of suggestions have been made regarding the chronology of certain forms and even wares, ranging from argued cases to rather loose suggestions based on less palpable observations.27 Although these may be considered, the decision was made to remain faithful to the existing typo-chronological frameworks, for the published evidence is based upon these. Yet Bonifay’s revised typochronological study for part of Hayes’ original ARSW typochronology is a case in point, and a comparison between the two based on the collected evidence is discussed elsewhere.28 About 95% of the database concerns the eight major wares discussed in chapter 2.29 Their respective typo-chronological frameworks allows us to attain, among other aspects, a much-desired degree of uniformity to be

However, an adaptation was made out of both practical and methodological considerations. Using the year as the basic chronological unit suggests a very secure knowledge of the chronological ranges of forms, which in most cases remains debatable. Instead of using 1-year units, it was

a case study on ARSW from Rome. 21  Reynolds 1995; Bonifay 2004; cf. supra, n. 8. 22  E.g. Orssaud 1980; Rossiter and Freed 1991; Harper 1980, 1995; Slane 1997; Rautman 2003; Lund 2006a, b. 23  For instance Schneider 1995, 1996a, b; Mackensen and Schneider 2002, 2006. 24  Wickham 1998; Sodini 2000; Bonifay 2003, 2005; Tomber 2004. 25  Poblome et al. 2006. 26  Peacock 1982a; Tomber 1993, 2004; Bonifay 2003, 2004, 2005; WardPerkins 2001; Wickham 2005. 27  For instance, Abadie-Reynal 2005b and Sieler 2004 summarising evidence arguing in favour of the revision of date ranges of certain forms. See Reynolds 2010 for ESA continuing well into the 3rd century; n. 126 for 3rd-century ESB. 28  Bes and Poblome 2009. 29  In addition to the eight major wares, further tablewares entered into the database include Gaulish Sigillata, Pontic Sigillata, thin-walled ware(s), lead-glazed ware(s) and ERSW.

ITS: Ettlinger et al. 1990; ITS stamps: OCK 2000; ESA, ESB, ESC, ESD: Hayes 1985a; ESC: Meyer-Schlichtmann 1988; ESC (Çandarlı Ware), ARSW, CRSW (LRD), PRSW (LRC): Hayes 1972, 1980; ARSW: Bonifay 2004. 31  Fentress and Perkins 1988, 205-214; Fentress et al. 2004, 147-162. 32  The chronological range can be based on other pottery classes, for instance stamped amphora handles or amphorae more generally, oil lamps, cooking ware, but also numismatic evidence, glass, historical circumstances, epigraphy, stratigraphic sequence, architectural setting, the style of mosaics, and so forth. Note, however, that the chronological ranges for production, distribution and consumption/use need not have run parallel: Peña 2007a. 33  Fentress and Perkins 1988. 34  Lund 1996b. 30 

5

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:30 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East

Figure 1. An example of the method of distribution by Fentress and Perkins 1988 (© Philip Bes/ICRATES Project).

Figure 2. Line chart showing the development, in absolute numbers, of the collected evidence per 15-year interval (n=15,665) (© Philip Bes/ICRATES Project).

a detailed knowledge of the chronological production range of a form or a ware, which, regrettably, is not the case for most tablewares of eastern manufacture, the exceptions being ESC and especially SRSW.36 Here, the pitfall of overinterpretation looms, which should — given the current state of affairs of the production infrastructure — be avoided at all costs. At the same time, it is felt that broader intervals (25, or even 50 years) might obscure some of the detail of the data. Therefore, Figure 2 show the data set for 15-year intervals.

decided to use 15-year intervals.35 This not only overcomes the problem of indecipherable charts, with the consequent tendency to over-interpret, as such charts represent a very detailed chronological and quantitative dimension: it also deals with the issue of the relative uncertainty of the chronological ranges of forms. As a 15-year interval appears a fairly good standard, it should be stressed though that if a form falls into a 15-year interval for only five years, it follows that it is attributed to that interval for only one-third. This practice closely follows the available typo-chronologies, yet by adopting it one avoids assigning a form to years during which it was (hypothetically) not produced and/or distributed. An example illustrating this method is shown in Figure 1. The 15-year interval implies 35 

Because this methodology requires the lower and upper limit of a form’s chronological range to be known, forms 36  ESC: Meyer-Schlichtmann 1988; Poblome et al. 2001b; SRSW: Poblome 1999.

Following editorial work on Early Italian Sigillata: Poblome et al. 2004.

6

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:30 UTC All use subject to https://about.jstor.org/terms

Introduction lacking one or both had to be omitted from the data set. Whereas all typo-chronologies contain forms whose chronological range is uncertain or only partly known, this actually poses no serious problem since often the quantity of such forms is (very) small. This is not the case with a number of stamps on ITS, though confusion can be forestalled by identification of the form using the Conspectus Formarum. The case is, however, more critical for ESB, as a considerable number, especially forms of Hayes’ ESB-I series, have no chronological range. Consequently, these forms are not included in this study.

ESC for the late Hellenistic-early Roman and the mid- to late Roman periods, respectively, is divided proportionally, based on the identified entries. 1.3.4 Data: Problems and Limitations The collected tableware evidence results from a long tradition of research. Although each publication has its value, needless to say, they vary greatly in quality. Each publication is obviously a product of its time. The content can also have been determined by the author’s or excavator’s agenda, or may have been forced to reconcile with certain circumstances. A critical study of the published data is nevertheless required in order to bring it to its fullest potential.

This methodology involuntarily also ignores the fact that a form may have been more popular during part of its chronological range. Although we may suppose this to have been the case, it is a rather grey zone of knowledge, often perception rather than fact. An important way out of this problem is the availability of high-quality, closed and well-dated archaeological deposits from comparable sites (for example major harbours, or from sites destroyed or abandoned suddenly which were not or only sparsely occupied, i.e. disturbed, afterwards) and contexts (for instance cistern fills and mosaic beddings). The popularity of a form can be tracked down more closely using such narrowly-dated deposits.

Appendices 1 to 4 offer geographical, typological and quantitative information concerning the distribution maps and charts used throughout: appendix 1a-b shows all the sites and surveys from which data was collected, and all other sites mentioned in the text, respectively; appendix 2 lists the sites attributed to each region (cf. infra, 1.3.6); appendix 3a-b lists the absolute and relative quantities per ware, per region; and appendix 4a-b captures the quantitative data for each of the nine chronological phases (cf. infra, 1.3.5).

What are some of the major problems encountered? First, the quantitative character of the publication. The custom of discarding undiagnostic (body) sherds is less and less common, but was, for instance, not exceptional into the 60s of the 20th century, and possibly even later in the case of the eastern Mediterranean. This mostly needs to be seen as an observation rather than a qualification: it shows a developing discipline. In basically all publications the data is presented as a catalogue, a selection composed arbitrarily or based on a set of criteria, whilst other publications contain both a catalogue and a quantified overview. However, only a handful of publications actually indicate whether their catalogue is representative of the material that was excavated, and without a quantification any form of control is rendered impossible. Creating a methodology capable of testing the nature of a ceramic catalogue seems farfetched. In addition, the representativeness of a catalogue is highly prone to circular reasoning as well as individual interpretation. Therefore, this study is firmly rooted in the belief of the intrinsic value of an overall approach. By collecting the material en masse (173 excavations and/or surveys were studied, see appendices 1a, 2) within a broad chronological and geographical framework, thereby taking into consideration the non-representative character of the material, the finer points will, perhaps, not be overlooked; although important, these are of less concern for this study. Rather, it aims to outline certain basic patterns that are undeniably present in the material. The matter, however, may also be reversed. Why would a large collection of published tableware fragments not be generally representative of ancient patterns of exchange? It is believed that context-, site- or publication-specific issues are smoothed away because of the geographical and chronological scope of the study. This of course applies specifically to the macroeconomic or supraregional level: after all, a site’s historical character is expressed in its archaeological record, distorted in most cases, and site-specific factors (location, for one) are taken into account in further interpretation. This archaeological record includes pottery, the interpretation of which gains added significance when compared with other sites and areas.

Finally, with regard to the share of ESC in each pie chart, throughout this study the quantitative data for unidentified

Second, the published pottery itself poses several problems and therefore needs to be critically judged. A first problem

Within the chronological and geographical framework (cf. infra, 1.3.5-6), an important question is how to accommodate forms whose date range falls roughly into two phases. This matter can only be taken into account to some extent. A form is attributed to a phase when the larger part of its standardised date range falls into that particular phase. Although this ignores the possibility that a form could also have arrived at a site within any year of its date range, the maps for the nine phases thus serve as (visual) guides for the distribution of the different wares. Further support is sought in the archaeological record, even if only a few relatively good and narrowly-dated deposits are published. Concerning the data presentation on the regional maps, even though a site has less than 100 fragments/entries, the relative quantities are presented in a pie chart. Fully aware that this approach does not adhere to statistical principles, it is thought that in this way each piece of data is at least equally represented visually; the text and appendices proper should then serve to clarify each map’s quantitative background.

7

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:30 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East

Figure 3. The nine-phase chronological framework used in this study (© Philip Bes/ICRATES Project).

that springs to mind is the delicate relationship between the dating of forms (and by extension, wares) and the dating of a deposit. For example, understanding the distribution pattern of the earliest forms of ESA, significant though this is, is possibly hampered if the dating of a deposit relies solely on the typo-chronological dating of these early forms: it becomes something of a self-fullfilling prophecy. One might also be confronted with the misidentification of a form, which can be the case for instance with small rim fragments of ESA forms 3 and 4. Not only a form, also a ware may be misplaced, or not identified at all.37 Also, certain illustrations-identifications raise serious doubt about the attribution. For one, we may be dealing with a locally- or regionally-made vessel inspired by a prototype in a commonly distributed ware. The database accommodates alternative suggestions when a more appropriate form can be recognised. The proportion of misinterpreted forms and wares is considered to be small, likely not exceeding an acceptable 1-2%. Striving for a high degree of uniformity, and attempting to have information as accurate as possible, is essential since all faults work through in the analyses and potentially influence the interpretation of the data. Something else that is occasionally encountered is the persistent use of terms such as ‘Samian’ and ‘Pergamene’. These can cause confusion, as both are reminiscent of Zahn’s ‘Pergamene’ and ‘Samian’,38 terms that are, in the main, no longer acceptable.

Catalogues of pottery can occasionally be highly confusing, or simply composed erroneously due to whatever cause.39 Sporadically, illustrations also pose problems because of their quality, thus hampering identification,40 or simply because of the quantity, as with over 1200 drawings of oftentimes small fragments of ESA.41 Perhaps a basic set of parameters needs to be more consistently published, in order to approach certain research questions.42 With software technology increasing at a rapid pace, one could favour pottery (and other artefact) catalogues or quantifications being published digitally.43 This would make available large numbers of artefacts that would otherwise not have been possible, or at least too costly, thus increasing research potential. 1.3.5 The Chronological Setting The archaeological evidence for the production and distribution of red slip tablewares as presently understood imposes the chronological limits of this study. For centuries, pottery with a (partly) red surface existed in many regions, before a fully red slip tableware began to be produced on a large scale somewhere in the northern Levant in the mid-2nd century BC, developing from an already existing framework of production.44 The lower For an example, Harrison and Hayden 2005, 57-79. See for example the illustrations in Adamsheck 1979. 41  Vanderhoeven 1989. Some profile drawings resemble ESD much more than ESA. 42  Poblome et al. 2006, 564-565. 43  See n. 13. 44  Jeroen Poblome, pers. comm.: unpublished chemical analyses by the Centre for Archaeological Sciences (CAS, University of Leuven) of early and mid-Hellenistic colour-coated tablewares and ESA from Kinet 39  40 

E.g. Hayes 2001, 278. For modern use of these terms in the east, see Wintermeyer 2004; see also Slane 1991a, referring to British archaeologists’ use of the term Samian.

37  38 

8

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:30 UTC All use subject to https://about.jstor.org/terms

Introduction

Figure 4. Map of the eastern Mediterranean showing the regional framework used throughout the text; each dot represents a site or area/survey from where data was collected (© Philip Bes/ICRATES Project).

limit of the mid-2nd century BC is established because it was at that time that the production process of slipped tablewares underwent a technological change of lasting importance: the shift from black to red slip, though red and/or brown slips were already favoured in for example parts of Asia Minor.45 This technological and cultural evolution came to determine the production and use of tablewares for the next nine centuries or so.46 The upper limit falls towards the end of the 7th century, when the large-scale production of red slip tablewares dwindled, though production (and distribution?) may have continued on a reduced, regionalised scale into the late 7th, if not the early 8th century.47 That being so, the topic proper dictates the chronological boundaries of this study: the period that began with the inception of terra sigillata in the mid-2nd century BC, right up to its disappearance in the late 7th, early 8th century, even if for the Roman imperial period the term red slip ware is generally used/preferred, a term

which nonetheless reflects the same ceramic product and its manufacturing procedures. In order to facilitate the discussion of the collected evidence, a nine-phase chronological framework was created (Figure 3; all dates are c.): the phases are defined when major changes occurred in the quantity of wares or the (dis)appearance of existing or new wares. Thus, the collected evidence directs this framework. The boundaries between these phases are not determined by a historical framework, though as will be discussed, the phasing does coincide with certain developments. As with the geographical framework, these nine phases are intended to facilitate the presentation and discussion of the collected material, and should not act as a directive, rigid framework. 1.3.6 The Geographical Setting The eastern Mediterranean is a vast area with a turbulent history, covering the modern countries of Greece, Turkey, Syria, Lebanon, Israel and the Palestinian territories, Jordan, Egypt and Libya. This may evoke the thought that the age-old distinction between west and east is yet again sustained by this study. This has at least two obvious reasons. The first is provided by the substantial quantity

Höyük show that these were made from the same clay (group). 45  Van der Enden et al. 2014, 85, fig. 3. 46  Élaigne 1999. 47  Hayes 2007, 435-436; Bonifay 2003, 128, 2005, 570 (ARSW); Armstrong 2007, 20, 24-25.

9

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:30 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East of published evidence for the eastern Mediterranean. Secondly, which is less of a practical motivation but historically prompted, the Roman Empire actually continued in the east. Large parts were subdued by Persian invasions in the 6th and early 7th century, followed by Arab incursions in the course of the 7th century. A relative caesura was established between west and east around the late 3rd century. Ties were not completely cut off: for instance certain legislative frameworks and institutions were taken over by ‘barbarian’ tribes,48 and the 6th century saw Justinian’s reconquest of large parts of the former western Empire. In the east, economic activities beyond any doubt continued, and perhaps even grew; yet, at the same time, the western Mediterranean was certainly not a dilapidated no man’s land.49 A great deal of work has been done on tableware finds50 and specimens of Riley’s Late Roman Amphora package from the east have been identified at sites such as Carthage, Marseille, Rome, and even Britain.51 The distribution of LRC in the west,52 and that of ARSW and African amphorae in the east are testimonies of a continued economic integration of west and east.

A grouping of sites according to the early Roman provincial framework, and for the mid- and late Roman periods following Diocletian’s system of themes and dioceses and Hierokles’ list of cities in his Synekdemos respectively,53 was ultimately dismissed in favour of an artificial geographical framework. This framework is principally based on the available evidence, although obvious cases such as Cyprus and Crete were taken into account. In this way 15 regions were ‘created’, illustrated in Figure 4. Appendix 2 lists, by region, the sites and surveys from which data was collected; see also appendix 1a. As with the chronological framework, this offers a flexible framework within which, at any given point, these regions can be further subdivided, or boundaries shifted. Suggestions for such are given below. This approach also provides an opportunity to investigate a possible relation between political-administrative, geographical and/or other boundaries on the one hand, and the distribution of pottery and other goods and commodities on the other. The order of presentation of the regions in chapters 3 and 5 is clockwise, starting with Africa-Cyrenaica and ending with Cyprus.

See e.g. Whittaker 1983; Liebeschuetz 1997, 2001. Whittaker 1983; Reynolds 1995; Wickham 1998. 50  See e.g. Berti et al. 1970; Carandini and Panella 1973, 1977; Tortorella 1986, 1987; Mackensen 1993; Bowman 1996; Campbell 1996; Bonifay 2004. 51  See Riley 1979 for his Late Roman Amphora package. For the distribution of some of these, see Kingsley 2001, 51-55, esp. 53-54 (fig. 3.4); Decker 2001, 76-77; Karagiorgou 2001. 52  Reynolds 2005a, 486, map 10.

53  Talbert 2000, maps 101-102; Hierokles may have written his travel guide around 535.

48  49 

10

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:30 UTC All use subject to https://about.jstor.org/terms

Archaeopress

Chapter Title: Definition of the Topic Book Title: Once upon a Time in the East Book Subtitle: The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East Book Author(s): Philip Bes Published by: Archaeopress. (2015) Stable URL: https://www.jstor.org/stable/j.ctvr43kch.7 JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected]. Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at https://about.jstor.org/terms

Archaeopress is collaborating with JSTOR to digitize, preserve and extend access to Once upon a Time in the East

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:33 UTC All use subject to https://about.jstor.org/terms

Chapter 2

Definition of the Topic ad Orontem (Antioch hereafter), casted doubt on the actual origin of Pergamene at Pergamon. In his contribution, he mentions the prolific presence of ‘Pergamene’ at Antioch and suggested an eastern provenance (hence the quotation marks), namely Alexandria.58 His study should be seen as the first full treatment of the topic; regarding ‘Pergamene’ he was able to distinguish an early (late Hellenistic) and a late (early Roman) phase, though he thought that production might have continued into his ‘Middle Roman Period’.59 Waagé was also the first to construct a typo-chronological framework, in spite of the generally poor stratigraphy at Antioch. Waagé also pioneered in paying attention to tablewares of the late Roman period. Interestingly, Waagé did adopt a more objective nomenclature for these wares, and defined ‘Late Roman A’, ‘B’, and ‘C’, and two wares of minor quantitative importance, ‘Late Roman D’ and ‘E’.60 Two further steps, which mainly offered additional evidence for the distribution of ‘Pergamene’, resulted from excavations at Tarsos and Athens.61 Jones, working on the Tarsos pottery, in fact suggested local involvement in the production of ‘Pergamene’.

2.1 History of Research 2.1.1 Pioneering Explorations The study of terra sigillata has come a long way over the past century or so. Dragendorff was the first to attempt structuring the study of terra sigillata, and he recognised that production had taken place in several regions of the Roman Mediterranean.54 Although nowadays his brief study is of little archaeological importance, it nevertheless still contains some valuable observations: that, for instance, Greece may not have been a major producer of red slip tablewares, and thus that this kind of pottery was largely imported. He also noted the morphological and contemporaneous connection between ITS and eastern products. This set the tone for subsequent studies. Not until Kenyon’s classificatory system for Eastern Sigillata was published in 1957 (cf. infra, 2.2.1), did such a system exist. However, in the decades prior to 1957 essential building blocks were uncovered. The first of these was Zahn’s contribution on pottery excavated at Priene, in southwest Turkey. Zahn distinguished two categories of red slip wares, ‘Klasse A’ and ‘B’. He was the first to put forward the idea that Italian potters or manufacturers founded a branch workshop in Asia Minor, an idea based on the stamps of his Klasse A that read C. SENT(IVS). Following Pliny’s list of places of manufacture, he suggested Samos was where the origins of his Klasse A (Samian) were to be sought. Zahn wondered whether his Klasse B could designate Pliny’s reference to Pergamon; indeed, the production of red slip tablewares at Pergamon had been suggested by Conze in 1903.55 Despite the endurance of some of his observations, Zahn did little to construct a typo-chronological framework. In addition, a third category of Eastern Sigillata was identified following the discovery of wasters and production implements at Çandarlı (ancient Pitane) in 1911.56 For decades to come, the Çandarlı finds remained the only archaeological evidence for the manufacture of tablewares in the east.

2.1.2 The Current Framework A major change regarding the study of ‘Pergamene’, ‘Samian’ and Pergamene Sigillata came with Kenyon’s treatment of the pottery excavated at Samaria-Sebaste.62 Kenyon applied a framework which was as objective as possible, based on macroscopic identification, and simply named the three classes ESA, ESB and ESC, regardless of their (supposed) provenance. Such a classification can be used until the actual production facilities are determined through archaeological and/or archaeometrical research. A decade later, Hayes added a fourth, which he provisionally labelled Cypriot Sigillata, based on the then known distribution pattern.63 In 1978, Rosenthal proposed to rename this ESD.64 In 1985, Hayes published typochronologies for all four major Eastern Sigillatae, drawing upon the lion’s share of the then published evidence, as well as a considerable amount of unpublished evidence.65 Subsequently, based on a number of cistern fills excavated in the urban centre of ancient Pergamon, MeyerSchlichtmann constructed a detailed typo-chronological framework for his pergamenischen Sigillata, or ESC.66

All three authors, Zahn, Conze and Loeschke, paid little attention to the construction of a typo-chronological framework. Their work did see some following in the first decades of the 20th century. Heberdey, Oxé, Technau and Waagé57 all identified one or more of these wares at Ephesos, Athens, Samos, and Athens respectively. Waagé, through his study of the pottery excavated at Antiocheia

Waagé 1948. Waagé 1948, 18-28 (early), 32-38 (late). For Waagé’s ‘Middle Roman Period’, see 39-42. 60  Waagé 1948, 43-58. 61  Tarsos: Jones 1950; Athens: Robinson 1959. 62  Kenyon 1957. 63  Hayes 1967. 64  Rosenthal 1978, 18-19. 65  Hayes 1985a; also Hayes 2008a. 66  Meyer-Schlichtmann 1988. 58  59 

Dragendorff 1897. Conze 1903. 56  Loeschke 1912. 57  Heberdey 1906; Oxé 1927; Technau 1929; Waagé 1933. 54  55 

11

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:33 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East More recently, the production of tablewares in the area of Pergamon was corroborated through the excavation of the remains of a number of pottery workshops in the Ketios Valley, east of Pergamon.67 Evidence for the manufacture of a fifth Eastern Sigillata was discovered in the late 1980s at Sagalassos, in southwest Asia Minor.68 The properties of the production of SRSW are now, following interdisciplinary efforts, well-established. Investigations over the past decade have largely focused on the production infrastructure in the so-called Potters’ Quarter, and new excavations in the years 2009-11 have greatly enhanced our understanding of the workings of pottery manufacture at Sagalassos in the late Roman period. In fact, ESC but above all SRSW are the only eastern red slip tablewares for which evidence of production is attested. Finally, pottery found at Tel Anafa in northern Israel, forms the most recent major contribution to the understanding of ESA, concerning typo-chronology, clay composition and distribution.69

the earliest scholarly attention, associated as these were with the heyday of the Roman Empire. This echoes the tendency of also this discipline to focus on the more representational aspects of the archaeological record. Fortunately, mentalities and minds have broadened over time. 2.2 Definition of the Eight Wares 2.2.1 Eastern Sigillata A (ESA) History of Research In his discussion of the material excavated at Priene during the years 1895-8, Zahn was the first to distinguish a ware with a distinctive pale (buff) clay and red-brown slip (Figure 5).74 He suggested that the origin of this ware was to be sought in Pergamon, as XAPIC-stamps were also found on tableware vessels from Egypt and Pergamon. However, of greater importance to Zahn’s argument was Pliny, who in his Naturalis Historiae mentions Pergamon as one of several important centres for the manufacture of tablewares. Hence, this ware was coined ‘Pergamene’.75 Zahn did express a certain doubt regarding his observations concerning the fabric colour, as he pointed out the similarity to sherds found at Puteoli. This made it clear that the finds from Pergamon were part of a much wider picture.

As far as tablewares of the late Roman period are concerned, it was Hayes’ seminal treatment of these wares that provided, for the first time, a detailed discussion of their typo-chronological framework and distribution. Although Waagé had already used the pottery from Antioch to attempt to produce such a discussion, because of the scarcity of published evidence at the time, it remains a somewhat isolated contribution. Hayes built upon Waagé’s work, which is indicative of its basic significance, distinguishing three major Late Roman Red Wares: ARSW, LRC and LRD, classes that Waagé had already identified, be it partly under different names; LRC was later renamed PRSW.70 His work gained wide adoption throughout the Mediterranean, and remains a seminal work till this day. Although LRC and LRD remained largely unexplored until more recently, ARSW received great attention, especially following large-scale excavations at Carthage in the 1970s and 1980s. Since then, much work has been done to lay bare the characteristics of the typo-chronological framework, the regional production infrastructure, and the clays used.71 Bonifay’s recent synthesis on pottery production in Tunisia provides the widest approach possible, bringing together not only ARSW but also amphorae, lamps and building ceramics into a regional model of pottery production.72

The American excavations in the Athenian Agora, initiated in the early 1930s, provided vast quantities of pottery with new opportunities for study. Notwithstanding Waagé’s emphasis on the tentative character of Zahn’s hypothesis, he closely followed the latter with regard to his ‘Pergamene’ ware. In addition, Waagé drew attention to the similarities between vessels from Çandarlı published by Loeschke, and his material from the Athenian Agora concerning fabric, vessel morphology and stacking technique.76 Given the proximity of Çandarlı to Pergamon, Waagé accepted this as evidence for Pergamon being the production centre of this ware. Where Zahn did not venture to discuss any chronological aspects, Waagé pointed out that the earliest fragments might belong in the 2nd century BC, and certain profiles hinted at a terminus ante quem of c. 100 for the end of the ware. Waagé would greatly elaborate on his own observations, and those of others, in his study of the material excavated at Antioch during the 1930s.77 This enabled Waagé, more so than regarding the Athenian material, to expand on the identification of wares and their typo-chronological frameworks. Waagé retained the term ‘Pergamene’ for one of these, although he was amenable to any critique against the use of such terms, and open for someone to come up with a good alternative. More importantly, he

The study of Eastern Sigillatae and Late Roman Red Wares has come a long way over the past century. That prior to Hayes’ LRP and A Supplement to LRP73 such wares are sparsely represented in the available literature, except mainly for Waagé and Jones, probably needs to be seen in the context of that time. It is surely no coincidence that late Hellenistic, but above all early Roman (moulded) red slip tablewares were exactly those wares that received Poblome et al. 2001b. Poblome 1999. 69  Slane 1997. 70  Hayes 1972, 1980. 71  For example Bonifay 2004; Peacock et al. 1990. 72  Bonifay 2004, also 2002, 2003, 2005. 73  Hayes 1972, 1980. 67 

Zahn 1904. Zahn 1904, esp. 447-449. Waagé 1933, esp. 285-286 (see for example no. 90, identified as Çandarlı Ware (ESC) on 289). 77  Waagé 1948, esp. 18-28 (Hellenistic ‘Pergamene’), and 32-38 (Early Roman ‘Pergamene’ pottery).

68 

74  75  76 

12

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:33 UTC All use subject to https://about.jstor.org/terms

Definition of the Topic

Figure 5. Late Hellenistic ESA (predating c. 60-50 BC) from the excavations at Kinet Höyük (Hatay, Turkey), directed by Marie-Henriette Gates (Bilkent University, Ankara) (© Philip Bes).

now renounced Pergamon as the place of manufacture for ‘Pergamene’ because of the fabric as well as its distribution pattern, and pointed out its common occurrence in the Levant.78 Following a shift to shapes strongly influenced by ITS during the reign of Augustus, Waagé advocated the creation of two morphological phases. Based on the Antiochene material, Waagé did not allow his Hellenistic phase to pre-date the 2nd century BC (Waagé’s ‘Late Hellenistic Red’ (‘Hellenistic Pergamene’)), whereas he remained vague in assigning a clear end date. Rather, he considered the period between c. 10 BC-15 as transitional, during which new shapes could develop under the influence of ITS (Waagé’s Early Roman ‘Pergamene’). Consequently, the shapes of his Early Roman phase of ‘Pergamene, which ran into the first half of the 2nd century AD’ were fully developed by the second quarter of the 1st century.79 Finally, Waagé suspected that production continued in his ‘Middle Roman period’, the 2nd and 3rd centuries,80 but it appears, however, that Waagé now no longer spoke of ‘Pergamene’, whose quality by now was deteriorating. It must be emphasised that Waagé relied heavily on numismatic evidence and external parallels in constructing these phases, greatly due to the stratigraphic conditions the excavators encountered.81

In her publication of the pottery excavated on the mound of Gözlü Kule at ancient Tarsos, Jones closely followed Waagé’s typo-chronological framework.82 Unfortunately, here also the stratigraphy was of such a nature that it did not allow elaboration on the chronological framework for this and other wares, despite large quantities of ‘Pergamene’. Jones concurred with Waagé that ‘Hellenistic Pergamene’ first appeared around the mid-2nd century BC, although she regarded the term ‘Pergamene’ as unsatisfactory.83 Like Waagé, Jones distinguished the ‘Roman Pergamene’ as influenced by western terra sigillata, whereby Jones stressed that the vessels did not mimic their prototypes, but are rather derivatives of, or shapes inspired by, ITS. Jones suggested that the production of ‘Pergamene’ represented a koinè, a cultural region or, in this case, a regional production framework sharing the same or similar characteristics in terms of tableware manufacture. She tentatively considered that Tarsos might have played an active role in this koinè as regards the production of ‘Roman Pergamene’, with production continuing during the 2nd and 3rd centuries. A profound change came with the publication of the material excavated at Samaria-Sebaste during the years 1931-3.84 Here, Kenyon threw overboard the use of terms such as ‘Pergamene’ and ‘Samian’ in favour of a more

Waagé 1948, 20. Waagé 1948, 36. 80  See now Reynolds 2010, 90. 81  Waagé 1948, 2. The archaeological agenda was greatly determined by literary sources. 78 

Jones 1950. New excavations on Gözlü Kule are carried out since 2001, by Boğaziçi University. 83  Jones 1950, 176, n. 80. 84  Crowfoot et al. 1957; Kenyon 1957; Crowfoot 1957.

79 

82 

13

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:33 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East objective classification, until the archaeological evidence of production workshops would supplant it. Vigorously arguing against the use of terms such as ‘Pergamene’, she proposed the term ‘Eastern Sigillata’, which contained both the kind of pottery and the general provenance. This was followed by ‘A’, ‘B’, etc. for each distinctive ware; thus were created the classes of ESA, ESB and ESC.85 Noteworthy about this publication are the two different opinions regarding the chronology of ESA. Kenyon considered that ESA appeared c. 100 BC and continued into the 1st century, and that it was manufactured in more than one place, whereas Crowfoot believed it appeared during the 2nd century BC and continued into the 3rd century.

to BSP was, more recently, summarised as follows: ‘Stratified evidence from the excavations at Tel Kedesh and Tel Anafa, both located about 30 km north-northwest of Gamla, provides precise chronological parameters for the beginning date of both wares. At Tel Kedesh, BSP vessels occur on the latest floors of the large administrative building that was damaged and temporarily deserted c. 144 BCE [...]. Not a single fragment of ESA appears on these floors, however, thus indicating that by the middle of the second century BCE BSP production had begun but ESA production had not. At Tel Anafa, both BSP and ESA vessels occur beneath the earliest floors of the large villa that was built shortly after 128 BCE [...]. Therefore, sometime between 144 and 128 BCE, Levantine potters began manufacturing ESA’.90 Since the abandonment of Tel Anafa predated the dwindling of ESA, the end of the typochronological framework could not be resolved. Secondly, Slane identified the so-called BSP, whose fabric relation to ESA was corroborated through archaeometrical analyses (cf. infra).91 Thirdly, the site provided no conclusive, but some indicative, evidence for the Levantine origin of ESA. Besides the variety of (early) forms and variants, Slane also identified several new forms (her TA-types), rarer shapes that can be identified in Cyprus and the northern Levant in small quantities.92 Such features are generally seen as indicative of a site’s proximity to a/the centre/centres of manufacture.93

Johansen, who had studied the ESA found at Hama (Epiphaneia hereafter) in Syria, closely followed the typology of Samaria-Sebaste and basically agreed with Kenyon in thinking that ESA did not predate the 1st century BC, though Johansen was not entirely convinced.86 Negev, following the excavation of a supposed pottery workshop at Oboda, argued that part of the ESA (some of which was found in the workshop proper) was locally manufactured.87 The seminal work on eastern terrae sigillatae, which appeared in 1985, was the Atlante della forme ceramiche II. Therein, Hayes published typo-chronologies for all four eastern sigillatae then known, using virtually all published evidence available at the time, as well as unpublished finds particularly from Athens, Corinth and Paphos.88 Although the study of eastern sigillata has progressed since, it can still be regarded as the most comprehensive study of the subject, both for the typo-chronological framework and the discussion. Hayes placed the earliest of his forms of ESA around the mid-2nd century BC (stretching back to c. 200 BC), the latest in the late 2nd century (his ESA tarda forms), though the later phases of ESA still remain poorly understood. Recent evidence from Beirut points out that the production and distribution of ESA continued to at least the mid-3rd century.89

Most recently, ESA was tentatively identified as rhosica vasa. Rhosica vasa are mentioned by Cicero in a letter of 50 BC to his friend Atticus. A second reference dates to the late 2nd or early 3rd century, by Athenaios, who recounts a story of King Juba II. These references provide an intriguing insight into the status a category of ceramic tableware may have achieved.94 Such identifications, however, are fraught with problems, not only because of a possibly more generic use of such terms in Antiquity, as occurred with vasa Samia; a geographical label such as rhosica need not necessarily mean the objects were also produced there.95

Finally, excavations at the site of Tel Anafa in northern Israel between the late 1960s and the early 1980s provide the most recent landmark concerning our understanding of the production and distribution of ESA and its typochronological framework. Slane was able to throw light on several fundamental issues. First, the stratigraphy included several floor levels and destruction deposits, allowing the creation of several sequences, whose absolute chronology, however, partly relied on numismatic evidence. It nevertheless confirmed that ESA appeared close to the mid-2nd century BC, and ESA’s chronological relation

90  Also see Hayes 2008a, 19; Berlin 2006, 13-14. Note that Berlin sees BSP disappearing in the late 2nd century BC. Rather, it could be that the distribution of ESA this ‘far’ south had not really begun yet: stratified evidence from Jebel Khalid might suggest that ESA appeared a bit earlier, just prior to c. 150 BC; see specifically Berlin’s review of Jackson and Tidmarsh 2011. 91  Originally suggested by Weinberg, see Slane 1997, 269-282, 394-401; also Slane et al. 1993, 1994. Unpublished chemical analyses on BSP, ESA and other, related wares confirmed this connection, see n. 44. 92  TA types 15, 28 and 31 do not appear in the Atlante; TA type 15, however, could represent Hayes’ rara a. Interestingly, TA type 28 is identified at Jebel Khalid (Jackson and Tidmarsh 2011, 353-354) and elsewhere, and all three TA types are identified at Kinet Höyük (Hatay), as well as other thus far unidentified shapes. Kinet Höyük’s Hellenistic pottery is studied by the author and Jeroen Poblome. I warmly thank Marie-Henriette Gates for the opportunity to study these finds and refer to them here. 93  Rare or uncommon shapes commonly present themselves at Sagalassos in southwest Turkey. Known to have manufactured red slip tablewares, deposits excavated particularly in or close to the Potters’ Quarter usually contain one or more new shapes, or variants of existing ones. 94  Malfitana et al. 2005; Lund et al. 2006. 95  Wallace-Hadrill 2008, 411.

85  Kenyon 1957, using a British system. Illife (1938), however, had already proposed to apply Waagé’s classificatory system of wares A-C onto Eastern Sigillata. 86  Johansen 1971, 55-57. 87  Negev 1974, 1986; see now Goren and Fabian 2008 for an alternative interpretation of this workshop as a bakery. 88  Hayes 1985a. 89  Reynolds 2010, 90.

14

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:33 UTC All use subject to https://about.jstor.org/terms

Definition of the Topic Rhosos, well located geographically but hardly a major urban centre in Antiquity some 30km south of modern Iskenderun, is thought to have played a role of some significance in the distribution of ESA. The distribution pattern of ESA during the mid-1st century BC is not very well understood, yet several decades earlier it did reach the central Mediterranean, and considerable quantities of late Hellenistic ESA are found in Italy.96 Through word of mouth it may have become known as rhosica vasa to traders and consumers, and if rhosica vasa are indeed to be paralleled with ESA, it implies that the consumption of ceramic tablewares may have been more democratic. This can be gathered to some extent from the archaeological record, which suggests that ESA is found in geographically, socially and functionally varied contexts, though it remains insufficient to make firm statements about the diversity in consumption. Cicero’s letter indeed evokes ‘the feeling that he has personal knowledge of what he is talking about’.97 However, the fern-patterns Cicero mentions, as well as Athenaios’ ‘gaily decorated’ and ‘bright colours’ do not readily match the typological and decorative repertoire of ESA.

a harder fabric, which is nevertheless still softer than the normal, very pale brown fabric’.98 Provenance The first laboratory experiments relating to ESA, performed over 70 years ago, using spectrographic and petrological methods, suggested the existence of two chemically distinct groups.99 NAA carried out by Gunneweg et al. on a collection of 161 samples, designated ETS-I, included fragments of ESA (from Tel Anafa in fact) but also black slip pottery and fragments of mouldmade bowls. Interpretation of the analyses showed that ETS-I formed a single group, and comparisons with the composition of Cypriot Bronze Age pottery led Gunneweg et al. to suggest a source on eastern Cyprus.100 A group of 126 sherds from Tel Anafa, including about 80-85 fragments of ESA, was also analysed using NAA, again demonstrating the homogeneity of ESA and, importantly, the relation between ESA and BSP.101 Slane et al., however, rejected eastern Cyprus as the source of ESA based both on a more thorough archaeometrical and statistical methodology, and on the distribution pattern of ESA.102 Rather, ‘[…] these factors allow us once again to suggest a north Phoenician or Syrian source for ESA as well as its associated groups.’.103 A recent plea for a Phoenician origin of ESA certainly raises interesting questions, yet the somewhat geocentric approach prevents it from being entirely convincing.104

Both archaeological and archaeometrical arguments overlap in pinpointing the provenance area of ESA to the northern Levant, where ancient Rhosos is located, thus geographically supporting the identification of ESA as rhosica vasa. However, not until archaeological evidence is found, as well as acquiring a better understanding of the harbours around the Gulf of Issos, must this remain a hypothesis, albeit an intriguing one.

Other analyses carried out over the past decades all point to a Levantine source. Schneider, who also carried out analyses on ESA using both XRF105 and thin-sectioning,106 concluded that ‘[t]he general chemical similarity points to only one clay deposit which may be very inhomogeneous in K and Na.’, thus confirming the interpretation of Gunneweg et al. and Slane et al. Schneider further emphasised that a comparison of results obtained through different methods (in this case XRF and NAA) is possible, allowing for the differences in the preparation of the samples and that not all methods are equally suited to detecting specific elements.107 Also, ‘this production area must have been not far from ophiolithic rocks. This, together with archaeological arguments, leads to a most probably [sic] provenance somewhere in the coastal area between Latakia in Syria and Tarsus in Turkey.’.108

Definition of the Ware ‘Eastern Sigillata A is defined by a combination of fabric and vessel shape; one of the chief characteristics of the ware is a very high degree of standardization of fabric, vessel shapes, and firing. The pieces discussed here are wheelmade throughout. The scars of up to three circular kiln supports placed in an equilateral triangle appear on the floors of plates and in a corresponding position on the foot, suggesting that such plates were stacked upside down for firing. Double-dipping streaks and scars from the kiln supports are most common in the Hellenistic period. The biscuit of the ware is normally hard and granular, without visible inclusions; its color varies between very pale brown (10YR 8/3) and pink (7.5YR 7/4). The slip, which is applied in most pieces by double dipping from the rim, is usually dark red with a non-reflective sheen. Variations from this norm are slight but distinct. A minority of pieces have a softer, pink biscuit (7.5YR 7/4), in which tiny sparkling inclusions are sometimes visible; the slip of such pieces is more orange and may be matte and worn or have a high luster. Pieces with a matte, worn slip have a powdery biscuit, while those with a lustrous slip have

96  97 

Slane 1997, 269-270. Crowfoot et al. 1957, 471-474. 100  Gunneweg et al. 1983, 11-14; the term ETS-I for ESA (which included BSP(?) and ESA, including Tel Anafa) is mostly used in the southern Levant. 101  Hayes 1985a, 9-10, 13, 23, already spoke of a ‘vernice nera’ for his forms 1 and 20, which belong to his earliest shapes and are morphological protagonists pur sang of Hellenistic tableware repertoires more generally. 102  Slane et al. 1993, 1994. 103  Slane et al. 1994, 64. 104  Regev 2007. 105  Schneider 1995. 106  Schneider 1996b, 194. 107  Schneider 1995, 416. 108  Schneider 1995, 415-416, 1996a, 131, 1996b, 194. 98  99 

Malfitana et al. 2005, 201, table 2. Lund et al. 2006, 498.

15

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:33 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East Antioch was recently put forward as a likely candidate for the origin of ESA. NAA performed on a number of black-glazed (in fact dark-slipped) and ESA fragments from Antioch and Jebel Khalid (and Pella in Jordan), indicated a probable common origin.109 Of the 22 samples from Antioch, ‘the rest of the samples [16] seem to fall well into the pattern already recognized for the Jebel Khalid [...] samples, suggesting they have a common origin’. Differences in the fabric and colour of ESA were also observed macroscopically at Jebel Khalid, yet ‘recent opinion [...] tended to favour a single source’, the differences ‘being the result of different firing temperatures rather than different clay sources’.110 However, even if we picture a major urban centre such as Antioch as a/the source and main motor behind ESA and other (earlier) slipped wares, also because of the ‘competence with which most of the vessels in this fabric [‘black-glazed’ Fabric 2] were produced’,111 the analyses — relevant nonetheless — are no strict proof for local Antiochene production, as these analyses merely show that the clays from which ESA and related slipped wares were made most probably originate from a single source. It is noteworthy that moulds for manufacturing mouldmade bowls were in fact retrieved at Antioch.112

(archaeological) argument for pinpointing the source of ESA. The distribution of ESD, however, seems to have had an influence in this as well (cf. chapters 3-4). On the other hand, the thin distribution of rare forms and variants appears to be confined to the northern Levant and Cyprus, which can now be supplemented by recently studied finds from Kinet Höyük. These include TA types that were not included in the Atlante, as well as a considerable number of BSP- and ESA-shapes not documented in either the Atlante or Slane’s Tel Anafa. Moreover, BSP and ESA appear to have been the continuation of early and mid-Hellenistic colour-coated tablewares, presumably originating in the northern Levant. 2.2.2 Eastern Sigillata B (ESB) History of Research At Priene, Zahn also distinguished a highly micaceous red slip tableware that he termed ‘Klasse A’.118 Zahn made two further observations that still are of primary interest to the study of terra sigillata. He not only noted the transition in firing technique from black to red slip, but also discussed the possible influence of ITS on the production of his ‘Klasse A’. Stamps bearing the name C. SENT(IVS), and to a lesser extent P. LVSIVS, provided the fundamental argument for his hypothesis of a ‘Filiale’, or branchworkshop(s), by (an) Italian potter(s) in western Asia Minor.119 Zahn suggested that his ‘Klasse A’ might have its origins on Samos, based on Pliny’s much-quoted passage: ‘maior pars hominum terrenis utitur vasis. Samia etiam nunc in esculentis laudantur. retinent hanc nobilitatem et Arretium in Italia et calicum tantum Surrentum, Hasta, Pollentia, in Hispania Saguntum, in Asia Pergamum. habent et Trallis ibi opera sua et in Italia Mutina, quoniam et sic gentes nobilitantur et haec quoque per maria, terras ultro citro portantur insignibus rotae officinis’,120 though he was still uncertain regarding the ware’s emergence.

The archaeological arguments put forward by Slane,113 and also mentioned by Schneider,114 include mapping the earliest and latest forms, as well as rare forms and variants. It is logical to reason that the closer one approaches a production centre, the greater the variety in shapes becomes, and this should at least suggest a general direction in which to search for the proverbial needle in the haystack. Although even Slane rightly stressed the scarcity of published ESA south of Antioch at the time, this can now be supplemented with published evidence from e.g. Kanatha, Petra, Mampsis and Ashkelon, sites where late 1st- and 2nd-century forms of ESA are attested in some quantity.115 Hayes’ earliest forms are now also known from Caesarea Maritima and as far south as Petra.116 The earliest, southern-most ESA find(s) in context come from Koptos.117 All this suggests that ESA entered the southern Levant during most of its period of production. Consequently, the distribution of the earliest and latest forms of ESA no longer seems an unequivocal

Waagé fully agreed with Zahn’s interpretation of Pliny’s text concerning vasa Samia,121 wondering why this identification was not already widely accepted. Waagé further observed that both ‘Pergamene’ and ‘Samian’ disappeared after the 2nd century (contra his original terminus ante quem of c. 100 for ‘Pergamene’, cf. supra, 2.2.1), when other wares appeared. Waagé paid only little attention to the matter in his 1948 publication, given the negligible amount found at Antioch, but insisted that use of the term ‘Samian’ should be maintained,122 in spite

Garnett 2011. Jackson and Tidmarsh 2011, 326-327, n. 124; Garnett 2011, 539. Jackson and Tidmarsh 2011, 283. 112  Waagé 1948, 29, fig. 9, nos 1-3, though Waagé doubted the date of no. 1. 113  Slane 1997, 272. 114  Schneider 1996b, 194. 115  Kanatha: Henrich 2002; Petra: e.g. Schneider (C.) 1996; Sieler 2004; Mampsis: Negev and Sivan 1977; Ashkelon: Johnson 2008a: 13-18. 116  Caesarea Maritima: Berlin 1992; Petra: Schneider (C.) 1996. 117  Herbert and Berlin 2003, architectural phase V 1.1, Hellenistic 3 assemblage, dated to the mid- or later 2nd century BC (12; 84-86, fig. 65 for the ESA), even if nos H3.6-7 are attributed to Hayes form 23, datable to c. 100-50 BC, yet at Tel Anafa this form (TA type 29) already started in the later 2nd century BC. Furthermore, both H3.3 (Hayes form 2A, c. 150-100 BC) and H3.7 (Hayes form 23, c. 100-50 BC) need not strictly belong to these forms. H3.3 also bears some similarity to Hayes form 3 (c. 125-1 BC), and the base profile of H3.7 is not unique to Hayes form 23, but can very well occur on Hayes form 22A. Moreover, H3.4 (Hayes form 18, c. 125-75 BC) is also not unlike Hayes form 22B and Tel Anafa type 25c; the latter is attested in early imperial phases at Tel Anafa (Slane 1997, 310). 109  110  111 

Zahn 1904, esp. 440-447. Zahn 1904, 431, 445; see, however, Comfort 1929, esp. 484-486. Pliny, Naturalis Historiae, Book 35.46.160-161: ‘the majority of mankind employs earthenware receptacles for this purpose. Among table services Samian pottery is still spoken highly of; this reputation is also retained by Arezzo in Italy, and, merely for cups, by Sorrento, Asti, and Pollenza, and by Saguntum in Spain and Pergamum in Asia Minor. Also Tralles in Asia Minor and Modena in Italy have their respective products, since even this brings nations fame, and their products also, so distinguished are the workshops of the potter’s wheel, are carried to and fro across land and sea’. 121  Waagé 1937 for an elaborate discussion of the use and abuse of a term as vasa Samia. 122  Waagé 1948, 38. 118  119 

120 

16

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:33 UTC All use subject to https://about.jstor.org/terms

Definition of the Topic of counter-arguments (also regarding ‘Pergamene’) put forward by Illife.123 Based on finds from the Athenian Agora, Robinson made a qualitative and morphological distinction between ‘Samian A’ and ‘B’, which was followed not much later by Hellström in his discussion of the finds from Labraunda, from excavations carried out in 1951, 1953 and 1960.124 Hellström, however, introduced the more objective terms ESB I and ESB II, following Kenyon’s framework. In addition to the Latin stamps mentioned above, a circular stamp on ESB I reading SERENI was published in 1977, and stamps on ESB reading ARRETI, ARRETINA and the like provide further evidence of an association with Italian potters.125 Hayes, who in fact has provided the only typo-chronology to date, maintained Hellström’s classification of ESB I and II, which correspond to Robinson’s ‘Samian B’ and ‘A’ respectively. In addition, Hayes defined a transitional phase, ESB I-II, and placed the appearance of ESB in the Augustan period; none of his forms postdate 150, though recent evidence points out that production and distribution of ESB continued well into the 3rd century.126

Provenance Originally, Zahn was somewhat hesitant to use Pliny’s vasa Samia as an argument to place the provenance of his ‘Klasse A’ on Samos.128 Until the publication of the Atlante, not a great deal changed in the discussion concerning the provenance of ESB. Its distribution pattern was a reason for Hayes to propose a source in Asia Minor, yet he drew attention to a much more compelling clue: ‘[i]n effetti Tralles, posta a circa 50 km all’interno, sembrerebbe maggiormente proponibile come luogo di provenienza, sebbene tale ipotesi non sia a tutt’oggi confermata da dati di scavo. Questa è une delle località i cui prodotti vengono ricordati da Plinio (N.H. XXXV.12.46.160); inoltre alcuni frammenti bollati di Notion, con il nome di Caesarea (la denominazione ufficiale di Tralles nel primo periodo dell’Impero) mastrano caratteristiche tipiche della Sigillata Orientale B I […].’; the stamps (two examples were found) in fact read ΕΚΚΑΙ/CΑΡΗΑΣ.129 A group of 20 sherds was analysed using AAS, indicating that ESB formed a distinct group, and the macroscopic distinction in ESB I and ESB II was probably not associated with the use of different clays.130 Laboratory analyses performed on ESB samples from Ephesos and other sites in the area indicated that ESB is a homogeneous group. Chemical analyses were also carried out on clay samples collected close to Aydın, whose composition showed great similarity to that of the ESB samples.131 This implies that Aydın — which is in fact ancient Tralles! — possibly along with one or more other centres in the Maeander Valley, was involved in the production of ESB.132

Definition of the Ware ‘La caratteristica tipica di questa classe ceramica, riconoscuita da tutti gli autori, è quella dell’impasto poroso e con forte presenza di mica, inconfondibile nella serie B 2, mascherata talvolta da un’accurata [sic] preparazione dell’argila nella B I. In contrasto con la Sigillata Orientale A, sia l’impasto che la vernice sono prodotti con la stessa argilla, più ferruginosa di quella della maggior parte delle ceramiche microasiatiche (se si eccettua quella di Çandarli), la quale generalmente dopo la cottura ragguinge un colore vermiglione vivo o rosso aranciato [Munsell 10R 5/16 (vernice); 10R 6/14 (argilla); gli esemplari di B I sono spesso nell’ambito 2.5 YR-10R 5/12]. La mica compare sotto forma do sottili scaglie o macchioline di muscovite, di aspetto argenteo; ad occhio nudo non si distingono altre impurità. La struttura dell’argilla si presenta scagliosa e stratificata, tranne che negli esemplari più antichi e più accurate, questa ceramica offre poca resistenza alla usura e presenta spesso una superficie molto danneggiata con la perdita dei particolari decorative più sottili. La vernice ha un aspetto opaco, saponoso, simile alla cera, ben lontano dalla brillantezza della ceramica aretina. Nella serie B I sono frequenti le strisce di doppia immersione, mentre nella serie B 2 esse vengono nascoste dalla vernice di maggior spessore’.127

2.2.3 Eastern Sigillata C, Çandarlı Ware (ESC) History of Research Pergamon and its surroundings were actively involved in the production of a wide variety of ceramic products, including tablewares.133 A clear testimony was the discovery, in 1913, of the remains of a workshop close to the gymnasion that included a kiln, wasters, and moulds for mouldmade bowls.134 Two years earlier, Loeschke had found pottery wasters at Çandarlı (ancient Pitane),

Zahn 1904, 430-431, 447 (‘Fabrikationsort’). In British academia, Samian generically designates red slip tablewares of the (early) Roman period. 129  Hayes 1985a, 49; Demangel and Laumonier 1923, 384-385, considering the stamps were Greek, not Roman. 130  Hatcher et al. 1980, 141, 146. 131  Schneider 1996b, 189. 132  See Ladstätter 2007, esp. 208-210, with references. A brief article by Takaoğlu (2006) discussing archaeological research in Tralles, mentions ‘[m]anufacturing debris’ and ‘waste dumps of the workshops’. Despite these tantalising remarks the article regrettably fails to live up to its expectations by discussing and illustrating these, and rather chooses to focus on the potters’ stamps, thereby making somewhat odd statements in the process. A series of recent analyses on 41 fragments from Troia contributes little to the matter: Tekkök and Pernicka 2012. See now also the instructive contribution by Civelek 2010. 133  Poblome et al. 2001b, 145-146. 134  Hepding 1957. 128 

Illife 1938. Hellström 1965, 28-33, esp. 30, 32-33. 125  For the SERENI stamp, Wrabetz 1977. For stamps reading ARRETI, ARRETINA and so on, see e.g. Mitsopoulos-Leon 1991, H52 (111), H154 (117), I22-23 (129); Zabehlicky-Scheffenegger 1996, 11.1-2 (256257, 270). 126  Hayes 1985a, 49-52; Ladstätter 2010a, 190, n. 220, Fundkomplex A-BIV/5-SR 5a, dated to 225-250 and which contains a ‘Spätform’ of ESB Hayes form 60, and 191-192, Fundkomplex A-BIV/8-SR 5c, of Severan date (c. 190-225/235?); Ladstätter 2008, 111, n. 752, 123, Schnitt 2/01-SE 21, no. K 3, Tafel 282, for a specimen in a deposit dated to the third quarter of the 3rd century; Reynolds et al. 2008, 75. 127  Hayes 1985a, 49-50. 123  124 

17

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:33 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East located some 30km from Pergamon as the crow flies.135 Between 1977 and 1988 remains of a potters’ quarter were found, including several kilns, now submerged following the construction of a dam.136 Prior to that, in 1962 Schäfer defined a ‘lokalpergamenisch’ (which in fact paralleled Çandarlı Ware), and as it became increasingly clear that tablewares made at Pergamon were clearly different from Zahn’s and Waagé’s ‘Pergamene’, Schäfer opted for abandoning the latter term.137 Following his classification of the latest series of Çandarlı Ware products in LRP,138 Hayes greatly expanded his form-series in the Atlante by defining an Augustan series and a series based on Loeschke’s finds.139 Subsequently, based on pottery finds from a number of cisterns in Pergamon‘s urban centre, Meyer-Schlichtmann wrote in 1988 that ‘[d]ie Vorlage der pergamenischen Sigillata soll einen weiteren Beitrag zur Erforschung der pergamenischen Keramik liefern und eine lang bestehende Lücke in der Bearbeitung der östlichen Sigillaten schließen.’.140 He considered his ‘pergamenischen Sigillata’ to have started during the second half of the 2nd century BC, placing the closing stages of production somewhere in the course of the 2nd century.141 Meaningful units were assembled only afterwards, as the cisterns were not excavated stratigraphically. Moreover, the dating relied heavily on numismatic evidence — so scarce as to be almost absent — oil lamps and stamped amphora handles, and other criteria to distinguish one ‘Füllung’ from the next. Other criteria concern sudden changes in the quantity of pottery and other material categories. Unfortunately, these aspects do not necessarily add to the reliability of the typochronological framework. Nonetheless, the chronological spectrum of the production of tablewares in the area of Pergamon was thus greatly expanded, now spanning the 2nd century BC to the later 3rd century, if not later. Recent archaeometrical research (cf. infra) increasingly lays bare the complexity of tableware production in the region of Pergamon, up to the point that it — like ITS and ARSW — should in fact be considered a regional production.142 As a matter of fact, tableware manufacture in this part of

western Turkey has been plausibly explained as a process of ‘Dezentralisierung’ in the course of the Roman period (if the origins of this process do not go back further in time), whereby (at least for the early Roman period) ‘sich die Töpfereien in den übrigen Siedlungen und Städten der Umgebung ebenfalls an der pergamenischen Keramiekproduktion orientierten’.143 The Ketios Valley workshops, and the findings at Çandarlı in 1911 appear to mirror this diachronic development.144 Definition of the Ware ‘Der Ton ist i.d. Regel fein bis mittelfein, meist hart gebrant, die Bruchflächen sind fest und unregelmaßig gesplittert. Glimmer ist im Ton nur in kleine Mengen vorhanden. Die Einteilung der Tonstruktur in fein und mittelfein richtet sich nach der Durchsetzung des Tons mit kalkigen Einschlüssen: fein bedeutet frei von solchen Einschlüssen, eine mittelfeine Tonstruktur weist einige dieser Einschlüsse auf. Scherben mit grober Tonstruktur, die stark mit Kalkpartikeln durchsetzt sind, kommen bei diesem Material äußerst selten vor’.145 Besides the pottery workshops discovered just outside the urban area of Pergamon,146 the production of tablewares was also initiated at Çandarlı during the 1st century.147 This production continued into the late 3rd, possibly even the early 4th century, and is characterised as follows: ‘[t]he distinguishing mark of the Çandarlı fabric is the occasional appearance in the clay of large flakes of golden mica. The ware is otherwise often virtually identical in appearance to late pieces of Italian sigillata, though the Çandarlı products are rather thicker-walled and can be distinguished by their low heavy feet and lack of decoration. As noted by Loeschke, two fabrics may be distinguished. The products of the first century are characterized by a finegrained orange body with a lustrous orange or orange-red gloss, generally thicker on the inside than on the outside, and often omitted under the base. In the second century and later the ware approximates more closely to that of Italian and Gaulish sigillata; it is almost always hard-fired, generally red-brown, maroon or even purplish in colour, with (on the inside, at any rate) a good gloss of a similar shade. The interior of these late products is normally finely smoothed and given a thick gloss surface, highly resistant to wear; the exterior, on the other hand, tends to be poorly finished, with numerous scratches from the final tooling, and its gloss coating is thin and lacking in lustre. On occasions the gloss contains quantities of fine silvery mica. Not infrequently the marks of three small pad-like kiln-supports are visible on the floor or the resting-surface of the foot. Decoration is kept to a minimum’.148 A recent program of analyses has shown the existence of a number of larger and smaller groups, which are not all considered to have had a regional origin (cf. infra).

Loeschke 1912. Poblome et al. 2001b. Other categories of tablewares, such as mouldmade bowls and Applikenkeramik, may also have been manufactured in these workshops; Meyer-Schlichtmann 1988, 11-12. 137  Schäfer 1962, 778-780. 138  Hayes 1972, 316-322, 369. Hayes’ forms 1-5 in Çandarlı Ware, in fact a different product from the same area, share considerable chronological and morphological overlap with Pergamene Sigillata, and recently it has been argued that macroscopically and archaeometrically distinguishing Çandarlı products from those made in Pergamon is no longer unequivocally tenable. Clay composition, as well as morphological similarities between Çandarlı Ware and LRC, indicate that tableware production in the Pergamon region continued into Late Antiquity. 139  Hayes 1985a, 71-78. 140  Meyer-Schlichtmann 1988, 10-11. 141  Meyer-Schlichtmann 1988, 11; Slane (1991b) has objected to the term terra sigillata, as vessels became fully red-slipped only from about 50 BC. 142  Perhaps this is further mirrorred by Abadie-Reynal’s (2007, 115) observation ‘que la forme Hayes 4 présente généralement une argile différente [...] ce qui peut permettre de suggérer que cette forme était produite dans d’autres ateliers et peut-être dans un autre centre’?; only until archaeometrical confirmation comes forward must this remain a hypothesis. Within this (geographical) context it is nonetheless interesting to note that LRC was (also) made in several places (cf. infra, 2.2.8). 135  136 

Japp 2009a, 210. Poblome et al. 2001b. Meyer-Schlichtmann 1988, 14. 146  Poblome et al. 2001b. 147  Loeschke 1912. 148  Hayes 1972, 316-317. 143  144  145 

18

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:33 UTC All use subject to https://about.jstor.org/terms

Definition of the Topic the House of Dionysos at Paphos.156 Negev claimed it as a product from Oboda, following excavations of a so-called potter’s workshop at that site, and coined the name ‘Nabataean Sigillata’.157 This claim relied solely on archaeological arguments,158 with no support from quantitative archaeometrical analyses.159 In keeping with Kenyon’s original scheme, in 1978 Rosenthal proposed to use the term ESD until firm proof of its origin should be found.160 The pottery from the House of Dionysos was finally published in 1991,161 in which Hayes argued that ESD developed out of a local tradition of colour-coated pottery, and appeared towards the very end of the 2nd century BC and continued into the second half of the 2nd century.162

Provenance Until the discovery of the Potters’ Quarter at Sagalassos in the late 1980s, ESC was the only Eastern Sigillata for which archaeological evidence of its manufacture was known (cf. supra). Fourteen samples from the Ketios Valley excavations were analysed using thin-sectioning and XRF; these indicated a mineralogical and geochemical homogeneous clay composition.149 Analyses on fragments from Pergamon and Çandarlı showed that ‘the chemical data cannot be split up into a group of Pergamum and another one of Çandarlı’; a recent program of analyses (NAA, Röntgenfluoreszenanalysen)150 indicates that separating products from Pergamon and Çandarlı is, archaeometrically, not always possible.151 This nevertheless sparked the notion that speaking of ESC and Çandarlı Ware as macroscopically/archaeologically distinct wares is basically no longer tenable, which obviously has implications for those working in the field, if not for data already published.152 However, more generally the analyses program was able to show that — with varying degrees of certainty, and on occasion backed up by archaeological arguments — the majority of the analysed fragments have their provenance in the wider region of Pergamon; the analysis of a clay sample from the environs of Pergamon is ‘sehr ähnlich zur Hautgruppe [sic] von Pergamon’.153 This picture of regional production is further coloured by finds from Elaia.154 The ‘[d]ezentralisierung’ of the manufacture of tablewares with similar morphological and chronological dimensions in the wider region of Pergamon, as well as in the city proper (one might speak of a koinè), invites to capture these different products under the appropriate heading of ESC; ESC will be used throughout the text. Whenever possible, the different products are to be distinguished as Pergamene/-ian ESC, Çandarlı(an) ESC, and so forth. As a matter of fact, a scenario of regional production applies to most of the eight wares under discussion here.

Definition of the Ware ‘The body-clay is extremely fine-grained, of a muddy consistency, and breaks cleanly; the only visible impurities are occasional lumps of hard lime, which in the poorer pieces can be of considerable size. Vessels are given a complete coating of a thin wash of the same clay, which generally appears a shade darker than the body in colour; this, depending on the firing-temperature, has a more or less pronounced metallic lustre, a feature peculiar to this ware, distinguishing it from the other Eastern fabrics. Double-dipping streaks are generally visible on dishes, though not on other forms; finger-marks can frequently be observed around the bases of vessels. The firing of the pots is generally good and uniform for any given piece, but shows considerable variations from vessel to vessel. Wellfired examples are mostly brick-red, maroon, purplishred or dark brown in colour, while the poorer-fired ones (common in the later phases) tend to be orange-red. Some early pieces were fired in a reducing kiln and have a metallic black or purplish-black gloss. The potting, particularly in the later phases, is rather heavy, with prominent tool-marks on the undersides of the vessels. In general, one may say that the earlier products are better finished, thinner-walled and better-fired than the later ones; a general decline in standards sets in during the second half of the first century A.D.’.163

2.2.4 Eastern Sigillata D/Cypriot Sigillata (ESD) History of Research This ware had been distinguished as a separate class of terra sigillata prior to Hayes,155 who proposed the term ‘Cypriot Sigillata’, following work on the material from

Provenance Hayes favoured a Cypriot origin, relying on the plentiful and varied finds from the House of Dionysos,164 a point emphasised in the final publication.165 Gunneweg et al., who termed it ETS-II, carried out NAA and refuted

Degryse et al. 2001, 115. Japp 2009a; Mommsen and Japp 2009; Schneider and Japp 2009. 151  Schneider and Japp 2009, 294, 296, 302-303, Gruppe 2; also see Daszkiewicz and Schneider 2011. 152  Schneider 2000, 533; Schneider and Japp 2009, 304; see, however, also Daszkiewicz and Baranowski 2011. 153  Predominantly Gruppen 1, 1a, 3-4, though not all samples were Roman-period, and Gruppe 4 in fact concerned post-Roman pottery; Schneider and Japp 2009, 294. Production wasters were occasionally included in the analyses (see 293, 295, 297, Tables 1-3); Mommsen and Japp 2009, 277. 154  Japp 2009a, 201; Mommsen and Japp 2009, 275; Schneider and Japp 2009, 298. 155  Jones 1950, 184, ‘[r]ed glazed ware with metallic luster’; Westholm 1956. 149  150 

Hayes 1967. Negev 1972. 158  Negev 1972, 1974, 1986. 159  Mentioned in Rosenthal 1978, 19, n. 88. 160  Rosenthal 1978, 19, and n. 81: ‘an ethnic or geographic definition should be replaced by a neutral one, and following Kathleen Kenyon’s sigillata classification, the term Eastern Sigillata D seems appropriate’. 161  Hayes 1991. 162  Hayes 1967, 1991, 37-38; cf. supra, n. 44, for a similar scenario concerning ESA. 163  Hayes 1967, 66; Kenrick 1985, 267; Daszewski 1995, 28. 164  Hayes 1967. 165  Hayes 1991, 37. 156  157 

19

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:33 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East Negev’s claim that it was produced at Oboda. Despite acknowledging the variety of clay sources on the island, they considered a Cypriot source unlikely;166 in fact, it seems that they firmly clung to their hypothesis that it was ESA (ETS-I) that originated on (eastern) Cyprus. Failing to find an association between their analyses and other categories of Cypriot pottery, ‘it was considered more profitable to turn away from Cyprus as well as Israel and Greece where much work on source material has also been done’.167

responsible for their manufacture’.173 Whereas Hayes had already observed the macroscopic resemblance between ESD and LRD,174 the typo-chronological gap between ESD and LRD remains to be bridged, despite important attempts to do so (cf. infra, 2.2.7).175 2.2.5 Italian Sigillata (ITS) History of Research ITS (Figure 6), like ARSW, was not manufactured in the eastern Mediterranean, though quantities can be considerable, particularly at sites that cluster along the axis Achaia-Cyrenaica. It began to be manufactured around 4030 BC — at least for northern Etruria it developed out of existing pottery traditions176 — and because of its western provenance its eastern distribution adds an interesting dimension to the tableware spectrum in the east. It was fairly widespread during the 1st century, and pockets continued to receive ITS into the 2nd century. However, it does appear that its overall quantitative impact in the east was fairly limited except for particular regions, at least so the stamps indicate (cf. infra, chapters 3-4).177

Their analyses were based on a small sample, which moreover concerned surface material, thus lacking any archaeological dimension. Nevertheless, they considered western and southern coastal Anatolia good candidates. ESD is indeed attested at sites in Lycia and Pamphylia, and the production of tablewares (LRD) similar to CRSW took place in late Roman times at a number of places in (southern) Asia Minor.168 In light of the evidence discussed in chapters 3 and 4, however, the claim made by Gunneweg et al. seems rather unlikely from an archaeological point of view. A range of archaeometrical methods, including thinsectioning and XRF, was employed in order to determine the chemical composition of a number of ESD (n=51) and LRD (n=5) samples, mostly from Paphos. Meyza deliberately also selected fragments which he considered to be transitional between ESD and LRD. Although the results were far from final, the analyses pointed out that ‘[m]ost samples belong to only one raw materials group’.169

Kenrick’s study of the finds from Berenice makes this the only city on the western fringes of the eastern Mediterranean for which the ITS has been published in full detail,178 and it seems to have been one of the few sites where ITS held an exceptional position (cf. infra, 3.2.1). Two instruments are crucial in the current study of ITS. The first is a by now widely-used typo-chronological framework, the Conspectus Formarum, and rather than using ITS, the authors prefer the term ‘Italian-type sigillata’. One of its aims was to provide a framework that moved beyond the different typo-chronologies then in use: ‘[e]ach system has its merits, but each is rooted in one particular national or cultural tradition, and tends to be better suited to the needs of one particular geographical area’.179 The second is the CVA, a project whose history can be traced back to the late 19th century.180 In the course of the 20th century it has grown into a powerful instrument for the study of (stamps on) ITS, now containing over 36,000 entries:181 ‘the compilation of OCK has made publicly accessible a vast amount of data in a structured form that can be used – with care, I must insist! – for all sorts of statistical analysis and historical interpretation; I am hopeful that some further improvements can yet be made’.182

The next important step was the work on the late Roman settlement at Kalavasos-Kopetra. A variety of pottery fragments and clay raw materials, including LRD and socalled ‘Red Tile’ fragments, were investigated through NAA. Because of the strong similarity in composition (‘chemically indistinguishable’) of LRD and the ‘Red Tile’ group, Rautman argued in favour of a shared production infrastructure located on Cyprus.170 Furthermore, analyses of terra rossa clays from the valley of Kalavasos-Kopetra indicated a faint similarity to the LRD and ‘Red Tile’ group, which suggests that these or similar clays were used in their manufacture.171 The relevance here is that analyses of ESD fragments also fall into this broad group:172 ‘[a] selection of early Roman Cypriot Sigillata sherds drawn from several locales on and off the island was found to fall within the range of variation of the CRS/RT reference group. This compositional uniformity corresponds with the consistent visual appearance of most specimens, and suggests that a single well-developed industry was

These are the two standard works to tackle ITS (also in the east) and are significant for dating purposes.183 Moreover,

Gunneweg et al. 1983, 14-15. Gunneweg et al. 1983, 15. 168  See section 5.3.7. Atık 1995, 161, mentioned in Vroom 2004, 294. See for example nos 355-357 on 164-165. Also see Poblome et al. 2001a; Poblome and Fırat 2011; Poblome forthcoming. At Ariassos, pottery wasters were noted by the author that bear very close resemblance to SRSW form 1B230 (Poblome 1999, fig. 38). As such, Ariassos probably also participated in this tableware koinè. 169  Meyza 1995, 183-185. 170  Rautman 1995b, 334-335, 2003, 268, 270. 171  Rautman et al. 1993, 251-261, 1995b, 333-336, 2003, 270; Gomez et al. 1996. 172  Rautman et al. 2003, 270. 166 

Rautman 2003, 270. Hayes 1972, 371. 175  Lund 1992; Meyza 1995. 176  Marabini Moevs 2006, 7. 177  Zabehlicky-Scheffenegger 2004, 79, pointing out that only about 5% of the collected stamps in OCK pertain to the eastern Mediterranean. 178  Kenrick 1985, 125-200. 179  Ettlinger et al. 1990, 1. 180  OCK 2000. 181  Kenrick 2006. 182  Kenrick 2006, 71. 183  Note that the Conspectus Formarum is largely based on sites in the

167 

173  174 

20

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:33 UTC All use subject to https://about.jstor.org/terms

Definition of the Topic

Figure 6. Three fragments of ITS (from left to right: Consp22.1.1; Consp18.2.1; Consp20.4), collected during intensive field survey at urban Tanagra (Boeotia, central Greece) by the Ancient Cities of Boeotia Project, directed by John Bintliff and Anthony Snodgrass (© Philip Bes).

Slane and Abadie-Reynal convincingly showed that it can be used as a tool to investigate economic and sociocultural issues within a site-based and regional context.184 In the present context, the OCK is carefully used in the interpretation of patterns of exchange in addition to sitespecific publications, as far as the evidence permits: ‘I drew attention in the second edition of the Corpus to the fact that any patterns which could be generated from it would be dependent both on the extent of actual publication in different regions, and on my coverage of those regions’.185

unidentified provenance; potters did not necessarily remain in one place during their active period.189 This said, some 15 places of manufacture have been identified by now through archaeometrical and/or archaeological research; some categories have, in fact, been defined solely through clay analyses, with no (precise) knowledge of the whereabouts of their place(s) of manufacture.190 Moreover, if we were to apply the term ‘Italian-type Sigillata’, this would also include Lyon (cf. supra, History of Research). Yet, ‘an archaeologist’ trying to distinguish products from the different workshops can ‘no longer presume to tell by the unaided eye where a piece of sigillata was manufactured’.191 Arezzo is no longer the manufacturing centre of ITS par excellence; the past decades have shown an increasing diversification of products and their origin. For instance, Kenrick originally had identified a ‘Tripolitanian Sigillata’, which was renamed ‘Campanian Orange Sigillata’ recently,192 as production is now attributed to the Bay of Naples.193 This category is fairly common at Berenice, and it would be surprising if (very?) small quantities did not arrive elsewhere in the east. The understanding of the production of ITS in the hinterland of Rome is also progressively better understood.194 The increasingly complex framework of regional production makes it impossible to give a single fabric description here, as laboratory analyses are in fact essential to distinguish the different workshops and their products.

Of a different dimension is the Early Italian Sigillata,186 which focuses on the Mediterranean at large with the aim of bringing west and east closer together, both historically and academically. It is clear, however, that the evidence from the eastern Mediterranean in general is insufficient to answer or approach some of the important questions.187 In the present context, the published evidence of ITS from the eastern Mediterranean is nonetheless incorporated if only in showing the increasing economic and cultural Mediterranean integration during the early Empire. Definition of the Ware and Provenance ITS, just like ESC, ARSW, LRC and now also LRD, is characterised by a model of regional production.188 A particular element in the study of ITS are the potters’ stamps. A considerable number of these have an western Empire. 184  Slane 2004; Abadie-Reynal 2004. 185  Kenrick 2006, 67. 186  Poblome et al. 2004. 187  Poblome et al. 2004, introduction. 188  Schneider and Daszkiewicz 2006; Klynne 2006; Kenrick 2004, 2006.

Klynne 2006. Olcese 2004; Kenrick 2004, 253-255, 2006, 66-67. Wells 1990, 2. 192  Kenrick 2004, 254. 193  Kenrick 1985, 283-302; Soricelli 1987, 2004, 300. 194  Kenrick 2004, 254; Olcese 2004. 189  190  191 

21

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:33 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East Based on the stamps, products from Arrezo, Pisa, Puteoli and Central Italy are the commonest finds of ITS in the eastern Mediterranean. A better understanding of the provenance of ITS for finds in the (eastern) Mediterranean will contribute to our knowledge of its system of supply. Here much depends on the identifiable stamps, yet as unstamped fragments are much more common, relying solely on macroscopic identification (through either fabric and/or (supposed) stamp provenance) remains difficult, thus prohibiting a more precise (quantified) mapping of the origins of ITS in the east.195

Bonifay’s200 synthesis presents the latest result of the investigations in Tunisia,201 bringing together the available evidence for different ceramic categories, including ARSW, amphorae and oil lamps. Based on available evidence, which was mostly drawn from Bonifay’s own fieldwork in Tunisia, he proposes a refreshed analysis of part of Hayes’ typo-chronological framework for ARSW. Like others, Bonifay convincingly shows how fundamental the archaeological and ceramological evidence is in reconstructing the regional production. The identification of the actual workshops (of ARSW, but also amphorae and other categories), and how these shifted diachronically within Africa Proconsularis, the understanding of clay composition and their provenance, the (agro-economic) basis on which Africa Proconsularis relied, as well as diachronic socio-cultural and technological aspects,202 are fundamental elements in unlocking this regional framework. Regarding the distribution of ARSW, Hayes’ typo-chronological framework remains in use for the eastern Mediterranean. Yet, with mounting and ever more diverse evidence from Tunisia, including Bonifay’s study, the time seems appropriate to re-evaluate the available evidence. This not only entails the potential of the evidence for ARSW in the east with regard to the typochronological and distributional frameworks, but also, which is as important, the assessment of what the research now in progress in Tunisia could mean for the eastern Mediterranean.203

2.2.6 African Red Slip Ware (ARSW) History of Research ARSW is, like ITS, of western manufacture, and its impact in the east was generally significant. Waagé was the first to pay considerable attention to late Roman tablewares in the east, first at Athens and later at Antioch,196 although Kübler had paid some attention to stamped ARSW from Athens.197 Whilst his study of the Athenian material remained exploratory, the Antiochene material allowed Waagé to elaborate and refine his wares A-C. Although the archaeological agenda of the excavations at Antioch was largely guided by literary evidence, Waagé’s scholarly focus, which amongst others concerned the recognition of an ‘intermediate’ group between ‘Late Roman A’ and ‘B’, was not neglected. Whereas the idea of a provenance in Egypt was not abandoned, based on the intensity of distribution and the presence of earlier forms, especially in North African museums, by suggesting that North Africa was a likely source Waagé was also more correct regarding the provenance of ‘Late Roman A’ and ‘B’. In retrospect, Waagé was also on the right path from a chronological point of view. The first phase of his ‘Late Roman A’ he now considered to span the mid-3rd to the early 5th century, followed by his middle phase, between the early 5th and ending around the mid-6th century. His last phase may have run into the first half of the 7th century. Despite the relatively poor stratigraphic record and primary reliance on numismatic evidence, Waagé also pioneered in constructing a more argued typo-chronological framework. In the context of the eastern Mediterranean, Waagé’s work would remain in isolation until Hayes’ seminal and most widely-used study to date, LRP and his A Supplement to LRP.198 These, for the first time, provided a relatively easyto-use handbook for tableware finds in the east for the midand late Roman periods. Hayes collected a great deal of evidence for his typo-chronological framework, yet since published evidence from the east remained scant, most was drawn from the western and central Mediterranean,199 a situation that has changed considerably since.

Definition of the Ware and Provenance Like ITS, the term ARSW should be seen as an umbrella: ‘[o]n regroupe sous le terme de sigillée africaine une multitude de productions d’aspect très varié’.204 Although a general fabric classification does exist, Bonifay has recently drawn attention to the fact that this system, comprising the fabric groups A, C, D and E (with subgroups),205 is becoming increasingly complex. In fact, this parallels the situation regarding the production of ITS. Several places of production have been investigated archaeologically, and/or their products are characterised through archaeometrical research. For instance, a workshop has been attested at El Mahrine in northern Tunisia, attributed to ‘D1’,206 and ‘C’ includes the large workshops at Sidi Marzouk Tounsi in central Tunisia.207 Products have also been fingerprinted archaeometrically, the places of production of which have thus far eluded archaeologists. The current state of at Carthage were instrumental: e.g. Hayes 1976a, 1977, 1978; Riley 1981; Fulford 1984; Tomber 1988. 200  Bonifay 2004, also 2002, 2003, 2005. 201  North Tunisia: Mackensen 1993; central Tunisia: Mackensen 1998b, 2003a, 2003b; Peacock et al. 1990; both areas, see Mackensen 1998a; Mackensen and Schneider 2002. 202  Bonifay 2007. 203  Bes and Poblome 2009. 204  Bonifay 2004, 45. 205  Hayes 1972, 287-292; Carandini et al. 1981; Martin 1999. See also Bonifay 2004, chapter 2 (with references); Mackensen 1993. For the manufacture of ARSW during the 2nd and 3rd centuries, see Mackensen and Schneider 2006; Mackensen 2006. 206  Mackensen 1993. 207  See also Peacock et al. 1990; Mackensen 2006.

195  Important attempts have of course been made. See e.g. Kenrick 1985, chapter 3, 125-218, esp. 129. 196  Athens: Waagé 1933; Antioch; Waagé 1948. 197  Kübler 1931. 198  Hayes 1972, 1980. 199  Adjustments and additions to the typo-chronological framework of ARSW have been made by Hayes and others. The UNESCO-excavations

22

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:33 UTC All use subject to https://about.jstor.org/terms

Figure 7. Sites and areas of production of ARSW in Tunisia (© Michel Bonifay forthcoming).

Definition of the Topic

23

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:33 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East knowledge thus begins to capture the complexity of that regional production, illustrated in Figure 7; the complexity of this model is further enhanced by the diachronic and geographic development of the (groups of) workshops. All in all, a more widespread use of the fabric classification could already contribute to a generally better understanding of the diachronic representation of the different production areas of ARSW in the eastern Mediterranean.

important question. A range of archaeometrical methods was employed for a number of samples, mainly from Paphos. In addition to standard ESD and CRSW/LRD fragments, Meyza purposely chose fragments which he considered to be transitional between the two. Although the results were far from final, the analyses at least pointed to compositional similarities for a number of samples from both wares.216 Although this did in fact not answer his continuity question, Meyza nevertheless proposed a scheme of how ESD forms could have developed into CRSW/LRD forms.217 Another matter that receives increasing attention is that of the homogeneity of both fabric and shape of CRSW/LRD. A decade ago, late Roman tablewares with an affinity to CRSW/LRD, but not being the ‘real stuff’, were amongst others identified at Perge.218 A recent plea was made to return to the use of the term LRD, whereby it was argued that (among others) CRSW/LRD and SRSW were not isolated cases but belonged to a wider region (a koinè) that partly shared the manufacture of stylistically-related tablewares, ‘a broader socio-cultural and socio-economic common language or koinè of pottery production’,219 and which encompassed (central) southern Asia Minor and Cyprus. The authors, however, emphasised that existing terminology, such as CRSW, should not be abandoned. Actually, these could be retained under an umbrella term, precisely one such as LRD, in order to make a clear hierarchical distinction.220 As such, differentiation could be made by combining both terms to denote both the provenance (when presumed or, better still, proven) and its morphological family, e.g. Cypriot LRD. This, in fact, very much echoes what concerns ESC (cf. supra) as well as LRC (cf. infra). In light of these recent developments, LRD is maintained throughout this text (with exceptions where applicable), given the knowledge that a broader family of related wares is increasingly recognised and that, for now, the majority was presumably produced in Cypriot potteries.

2.2.7 Cypriot Red Slip Ware/Late Roman D (LRD) History of Research Following his work at Athens, Waagé recognised ‘LRD’ and ‘E’ as further distinct classes of tableware, though Waagé also classed locally-made pottery under his ‘LRD’.208 In LRP, Hayes renamed both CRSW, given its regular occurrence on Cyprus and its macroscopic similarity to ESD; ‘Late Roman E’ designated the earlier specimens, characterised by ‘a rather thicker slip with a fine lustrous appearance’. He considered its manufacture to span the late 4th through the 7th century.209 Meyza provides an excellent synthesis of the research into CRSW/ LRD,210 and more recently a case was made to return to the use of LRD.211 Definition of the Ware The fabric is described as ‘consistently very fine and smooth, with no grain visible, and breaks cleanly, even when ill-fired. The only obvious impurities are occasional lumps of lime, often quite large, which tend to rupture the surface. The presence of these proves that the clay does not owe its fine texture to careful levigation; it presumably comes from alluvial deposits. The degree of firing and colour of the fabric vary enormously, the latter ranging from near-yellow through various shades of orange, brown and red to a deep maroon, purple or sepia; a pinkish or maroon tint is commonest.’.212

2.2.8 Phocaean Red Slip Ware/Late Roman C (LRC)

Provenance

History of Research

For an overview of the archaeological and archaeometrical arguments regarding the possible origin of CRSW/LRD, see also 2.2.4 (cf. infra).213 An ongoing issue is that of the continuity between ESD and CRSW/LRD. A macroscopic similarity was already observed by Hayes,214 and despite archaeological evidence pointing to settlement continuity, pottery shapes of the 3rd and first half of the 4th century remain absent.215 However, an increasing body of evidence appears to provide some tentative answers regarding this

Waagé classed his ‘Group C’ next to ‘A’ and ‘B’, and already noted the ubiquitous keeled rim that later would become Hayes form 3 (Figure 8). Waagé, following Kübler,221 suggested an Egyptian origin for this and his ‘B’ ware. He based his relative chronology primarily on the style of stamps, combined with archaeological considerations. Palmettes etc. were the first stamps, followed by animals, while Christian symbols came last. Waagé assumed that ‘Group C’ appeared in Athens in the second half of the 4th century, pointing out that it occurred in conjunction

Waagé 1933, 304-308. Hayes 1972, 371, 2007, 435-436. Regrettably, out of practical considerations Meyza’s valuable work could not be incorporated more fully into this study. 210  Meyza 2007, 15-16. 211  Poblome and Fırat 2011. 212  Hayes 1972, 371. 213  In addition here may be mentioned Gomez et al. 1996. 214  Hayes 1972, 371. 215  Lund 1992; Hayes 1980, 528. 208 

Meyza 1995, 179-184. Meyza 1995, figs 5-6. 218  Fırat 2000; Poblome et al. 2001a. 219  Poblome and Fırat 2011, 49; Kenkel 2007; Meyza 2007, 13-20. Assuming that a number of manufacturing centres shared this common language yet spoke in different tongues, similar to ITS and ARSW for instance, Ariassos belonged to it as well (see ns 168, 878). 220  Poblome and Fırat 2011, 54. 221  Kübler 1931.

209 

216  217 

24

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:33 UTC All use subject to https://about.jstor.org/terms

Definition of the Topic

Figure 8. Several types of Phocaean LRC Hayes form 3, collected during intensive field survey at urban Tanagra (Boeotia, central Greece) by the Ancient Cities of Boeotia Project, directed by John Bintliff and Anthony Snodgrass. Note the discolourations of some of the rims (© Ancient Cities of Boeotia Project).

with numismatic evidence of the 7th century, though he expressed some reserve about production continuing into this period.222 In his discussion concerning the pottery from Antioch, Waagé maintained his ‘LRC’ as a distinct category. Waagé now altogether rejected a western provenance, which in turn supported his argument for an African source for his wares ‘A’ and ‘B’.

Phokaia.225 Some progress has since been made, notably by Vaag, who positioned the Phokaia workshops as the most important with regard to (distributed) output, which should then be differentiated from the products manufactured by secondary workshops in western Asia Minor through the use of a geographical label: Phocaean LRC, for instance (cf. ESC and LRD).226 These other workshops supposedly catered for more localised demand, and their products only occasionally travelled beyond the region (cf. infra).227

It was not until the publication of LRP that new light was thrown on the matter. Hayes retained Waagé’s ‘LRC’; he not only noted that the ware was most commonly found in the Aegean, but also forwarded the hypothesis for a provenance in western Asia Minor, drawing attention to the similarity in macroscopic characteristics between Çandarlı Ware and LRC.223 In A Supplement to LRP, Hayes took this one step further, and LRC was renamed PRSW, following Langlotz’s discovery of wasters at Foça (ancient Phokaia).224 Not much later, however, it was argued that the production of PRSW need not have been restricted to

Empereur and Picon 1986; as in their identification of amphora kilns (Empereur and Picon 1989), the authors provide meagre archaeological observations to support their argument. As with the amphora kilns, much may have disappeared since, due to intensified construction along coastal zones. 226  Vaag 2005. 227  Vaag 2005; also Hayes 2008a, 83. Among the pottery from the surveys in Boeotia (central Greece) and on Skyros (central Aegean), as well as at Ammata (Jordan), several specimens were noted in a rather soft and powdery, pale orange-brown clay with no (remains of) slip. One might expect more sherds to share these features given their exposure to the elements, if such played a role; these sherds could in fact be similar to those mentioned by Vaag 2005, 133-134. Also note for instance the ‘Imitation Late Roman C’ from Akra Sophia: Gregory 1985, 424-426, nos 1-2, 17, fig. 4. 225 

Waagé 1933, 303-304. Hayes 1972, 369-370. 224  Langlotz 1969; Hayes 1980, 525. 222  223 

25

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:33 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East Empereur and Picon suggested that both Phokaia as well as Gryneion were involved in the production of PRSW/ LRC, a claim that was backed up by archaeometrical analyses.230 In addition to Hayes’ observation about the similarity in macroscopic properties they also noted morphological similarities between certain forms of Çandarlı Ware and PRSW/LRC.231 Next to these two sites, local production has recently been forwarded for Ephesos and Sardis, and is now attested at Çandarlı.232 Although the often discoloured rims are considered a recognisable and discriminating feature for products from the Phokaia workshops, this is apparently not exclusive.233 Though our knowledge at present remains relatively superficial, the increasing body of evidence implies that PRSW/LRC, like ITS, ESC, ARSW and now also LRD, is characterised by a regional framework of production (koinè). It is therefore that LRC is preferred and used throughout this text, with the knowledge that the majority was presumably manufactured at Phokaia. Naturally, in order to understand the geographical and (macro-)economic impact of each of the individual production centres, it will be vital to identify these archaeologically and macroscopically in the field much more.

Definition of the Ware ‘The clay is a fine-grained red, not unlike that of the finer African products, but having a different range of impurities. A large number of small lime particles are frequently (but not always) present in it, producing in the case of the darker, harder-fired examples a distinctive fine-speckled appearance; mica, on the other hand, is rare–hardly ever more than a few specks–and other impurities are absent. The ware is generally hard-fired, with a brownish-red, purplish-red or maroon tint, and breaks cleanly without splintering. The red slip, which covers the whole surface, is generally no more than a fine film, which fuses with the body-clay; sometimes however, it is more thickly applied on interiors. It is merely a refined version of the body-clay, dull or occasionally slightly metallic in appearance, with none of the gloss of the terra sigillata wares’.228 Provenance By the time LRP was published, Langlotz had drawn attention to kiln infrastructure and misfired vessels from ancient Phokaia,229 indicative of the local production of tablewares. Following archaeological finds at Gryneion,

Empereur and Picon 1986; Mayet and Picon 1986. See Hayes 2008a, 87, for a dish that is context-dated to c. 300, an ‘unusually early occurrence of the classic fabric’, thus emphasising the (morphological) relationship between Çandarlı Ware (ESC) and early LRC forms. 232  Rautman 1995a; Ladstätter and Sauer 2002, 2005; Ladstätter 2010b, 101; Ladstätter and High 2010, 163, K687; Yilmaz 2007, noting Phocaean, Gryneian and Ephesian (?) LRC at Priene. Schneider and Japp 2009, 296, 298, 304 (Gruppe 5); See also Quercia et al. 2011, 57. 233  Yilmaz 2007, 127. 230  231 

228  229 

Hayes 1972, 323. Langlotz 1969.

26

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:33 UTC All use subject to https://about.jstor.org/terms

Archaeopress

Chapter Title: The Late Hellenistic and Early Roman Periods: Discussion and Chronological Phasing Book Title: Once upon a Time in the East Book Subtitle: The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East Book Author(s): Philip Bes Published by: Archaeopress. (2015) Stable URL: https://www.jstor.org/stable/j.ctvr43kch.8 JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected]. Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at https://about.jstor.org/terms

Archaeopress is collaborating with JSTOR to digitize, preserve and extend access to Once upon a Time in the East

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:35 UTC All use subject to https://about.jstor.org/terms

Chapter 3

The Late Hellenistic and Early Roman Periods: Discussion and Chronological Phasing

demise of the six wares that occur during this period. ESC and ARSW continue into the late 3rd and the 7th century respectively, which is picked up again in chapters 5-6.

3.1 Presentation of the Collected Published Evidence This section introduces the six main wares for the late Hellenistic and early Roman periods: ITS, ESA, ESB, ESC, ESD and ARSW. The pie charts that present the relative proportions of the different wares for each region (3.2.115) are based on what has been collected in the database. On occasion, for the sake of argument (un)published data is additionally referred to — both here and in the mid- and late Roman chapters (cf. infra) — and is indicated as such. The line charts that capture the quantitative-chronological developments for each region are without exception based on a lower quantity than the pie charts: fragments that could be attributed to a ware, but not to a form (or its typochronological range), are included in the pie charts.

The mid-2nd century BC characterises the emergence of ESA and ESC;234 the former fairly quickly grew in quantitative importance. From around 100 BC they were joined by ESD, that together with ESC initially remained of modest significance concerning quantitative-geographical distribution. Following a brief stagnation, towards the mid-1st century BC ESA began a major period of growth, and reached an overall peak in the decades around 1. ESC also showed a gradual increase from the mid-1st century BC onward, and peaked towards the mid-1st century. ESD reached a (first) peak in the early 1st century. In the late 1st century BC two more wares appeared: ESB showed a gradual increase from the start of its production; ITS immediately rose very strongly and was most common in the first half of the 1st century. During this same period ESA dwindled markedly, whilst ESD witnessed a brief low, and the growth of ESB matured. ESA presented a

A brief discussion of the total collected evidence for the late Hellenistic and early Roman periods introduces the regional evidence, so as to gather a general notion about the significance of each ware. This overall chart (Figure 9) illustrates the appearance, development and eventual

Figure 9. Line chart showing the development, in absolute numbers, of ITS, ESA, ESB, ESC, ESD and ARSW during the late Hellenistic-early Roman period (n=8769) (© Philip Bes/ICRATES Project).

234 

27

Not until the mid-1st century BC did ESC appear fully red slipped.

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:35 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East 3.2.1 Africa-Cyrenaica Data was gathered from only two sites, Berenice and Cyrene; the pottery from both sites was studied and published by Kenrick.236 Proportions (n=578) are shown in Figure 10. The percentage of ITS is c. 66%; in fact, this region presents the highest percentage of ITS of all regions. This is followed by ESA, representing just over 16%, a fairly high percentage since Africa-Cyrenaica is the most distant region in relation to ESA’s presumed area of production. ESB and ESC represent 6.6% and 5.7% of the collected data respectively. ESD remains just under 3% which, as will be shown, compares rather well to other regions of the central Mediterranean (the Aegean and Crete).237 ARSW is represented by c. 3%, and is fairly common compared to other regions. The site-specific proportions are illustrated in Figure 11.

Figure 10. Africa-Cyrenaica: pie chart showing the proportions of the different wares (n=578) (© Philip Bes/ICRATES Project).

The developments in absolute quantities for AfricaCyrenaica mainly reflect the Berenice material, and are shown in Figure 12. The four main wares follow neatly one after another. ESA, which is found quite frequently in an area so distant to its source, was most plentiful in the late 1st century BC and early 1st century. It remained present throughout the 1st century yet gradually dwindled, disappearing in the course of the early 2nd century. ITS was the most common ware throughout the very late 1st century BC and the 1st century but fairly swiftly decreased in the decades around 100. ESB is predominantly represented by Hayes’ later ESB II-series, and was found with ITS in the second half of the 1st and early 2nd century. Interestingly, the quantitative developments for ITS, ESA and ESB are roughly comparable to the AegeanSouthwest, although the proportions are rather different, which in the case of ESB probably reflects the AegeanSouthwest’s greater proximity to its area of manufacture. ESD and especially ESC clearly did not play any substantial role, though occurred in some form-variety. ESC was somewhat common during the first half of the 2nd century, yet scarce both after and especially before that period.238 ESD conforms to the general pattern for the central Mediterranean, though the forms attested represent the entire period of manufacture.239 ARSW, finally, conforms rather well to the general trend for the eastern Mediterranean during this period. Though ARSW forms predating c. 200 are attested in the eastern Mediterranean, their quantities are very small. In Africa-Cyrenaica, however, ARSW is ever so slightly more common (see chapters 5 and 6 for the continuation of this development); possibly, the proximity to Africa Proconsularis was a decisive factor.

brief revival during the mid-1st century, but began to decrease afterwards. ESC experienced a brief low (as ESD did several decades earlier), but had a second floruit during the first half of the 2nd century. ESD showed a general increase and reached a peak in the late 1st and the first half of the 2nd century. ESB clearly experienced its main phase during the second half of the 1st and early 2nd century. ARSW appeared on the eastern markets in modest quantities by the late 1st century. ESB and ITS had disappeared by the mid-2nd century, and were followed by ESA and ESD towards the late 2nd century.235 3.2 Regional Developments The next 15 sections present and discuss the collected data region by region (see appendix 3a-b), according to the framework defined in 1.3.6, thereby following a template: first, a pie chart captures the relative proportions of the different wares, which allows to compare the different regions with one another. Second, a map shows the sitebased proportions, which helps to identify intra-regional similarities and differences. Third, the collected data that is typo-chronologically identified and quantified (cf. 1.3.3) enables us to trace each ware’s chronologicalquantitative development, regional significance as well as the interplay between these wares. Finally, the evidence of archaeological deposits — the availability and quality of which differs greatly between and within regions — serves, where feasible, to evaluate the strength of the other three approaches.

At the sanctuary of Demeter and Persephone at Cyrene, ‘[t]he vast majority of the Hellenistic and Roman fine ware Berenice: Kenrick 1985; Cyrene: Kenrick 1987; Martelli 1976. ESD is attested at Carthage (Ortisi 1999) and Lepcis Magna (Reynolds 1997) in very small numbers; interestingly, the chalice Hayes form 37(AB) has been identified at both sites. 238  Kenrick 1985, 257, 259. 239  Kenrick 1985, 267. 236  237 

The production of ESA and ESB continued well into the 3rd century. This could, typo-chronologically, not be taken into account here. It is the author’s impression, and nothing more, that the distribution patterns of both wares had contracted to encompass their respective heartlands.

235 

28

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:35 UTC All use subject to https://about.jstor.org/terms

The Late Hellenistic and Early Roman Periods: Discussion and Chronological Phasing

Figure 11. Africa-Cyrenaica: map showing the ware proportions of the different sites (© Philip Bes/ICRATES Project).

Figure 12. Africa-Cyrenaica: line chart showing the development, in absolute numbers, of the different wares (n=412) (© Philip Bes/ICRATES Project).

29

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:35 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East was found in either disturbed or superficial levels, and therefore makes little direct contribution to the interpretation of the structures in the sanctuary.’.240 Although two contexts are of possible interest, since different wares occur side by side, not too much emphasis may be put on these considering their nature and quality.241 Archaeological information is more plentiful for Berenice, yet since the large number of deposits rules out a detailed discussion, several were selected to highlight the major trends. A first benchmark is deposit 168.1, of the late 2nd to early 1st century BC, which included Campana A and C; ESA was sparsely represented.242 Of slightly earlier date is deposit 32 (and to a lesser extent 33), dated to the first third of the 1st century BC; next to e.g. Campana A and C and Knidian Grey Ware, it documents the arrival of the first considerable quantities of ESA.243 Deposits of about the second half of the 1st century BC seem rather scarce. The line can be picked up again in the Augustan period with deposit 38, which included ESA and ITS (including ‘Tripolitanian Sigillata’).244 This pattern continued in other deposits of the early 1st century, such as 54, and deposits 52-53 of Tiberian date.245 All three contained ITS (including ‘Tripolitanian Sigillata’) and ESA. The large deposit 46 (c. 25-30) contained mainly ITS (chiefly ‘Tripolitanian Sigillata’) and ESA, and recorded an early specimen of ESB. Deposits of mid-1st century date were quite common, notably 44, 48, 55, 60-61, and 170-171.246 These deposits reflect the prominent position of ESA and ITS (again including ‘Tripolitanian Sigillata’) in the import patterns. Sparse examples of other major wares could be identified, such as ESC and ESD (and Pontic Sigillata). ITS remained a significant imported tableware in the late 1st and early 2nd century, as shown for example by deposits 63, 69 and 72;247 these also contained single or a few fragments of the other major wares: ESB, ESC and ESD (and again Pontic Sigillata), notably in the large deposit 69. Deposits of the second half of the 2nd and early 3rd century were not very common, yet deposit 81 seemed to document the growing importance of ESC and ARSW.248 Deposit 84 (and to a lesser extent 82) of the early 3rd century contained a fair quantity of ARSW, thus anticipating the mid-Roman period; other categories of red slip tablewares were by now largely residual.249

Figure 13. Crete: pie chart showing the proportions of the different wares (n=735) (© Philip Bes/ICRATES Project).

island, most of which also provide pottery evidence for the mid- to late Roman periods. The urban sites of Knossos and Gortyna figure prominently in the published record. Although published data from other sites and areas is less common, it nevertheless allows us to outline the island’s tableware consumption, also beyond the urban level. Material was collected from Gortyna, Knossos, Kommos, Kydonia/Akrotiri and Phalasarna, and from a survey in the Vrokastro area.250 A total of 735 identifiable entries was collected; the proportions are shown in Figure 13. Of these, just over 37% is attributed to ITS and ESA comprises c. 18.5%. Aegean tablewares (ESB, ESC), however, are the most commonly imported group on the island, accounting for nearly 40%. ESD represents just below 4%, and ARSW is just attested with c. 1%. The scarcity of ESD is fairly typical for the Aegean at large, and Cypriot tablewares in fact remain rare throughout the Roman imperial period (cf. infra, chapters 5-6).

3.2.2 Crete Work on late Hellenistic and early Roman Crete has concentrated on a number of sites and regions across the

The geographic distribution is shown in detail in Figure 14.251 The patterns for Knossos and Gortyna do not differ too much, except for ITS and ESD (the evidence for ITS predominantly comes from Knossos and Gortyna): ITS comprises over 34% at Knossos, which is not unlike the average for Crete; at Gortyna it is c. 41%.

Kenrick 1987, 11. For a complete context-list, see 12-18. Kenrick 1987, 12-13, strata C10/11 A,α 1 and A,β 2. 242  Kenrick 1985, 487-488. 243  Kenrick 1985, 419, 432-433: the intrusive material ‘is not sufficient in quantity to cast doubt upon the date indicated by the vast bulk of the material’. 244  Kenrick 1985, 434-435. Also identified are Knidian Grey Ware and Pompeian Red Ware. 245  Kenrick 1985, 440-441. Pompeian Red Ware was found in deposits 52 and 54. 246  Kenrick 1985, 436-437 (deposit 44), 439 (deposit 48), 441 (deposit 55), 442-444 (deposits 60-61), 490-491 (deposits 170-171). 247  Kenrick 1985, 445-446 (deposit 63), 447-449 (deposit 69), 450 (deposit 72). 248  Kenrick 1985, 456-457, besides residual and intrusive material. 249  Kenrick 1985, 457 (deposit 82), 457-459 (deposit 84). 240  241 

250  Gortyna: Dello Preite 2004; Lippolis 2001; Magnelli 2001; Martin 1997a; Monacchi 1988; Rendini 2004; Rizzo 2001, 2004; Knossos: Coldstream 1973; Hayes 1971, 1983; Sackett 1992; Kommos: Hayes 2000a; Kydonia/Akrotiri: Raab 2001; Phalasarna: Frost and Hadjidaki 1990; Vrokastro: Harrison and Hayden 2005. See now Baldwin Bowsky 2014 concerning ITS stamps from Crete, in particular those from Aptera. 251  Recently published quantified evidence from Gortyna (Panero 2008) reflects very well the pan-Mediterranean character of the imported tablewares (include Gaulish Sigillata) during the 1st, 2nd and 3rd centuries. A chronological breakdown of these proportions is eagerly awaited.

30

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:35 UTC All use subject to https://about.jstor.org/terms

The Late Hellenistic and Early Roman Periods: Discussion and Chronological Phasing

Figure 14. Crete: map showing the ware proportions of the different sites and surveys (© Philip Bes/ICRATES Project).

Archaeological deposits support a more detailed picture of the development of these wares during the late Hellenistic and early Roman periods. Unfortunately, the available literature did not contain many relevant deposits. At Knossos, an early 1st-century BC lower well deposit contained ESA (and Italian Thin-Walled Ware); most of the 22 ESA fragments from the well deposit as a whole are preAugustan.252 Also at Knossos, the lower fill of well MW/589, dated to the later 1st century BC, contained ESA, ESD, Italian Thin-Walled Ware and a Koan/Knidian carinated bowl.253 Interestingly, a contemporary deposit, the lower fill of well RR/K/60, also contained only (or mainly) ESA and Italian Thin-Walled Ware.254 Both well deposits did not contain ITS, which contrasts significantly with Corinth (cf. infra, 3.2.3), where ITS appeared already in small quantities during the early and mid-Augustan period.

Generally speaking, the importation of ITS and ESB spanned the 1st century, whilst the other wares predominantly pertain to the periods before and after. Even if overall quantities are small, this is borne out by the line chart in Figure 15. ESA was mainly limited to the 1st century BC and the Augustan period, yet near-absent after the Tiberian period. ITS clearly dominated the imported tableware spectrum during the Augustan to Neronian period. ESB and ESC only became more frequent towards the mid-1st, and the early 2nd century respectively. Both ESB and ESC increased quantitatively around the time when ITS dwindled and only the occasional ESA vessel still reached Crete. ESD is rare although here also forms from the 1st centuries BC and AD were recognised. Finally, ARSW forms that pre-date the 3rd century have been identified sporadically; clearly ARSW must have occupied only a minor position during the 2nd century.

On Crete, ITS only became common in deposits dated to the Claudian period. Three deposits from Knossos, all dated to c. 40-60, contained a majority of Italian tablewares, both ITS and Italian Thin-Walled Ware, with a few specimens

A first observation pertains to ITS, which recurrently features in the other sites and surveys. The source of the ITS found in the Akrotiri peninsula may have been the town of Kydonia; similarly, the ITS found at Kommos may have arrived through Gortyna. Knossos, Kydonia and Gortyna were most likely primary receivers of ITS, part of which was subsequently redistributed.

Hayes 1971, 251-252, 263-264, 266, deposit KW/51/13 (‘[e]arly’). Hayes 1971, 250-251, 257-259. 254  Hayes 1971, 249-255. 252  253 

31

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:35 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East

Figure 15. Crete: line chart showing the development, in absolute numbers, of the different wares (n=489) (© Philip Bes/ICRATES Project).

of ESA, ESB and ESD each.255 Deposits at the Villa of Dionysos generally span the 2nd and 3rd century, and most of the imported tablewares found in deposits of the 2nd century were ARSW and ESC, the latter being more than twice as frequent.256 Finally, the early Roman phases in the Unexplored Mansion at Knossos present a comparable pattern, with slight yet interesting differences. PreAugustan deposits were seemingly scarce, and contained only a handful of relevant material (ESA and Italian ThinWalled Ware). The Augustan group A2 mainly contained ESA, and ESD came second, whilst ITS and ESB were represented by only a few specimens each.257 Group A1, dated to the late 1st century BC, also contained ESA (more common) and ESD. Groups B1 and B2, of Tiberian date, reflect the position now held by ITS and ESB, and even if ESA was still imported, part of it was by now possibly residual.258 In deposits of group C, dated to the mid-1st century, the position of ESA was further reduced (if by now it was not increasingly residual) in favour of ITS and ESB.259 The very peak that ITS reached during the late Augustan-Claudian period falls a bit earlier than the trends outlined by the author. ESB increased from the mid1st century onwards, and was most common in groups

datable to the Hadrianic period.260 ESB and ESC were both commonly imported tablewares from the later 1st century onwards, though the former remained the most common, probably until the mid-2nd century. This situation then shifted in favour of ESC that virtually became the only imported tableware, with ARSW making the occasional appearance.261 To a large extent the line chart for Crete conforms to the eastern Mediterranean evidence. One may of course question, here and elsewhere, whether these deposits are all necessarily of the same ‘archaeological’ quality, and thus provide a good basis for comparison. In some cases one would be right to ask such a question yet, in general, the patterns borne out by the evidence illustrated in Figure 15 conform to the contextual evidence. A comparison between the trends discussed and the line chart, shows that only ITS has a rather dissimilar course, with the peak falling later than the line chart shows. Since this chart is based on the date-ranges of the identified forms and stamps of ITS, this raises the question whether or not the ITS in mid-1st century deposits is partly residual in those deposits, or knew a longer life span, as by form-identification they pertain to an earlier period. However, this residuality-argument can be countered by arguing that it can also apply to the other wares, or to deposits in general (as part of the issue of the relative quality of deposits). Moreover, manufacture, shipping and subsequent distribution, consumption, use and finally reuse

Hayes 1971, 250-252, 257-259, 263-266, 270-271, the upper fill of well MW/58-9, the late phase of well KW/51/13, and an abandonment feature in RT/60. 256  During the 25th International Congress of the Rei Cretariae Romanae Fautores (RCRF) in Dürres, Albania, John Hayes observed that most (or all?) of his deposits published from the Villa of Dionysos excavations (1983) should now be dated about 30 years earlier. 257  Sackett 1992, 183-187. 258  Sackett 1992, 190-195. 259  Sackett 1992, 197-200, 203-207. 255 

Sackett 1992, 229-236, 239-242. Sackett 1992, 158-159, 163 (fig. 6), 242-249, groups R and S; Forster 2005, 129-130, summarising the trends described by Sackett. 260  261 

32

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:35 UTC All use subject to https://about.jstor.org/terms

The Late Hellenistic and Early Roman Periods: Discussion and Chronological Phasing or recycling were disparate operations, and thus need not have followed chronologically linear tracks, thus offering potentially other explanations for the disparity between publications and the chart(s) concerning the development of tablewares. 3.2.3 Aegean-Southwest The number of publications for this region from which data was collected is considerable, and reflects its generally healthy state of research into Hellenistic and Roman pottery: data from the sites of Aigeira, Argos, Athens, Corinth, Delos, Delphi, Eretria, Isthmia, Kabirion, Kenchreai, Kythera, Olympia, Patras, Phlius, Siphnos, Sparta, and Tenos,262 and the surveys of Aetolia, Asea, Berbati-Limnes, Hyettos, Keos, Koroneia, Laconia, Methana, Nemea, Oropos, Pylos, Tanagra, Thespiae and Veloukovo was collected.263 The relative proportions of this collected evidence, numbering 1323 fragments, are shown in Figure 16. Additionally, in a recent survey on Antikythera only ESC was identified with certainty, whilst ITS was relatively common at several sites in three extensive surveys in Achaïa, west of Patras; a small collection from Nikopolis includes ITS, ESB and early ARSW.264

Figure 16. Aegean-Southwest: pie chart showing the proportions of the different wares (n=1323) (© Philip Bes/ICRATES Project).

accounts for c. 12.5% of the finds. Last, ESD and ARSW are scarce also here, comprising c. 1.4% and 0.4% of the finds respectively.

ITS is the most common tableware imported here, at well over 43% of the collected total. These proportions, however, are considerably distorted due to the evidence from Corinth, and taking this out results in quite different percentages especially for ITS (c. 34%) and ESA (c. 28%). For now the evidence is considered as a whole, thus allowing a comparison with the other regions. In addition to ITS, both ESA and ESB occur in roughly similar percentages (22% and 20% respectively), whereas ESC

The predominance of ITS at Corinth is also borne out by the distribution map (Figure 17) for this region, that is, it is not distorted because of a very small quantity (see e.g. Aigeira and Delphi).265 At Corinth, ITS comprises over 63% of the imported tablewares. The fact that the typo-chronological range of ITS (c. 40/30 BC-150) does not span the entire period under study in this chapter (c. 150 BC-200) illustrates the frequency of ITS at Corinth even more. In general, however, ITS regularly turns up at sites in this region, and although quantities are sometimes (very) small, it shows that ITS achieved a wide distribution and penetrated into areas beyond the urban framework; see the Boeotian survey evidence, and that from the Kabirion. Most of the other wares are common, even if in varying proportions throughout the region. Sub-regional patterns do not immediately surface, except perhaps for ESA which seems to be more common in the region’s eastern part, an aspect that resurfaces in the late Roman period.

262  Aigeira: Hagn 2003; Thrümmer 1986; Argos: Abadie 1984; AbadieReynal 1995; Aupert 1980, 1986; Sève 1980; Athens: Oxé 1927; Robinson 1959; Thompson 1934; Vogeikoff-Brogan 2000; Waagé 1933; Corinth: Comfort 1929; Hayes 1973a; Robinson 1969; Sanders 1999; Slane 1986, 1987, 1990, 1994; Williams II 1980; Williams II and Fisher 1975, 1976; Williams II and Zervos 1982, 1983, 1985, 1986, 1989; Wiseman 1967; Wright 1980; Delos: Brun and Brunet 1997; Bruneau 1970; Peignard 1997; Delphi: Perdrizet 1908; Déroche and Pétridis 1992; Eretria: Metzger 1969, 1993; Schmid 1999, 2000; Isthmia: Gregory 1995; Hayes 1998a, b; Marty 1993. See now Rife 2012 (99-102) for some early Roman, typologically unidentified fragments; Kabirion: Heimberg 1982; Kenchreai: Adamsheck 1979; Kythera: Coldstream 1972; Olympia: Martin 1995, 2001; Schauer 2001; Sinn et al. 1993; Patras: Hübner 1996, 2003; Phlius: Sutton 1990; Siphnos: Brock 1949; Sparta: Bailey 1993; Hayes 1995b; Pickersgill and Roberts 2003; Tenos: Etienne and Braun 1986. 263  Aetolia: Bommeljé and Vroom 1995; Asea: Karivieri 2003; BerbatiLimnes: Forsell 1996; Keos: Cherry et al. 1991; Laconia: Lawson 1996; Methana: Mee and Forbes 1997; Nemea: Sutton 1990; Oropos: Sakka 2001; Pylos: Berlin and Heath 2005; Tanagra: for preliminary reports, see Bes accepted a, b; Bes et al. 2011; Peeters et al. in press; Thespiae: unpublished finds originally studied by John Hayes; Bes and Poblome accepted. Late Hellenistic and Roman (imported) ceramic finds from Hyettos (originally studied by John Hayes) and Koroneia are (re)studied by the author, for which see Bes accepted a, b. The same nomenclature is applied to all four sites, so as to obtain a uniform basis for further regional research and comparison. I warmly thank John Bintliff and Anthony Snodgrass, directors of the Ancient Cities of Boeotia Project, allowing me to study these finds; Veloukovo: Vroom 1993. 264  Antikythera: Quercia et al. 2011, 51: Achaïa: Bes 2010; Nikopolis: Angeli 2007.

The evidence for the Aegean-Southwest reflects the more detailed chronological developments as shown in Figure 18. During the 1st century BC, the main ware coming in was ESA, with minor quantities of ESC and ESD. This situation changed considerably from the Augustan period onwards. ESA increased, and ESB emerged, yet it is ITS that dominated the market for imported tablewares during the Augustan-Flavian era (this pattern, again, is influenced by the evidence from Corinth). ESA fell off sharply towards the mid-1st century, after which ITS also began to diminish, whereas ESB maintained a steady growth. ESC and ESD266 were sparsely represented during the entire 265  266 

33

See Slane 2004. As, for instance, at Athens (Hayes 2008a, 53-54) and Argos (Abadie-

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:35 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East

Figure 17. Aegean-Southwest: map showing the ware proportions of the different sites and surveys (© Philip Bes/ICRATES Project).

Figure 18. Aegean-Southwest: line chart showing the development, in absolute numbers, of the different wares (n=776) (© Philip Bes/ICRATES Project).

34

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:35 UTC All use subject to https://about.jstor.org/terms

The Late Hellenistic and Early Roman Periods: Discussion and Chronological Phasing period; small quantities of ARSW turned up in the course of the 2nd century.

A similarly dated deposit on Tenos contains ESA (2) and Campana Ware (1).278 Imports datable to the 1st century BC are rare at Argos, represented by Campana Ware and ESA.279 Somewhat later, towards the mid-1st century BC, ESA is encountered on Delos (related to destruction activities in 88 BC, and destruction and subsequent abandonment in or just after 69 BC), and started to arrive at Athens in quantity from 75-50 BC onwards.280 Even if deposits datable to this period are scarce, this period hardly figures in the published record for Corinth, and a contrast of some interest emerges between Corinth and the sites mentioned above. This may well be associated with the focus of exchange patterns, which may largely (though certainly not completely)281 have ignored Corinth following its destruction in 146 BC. However, its refounding as a colonia in 44 or 42 BC probably saw the renewal or intensification of economic and other contacts, (re)establishing a role both as consumer and port of transit between west and east.

Notwithstanding the fact that this region presents a relatively dense pattern of published evidence (Figure 17), a discussion of relevant archaeological deposits and wider literature provides a more detailed picture. Corinth offers a good starting point, with a deposit thought to contain debris that resulted from the destruction of the city by Mummius in 146 BC (which serves as an historical terminus post quem). This deposit includes a complete ESA vessel (Hayes form 16), a both rare and exceptionally early occurrence.267 A late 2nd-early 1st century BC deposit contains mostly local or regional tablewares, and although no ESA was noted, importantly enough it contains Italian tablewares (Campana Wares) and amphorae.268 Only towards the later 1st century BC does the evidence increase, which probably reflects the urban development of Corinth at the time: two deposits of the later 1st century BC and early 1st century contain mainly ESA and the much less common ITS, besides several other imported wares: the ITS appears to echo the situation of Corinth at that time.269 In a deposit dated to the third decade of the 1st century, ITS had developed into a major imported tableware at Corinth (or at least in that deposit);270 ESA and the newly appearing ESB attain 15% and 11% respectively during the first quarter of the 1st century.271 ITS remains common up to the third quarter of the 1st century, when ESB starts to gain in importance; ESC and ESD remain scarce.272 In essence, this scenario recurs in the South Stoa deposits published by Hayes: ITS dominates, followed by ESB and ESA, while ESC and ESD constitute only a minor part of the imported tablewares.273 The importation of ITS continues into the 2nd century,274 but by the middle of that century several small deposits indicate that by now ESB had become the major imported tableware.275 ARSW makes its first sporadic appearances from the early 2nd century onwards, though during the second half of the 2nd century it is only ESC that is identified with certainty.276

This situation changed considerably from the later 1st century BC onwards with the arrival of ITS. At Argos, ESA comprised about 63-65% of the tablewares, although overall quantities seem to have been fairly low.282 At Athens, ESA dominates in several early 1st-century deposits in which ITS is sparsely represented, though these do contain noticeable quantities of Italian Thin-Walled Wares.283 The so-called ‘Underwater Deposit’ at Kenchreai, Corinth’s eastern harbour, broadly dated to c. 1-75, predominantly contains ESA, whilst ITS and Italian Thin-Walled Wares play only a secondary role.284 At Tenos, the 1st-century groups Fa and Fb contain mainly ESA, followed by ESC and ESD, but no ITS.285 At Olympia, several contexts of the 1st century associated with the so-called Southwest Building mainly contain ITS, a picture not different from slightly later deposits of the second half of the 1st century.286 In fact, ITS is most common at Olympia in the third quarter of the 1st century.287 A contemporary deposit at Argos also contains mostly ITS.288 As a matter of fact, ITS comprises c. 45% of the (imported) tablewares at Argos during the 1st century, even if the majority seems to postdate the first quarter of that century.289 Groups G layer II-b, and group M layers II and III at Athens (that both run into the 2nd century) reflect the growing significance

What does the evidence from the other sites and surveys contribute, and to what degree can similarities or differences be observed? A late 2nd-, early 1st-century BC deposit from Athens contains several ESA vessels where Athenian products dominate, and Thompson’ group E and Robinson’s group G, layer I present a similar situation.277

Etienne and Braun 1986, group Ea, 220-222, figs 119-120. Abadie-Reynal 2005a, 37-39; for ESA at Argos in a 1st-century BC deposit, Croissant 1971, 749-750 (fig. 22). Abadie-Reynal 2007, 60, 63, for some very late 1st-century BC ITS, though no ITS apparently predates c. 15-10 BC. 280  Brun and Brunet 1997; for Athens: Hayes 1996a, 10. 281  Bald-Romano 1994. 282  Abadie-Reynal 2005a, 39, 2007, 88. 283  Robinson’s groups F and G layer II-a; for the Italian Thin-Walled Wares, see Rotroff 1997b; ESA reached a peak in the Augustan period: Hayes 1996a, 10. 284  Adamsheck 1979, 44-45. 285  Etienne and Braun 1986, groups Fa and Fb, 224-227, fig. 122. 286  Martin 2001; Schauer 2001; Sinn et al. 1993. 287  Martin 1997b, 2004; see Schauer 2008, 232 for a recently published Consp34 fragment. 288  Abadie-Reynal 1995, 296-297; Consp3 is rather common among the repertoire at Argos: Abadie-Reynal 2007, 62. 289  Abadie-Reynal 2007, 60, 63. 278  279 

Reynal 2007, 133-134). 267  Robinson 1969, 27-28, manhole 10, fill IV. 268  Bald-Romano 1994, though now see Slane 2008b, 237, n. 3. 269  Slane 1986; Williams II and Zervos 1982; Slane 2004 for early ITS, referring to several Augustan deposits. 270  Slane 2004, 32, ITS being most common in the second quarter of the 1st century. 271  Slane 1989, 221-222. 272  Slane 1986, 1989, 222. 273  Hayes 1973a. Some debate concerns these deposits, originally dated to the reign of Nero. 274  Slane 1987, 2004, 32. 275  A manhole published by Williams II and Fisher 1975; pit 1982-2 published by Williams II and Zervos 1983; building 3 and test trench A in Williams II and Zervos 1985. 276  Slane 1989, 223-224. 277  Vogeikoff-Brogan 2000; Thompson 1934; Robinson 1959, 22-24.

35

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:35 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East of ESB towards the later 1st century. At this point, ESB prevails, accompanied by small quantities of contemporary tablewares, including ITS.290 These late 1st-century deposits are the preamble to patterns that characterise the 2nd century. At Argos, ESB is the major tableware till the mid-2nd century, when ESC takes over, whilst ARSW arrives only sporadically.291 A similar scenario applies to Athens where ESB becomes very frequent from the mid-1st century onwards, whilst contemporary so-called Late ITS is rather scarce.292 Group Ga from Tenos, dated to 100-150, contains but few fragments among which ESC is most common.293 A deposit of the mid-2nd century from Sparta also contains very few tablewares, and of the four fragments two belong to ESC, whereas another, only slightly broader dated deposit primarily contains ESB.294 Two deposits found at Isthmia and Corinth with small quantities of pottery and dated close to the mid-2nd century, contain only ESB.295 Finally, ESC is the commonest tableware in a deposit at Argos of the second half of the 2nd century,296 while group Gb from Tenos is more mixed.297

Figure 19. Aegean-North: pie chart showing the proportions of the different wares (n=2053) (© Philip Bes/ICRATES Project).

This overview mainly focused on deposits published from the major (urban) sites in this region, because these are relatively plentiful and thus offer some quantity for intersite comparison. The map shown in Figure 17 provides some insight into the distribution of imported tablewares in the Aegean-Southwest beyond the (major) urban framework, making it clear that ITS was not merely an urban and coastal phenomenon.298 Survey finds hardly contribute chronologically, though it is worth noting that in Boeotia ITS, ESA, ESB and ESC (and a few ESD and early ARSW) are all represented, particularly at Tanagra in eastern Boeotia.299

(Figure 19) are in marked contrast with those from Crete and the Aegean-Southwest. In the Aegean-North, products from Ephesos/Maeander Valley (15%), but above all those from the region of Pergamon dominated the import of tablewares (c. 74%). ITS and ESA arrived only in small numbers, whereas the token presence of ESD is in keeping with the evidence from the other Aegean regions and Africa-Cyrenaica. Apart from Abdera and Stobi, a pattern emerges on the map shown in Figure 20. ESC thrived at Abdera, Amphipolis, Assos, Smintheion, Thasos and Troia, admittedly with certain relative differences between the sites. This is followed by generally smaller percentages of ESB, while ITS and ESA generally remain uncommon. Although the percentages differ from site to site, this can be taken as the basic pattern. At Abdera ESB is not all uncommon, though with the predominance of imports from western Asia Minor it still matches the pattern observed for the other sites; ESB and ESC were also attested at Demetrias.301 At Stobi, however, completely different proportions are noted: ESB and ITS dominated, whereas ESA and ESC are not common. The high percentage of ITS (c. 47%) suggests that the site, situated along the Via Egnatia, was well-located with regard to Adriatic exchange patterns,302 as well as Aegean and, to a lesser extent, eastern supply routes. In summary, however, imported tablewares from western Asia Minor clearly dominated the Aegean-North.

3.2.4 Aegean-North Compared to the Aegean-Southwest, the number of sites in this region (which covers the northern Aegean and northwestern Turkey) from which evidence was collected is small. A total of 2053 finds (all from excavations) was gathered, and pertains to the following sites: Abdera, Alexandria Troas, Amphipolis, Assos, Kepia, Samothrake, Smintheion, Stobi, Thasos and Troia.300 The proportions Robinson 1959, 22-23. Abadie-Reynal 2005a, 43-47, 2007, 128, 170-171. Hayes 1996a, 10, 2008a, 32, 44. 293  Etienne and Braun 1986, group Ga, 227-230, figs 123-124. 294  Bailey 1993, deposits RSC 2W 4119, 240-242, fig. 12, and RSC 1SW 4113, 228-230, fig. 6. 295  Isthmia: Marty 1993, deposit 90-1; Corinth: Williams II and Fisher 1975. 296  Abadie-Reynal 1995. 297  Etienne and Braun 1986, group Gb, 227-228, 230, fig. 124. 298  Bes 2010; Bes accepted a. 299  For a full discussion, see Bes accepted a. 300  Abdera and Amphipolis: Malamidou 2005 (also now see Papaioannou 2010); Alexandria Troas: Japp 2007; Assos: Zelle 1997; Kepia: Malamidou 2005; Samothrake: Love 1969; Smintheion: Akyürek 1992; Stobi: Anderson-Stojanović 1992; Thasos: Abadie-Reynal and Sodini 1992; Malamidou 2005; Troia: Hayes 1995a; Kozal 2001; Tekkök et al. 2001. Also see Tekkök 1996, which unfortunately could not be consulted 290  291  292 

in time. Hayes (1992) reports residual LH-ER terra sigillata fragments found at Saraçhane. 301  Eiwanger 1981, cat. nos IVA.1-12, including Hayes forms 58, 80, possibly also 77; for possible ESB, also see cat. nos III.120-136, IIIa.31, Tafel 37, including later examples of Hayes form 60? For the ESC, cf. infra, 5.3.4. 302  Slane 1996a.

36

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:35 UTC All use subject to https://about.jstor.org/terms

The Late Hellenistic and Early Roman Periods: Discussion and Chronological Phasing

Figure 20. Aegean-North: map showing the ware proportions of the different sites (© Philip Bes/ICRATES Project).

Figure 21. Aegean-North: line chart showing the development, in absolute numbers, of the different wares (n=1548) (© Philip Bes/ICRATES Project).

37

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:35 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East Figure 21 covers the chronological development of the different wares in more detail. This line chart reflects both the general proportions of wares and the site-based evidence, yet does not represent well the patterns of Stobi and Abdera. Products from the area of Pergamon dominated the picture from the early 1st century BC onwards, and reached a peak in the first half of the 1st century. A decrease then followed around the mid-1st century and the subsequent increase was generally to last until the late 2nd century. Clearly, as far as ESC is concerned, the pattern here is altogether different from those for Crete and the Aegean-Southwest, whereas the evidence for ITS and ESB for the three regions basically matches, even though collected totals differ. ESB, the imported tableware second in importance, is mainly represented by forms of the later series, Hayes’ ESB II, possibly to be associated with the temporarily dwindling of ESC. ITS mostly occurs in the late Augustan-Claudian period. ESA occurred only rarely in the 1st centuries BC and AD, and ARSW sparsely appeared by the later 2nd century.

terra sigillata (2%), ESB (6%), ESA (2%) and several other categories.309 Of the six pieces catalogued from part of a late Augustan well fill from Troia, three are ESB, one is ITS.310 Finally, the catalogue (‘a representative assemblage of materials’ [!]) from another well fill, dated to the early 2nd century, mostly included ESC and ESB; other (some possibly contemporary) classes of tablewares are represented by no more than one or two fragments.311 3.2.5 Aegean-East This region, the last of four that make up the Aegean, covers western Asia Minor from Pergamon southwards and includes the Datça/Knidian peninsula; the Troad was assigned to the Aegean-North. The evidence collected comes from two clusters of sites. The first is centred on the wider area of Pergamon (Methymna, Pergamon, Emporio, Sardis),312 while the second covers Notion, Ephesos, Samos (including Kastro Tigani), Priene, Labraunda, Iasos, Didyma and Halikarnassos.313 A total of 1359 finds was collected (Figure 22).314

Archaeologically speaking, the material from Assos, a major share of the collected evidence, is found in disturbed deposits and thus has no stratigraphical clues to offer as to when wares other than ESC occurred, although their context (a necropolis) is worthwhile from a ceramic-functional point of view. In fact, one of the vessels from Samothrake, Çandarlı Ware (ESC) Hayes form 3, was also found in a burial context;303 further likely examples of the attestion of ESC in burial contexts comes from Kenchreai.304 In general, however, published deposits from this region are rather scarce and really only available for Stobi and Troia. As no single deposits with some quantity of tableware material are available from Stobi, lots that were similarly dated are grouped, so as to see what these signify regarding the import of tablewares.305 Interestingly, a number of lots point out that Stobi received Italian Thin-Walled Wares during the 1st centuries BC and AD.306 It has been suggested that this was related to Stobi’s location on the fringes of republican territories, where it acted as an important centre for communications and military control.307 ESA appeared sporadically by the mid-1st century BC. ITS is found both in deposits of the Augustan-Tiberian period and in post-50 lots, when it is apparently residual. This pattern continues in post-100 lots, although ESB clearly increased by this time. It nevertheless remains quite difficult to gauge the archaeological quality of the lots from Stobi. This situation appears to be better for Troia. Here, ESC is prevalent in five deposits dated to the Augustan-Flavian period, with very small numbers of other classes of tablewares.308 This basically reflects a wider quantification program for Troia, showing the predominance of ESC (c. 86%) over western

ESD conforms to the Aegean as a whole, and rarely occurs. ITS (c. 9%) is much less common in comparison to Crete and the Aegean-Southwest, although it does occur at most sites. ESA represents about 15% of the collected data, though the proportions of ESA between sites differs substantially. Aegean tablewares again predominate, accounting for c. three-quarters: ESC represents 10.6%, ESB is the most common terra sigillata with over 64%. Tekkök 2003, 237. Hayes 1995a. Tekkök et al. 2001, well C29, 347-351, 372-373. 312  Methymna: Lamb and Pryce 1940; Pergamon: Bounegru and Erdemgil 1998; Schäfer 1962; Emporio: Boardman 1989; Sardis: Rautman 1995a; Wrabetz 1977; a scatter of finds come from the Byzantine shops at Sardis (Stephens Crawford 1990). The cistern fills from Pergamon published by Meyer-Schlichtmann (1988) could not satisfactorily be incorporated, as the author provides only general hints about the popularity of forms. The gathered evidence nevertheless provides sufficient proof that ESC dominated the tableware market at Pergamon. 313  Notion: Demangel and Laumonier 1923; Ephesos: Beyll 1993; Gassner 1997; Heberdey 1906; Ladstätter 2003; Meriç 2002; MitsopoulosLeon 1991; Outschar 1997, 2000; Zabehlicky-Scheffenegger 1996. Ephesos sees a steady and valuable publication output in recent years that could, unfortunately, not be included here; Samos: Technau 1929; TölleKastenbein 1974 (Kastro Tigani); Unterkircher 1983; Priene: Zahn 1904; Labraunda: Hellström 1965; Iasos: Berti 2000; Didyma: Wintermeyer 1980; Halikarnassos: Vaag 2002. See Berndt 2003 for evidence from Miletos, regrettably noted too late. 314  The recently published, thorough work on the Knidian manufacture of ‘Feinkeramik’ during late Hellenistic and early imperial times also contains valuable observations on the imported sigillata wares, and as such fills an important geographical gap in our knowledge. ESA was the dominant import during the late Hellenistic period, when ESC also seems to have been fairly common, particularly in group G, dated to the first third of the 1st century BC. The import of ESA fades after the mid-1st century, and ESC is rarely noted during that century. ESB is occasionally noted in the Augustan-Tiberian period, but takes over the role as dominant import between c. 50 and 150, a period when other imported sigillata wares are in fact uncommon. ITS occurs in a variety of forms, yet its arrival at Knidos practically stops after the mid-1st century. Finally, the scarcity of ESD reflects the general pattern for the Aegean. Kögler 2010, 369-370. Despite the problems the author was presented with concerning the material, some indications of the relative presence of imports is given, backed up by the archaeology (63-80): ‘[s]o ist in den bisher eingehender analysierten, normal durchmischten Fundkomplexen in der Regel ein Importanteil von deutlich über 5% zu beobachten’ (35). 309  310  311 

Love 1969, 196-197. Rife et al. 2007, 171, fig. 23 for ESC (and ESB). 305  Clearly this is a less than ideal practice, also because the problem arises — a problem with a more general character — that often a deposit is dated on ceramic finds, tablewares in particular. 306  Anderson-Stojanović 1992, e.g. lots 134, 250, 552, 1094 and 1636. 307  Anderson-Stojanović 1992, xxi. 308  Kozal 2001, 327-230, figs 1-6, especially the so-called ‘Street Dump’. The area only saw renewed occupation from Augustan times onwards. 303  304 

38

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:35 UTC All use subject to https://about.jstor.org/terms

The Late Hellenistic and Early Roman Periods: Discussion and Chronological Phasing Figure 23 appears to allow for a subdivision of this region into a northern and a southern sphere. In the northern part, which includes Emporio and Pergamon, it is ESC that dominates the scene, although it appears that ESB is relatively common at Mytilene, perhaps testifying to its export up north, including the Black Sea.315 Even though evidence from sites other than Pergamon remains rather scant at present, the data from Pergamon indicates the near-monopolistic position the ware held for centuries.316 Also at Mytilene, for example, ESC appears to have been very common.317 Yet again, insufficient evidence is presently available to fully tackle the question of what happened outside Pergamon. At Sardis, however, both late Hellenistic ESC and appliqué ware from (the area of) Pergamon are attested, both, it appears, in some quantity.318 Furthermore, ESC prevails among the recently published evidence from the ancient agora of Smyrna (besides ESA, ESB, ITS and Gaulish Sigillata), which also helps to look into the question how far south the dominance of ESC extended.319

Figure 22. Aegean-East: pie chart showing the proportions of the different wares (n=1362) (© Philip Bes/ICRATES Project).

Figure 23. Aegean-East: map showing the ware proportions of the different sites (© Philip Bes/ICRATES Project).

Williams and Williams 1989, 190. Meyer-Schlichtmann 1988. Williams and Williams 1989, 171, 173-174, 178, 1990, 189-190. 318  Rotroff and Oliver 2003, 84-87, 152-166. ‘[t]he impression is that’ ESC ‘was in fact imported in some quantity’. 319  Erol 2007. 315  316  317 

39

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:35 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East

Figure 24. Aegean-East: line chart showing the development, in absolute numbers, of the collected evidence (n=648) (© Philip Bes/ICRATES Project).

ESB, manufactured in Ephesos, Tralles and possibly other sites in the Maeander Valley, appears to have been the most commonly-used tableware further south, even if also here site-based differences can be observed. Recent evidence from Knidos suggests that ESA and ESB were the predominant sigillata imports during the late Hellenistic period and the period between 50-150 respectively. Whereas one might expect ESA to be not uncommon in these parts, considering the area of manufacture and the routes along which ESA (may have) travelled, the evidence is too diffuse at present to substantiate this claim. Moreover, tablewares manufactured at or around Knidos had a considerable regional impact.320 However, ESA clearly predated the introduction of ESB (Figure 24). Last, ITS and ESC do occur in this southern zone but clearly did not play a major quantitative role.321 This generally echoes what has been observed for excavated material from Halikarnassos.322 All in all, the present evidence does little to trace the clear outlines of a north-south pattern, and to demarcate regional economic patterning in this period (cf. infra, 3.2.9, 5.3.9, for similar sub-regional cases).

the quantities of ITS were much lower here compared to Crete and the Aegean-Southwest, chronologically the import of ITS into the Aegean-East compares rather well: the available evidence indicates that it spanned the midAugustan and (early) Tiberian periods.324 ESC and ARSW were generally uncommon, but again, ESC dominated the northern part of this region. ESD is near-absent: an unidentified fragment has been found at Ephesos, and it has been noted at Knidos.325 The development of ESB fits these developments rather well. One would be inclined to think that with the production of ESB the import of ESA, ITS and other tablewares into this region would have (largely) ceased: in other words, local production was creating a threshold which would make the import of other tablewares unnecessary. Indeed, Figure 24 can be read as such. ESB began to be manufactured around 30-20 BC; it quickly reached a peak towards the mid-1st century. Interestingly, thereafter ESB began to dwindle, which is in marked contrast with virtually all other regions, where ESB in fact increased from the mid-1st century onward; perhaps these two developments, if true, were related in a way that presently escapes us.326 Despite the dominance of the local Knidian products, recent evidence from Knidos, however, indicates that ESB was the dominant imported sigillata between c. 50 and 150.327

The chronological development of the different wares is illustrated in Figure 24. This line chart helps to clear some of the questions concerning the occurrence of, for example, ESA. Indeed, ESA played a considerable role but its import was largely confined to the late Hellenistic and Augustan periods; thereafter it arrived only sporadically.323 Although

These observations find support by looking at the collected evidence on the level of the archaeological deposit in

Kögler 2010. For Ephesos, Zabehlicky-Scheffenegger 2004; Martin 2003. 322  Hansen 2003, 198. 323  E.g. Gassner 1997, 121-122; Kögler 2010, 370.

Zabehlicky-Scheffenegger 2004, 73. Kögler 2010, 554, fig. 49, no. Hb.63. 326  Lund 2003, 129-130. 327  Kögler 2010, 369.

320 

324 

321 

325 

40

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:35 UTC All use subject to https://about.jstor.org/terms

The Late Hellenistic and Early Roman Periods: Discussion and Chronological Phasing Ephesos. The oldest relevant deposit dates to the last third of the 2nd century BC, and contains a fragment of ESA form 4A, which is probably one of the earliest specimens recorded in the Aegean.328 A next fixed point is a deposit sequence found in a well in room SR 9C of Hanghaus 1 at Ephesos.329 Particularly the middle deposit, dated to around 100 BC, contains mouldmade bowls of local manufacture as well as a small yet significant quantity of ESA. A roughly contemporary deposit from Halikarnassos contains three fragments of early forms of ESA, consequently expanding the geographical reach of early ESA beyond Ephesos.330 The four specimens of the early ESA form 17B (c. 150-100 BC) found at Samos (Kastro Tigani), are certainly worth mentioning here even though their archaeological context is not known.331 The upper part of the deposit sequence found in the well in room SR 9C, dated to c. the second half of the 1st century BC, in addition to ESA marks the first (modest?) appearance of ESB.332 Ephesos has recently furnished further evidence: the Augustan period, more specifically the last two decades of the 1st century BC, illustrate the continued dominance of ESA (84%) but also the appearance of ESB (11%), which finds itself on the threshold of becoming the major terra sigillata in Ephesos.333 Still somewhat later, Füllung A of a well on the Staatsmarkt is fairly representative of the variety of tablewares in circulation in the Augustan period, and mirrors Figure 24. Besides considerable quantities of Ephesian (?) vessels, ESA, ESB, ESC as well as ITS are identified there, which provides another benchmark for the early distribution of ITS as well as the early phase of ESB.334 This generally echoes a recently published mid-Augustan complex from Priene, which contains ESA and rather early forms of ESB and ITS.335 One of the most significant deposits found at Ephesos with respect to tablewares is the large pottery dump found on the Tetragonos Agora, which is thought to have been damaged by fire and subsequently disposed of.336 It dates to the midto late Augustan period, and comprises a large quantity of ESB, as well as small quantities of ITS and (residual) ESA. This deposit not only advances the understanding of the typo-chronological and economic framework of ESB, since a range of new shapes (including closed vessels) and a large number of stamps of C. SENT(IVS) was identified; but, equally important, it illustrates the quantitative position ESB had by now acquired.

ESB, though imported terra sigillata continued to arrive. On the other hand, ESB forms only a minority in a small deposit that is roughly contemporary with B2.338 Füllung B3, of the third quarter of the 1st century, is a further sign of the matured phase of ESB, as is a recently published, contemporary deposit from Hanghaus 2, in which ESB comprises 90% of the terra sigillata;339 ESA and ESC are present in only small numbers.340 The same applies to stratum 8 from the Südtor on the Tetragonos Agora, wherein ESB quantitatively overshadows other classes of terra sigillata.341 Only ESB and ESC can be identified in Füllung D1, of the late 1st century; ESC now accounts for a considerable share of the catalogued items.342 Füllung D2, dated to the early 2nd century, does not hold much material but continues the pattern of Füllung D1. 3.2.6 Asia Minor-West Central Asia Minor-West Central is poorly represented regarding both the quantity of the evidence and the number of sites from which this is drawn. Material was collected from four sites: Aizanoi, Amorion, Antiocheia ad Pisidiam and Kozluca;343 the proportions are shown in Figure 25.

Figure 25. Asia Minor-West Central: pie chart showing the proportions of the different wares (n=42) (© Philip Bes/ICRATES Project).

This makes it ever so difficult to delineate the (artificial) boundaries of this region, but it roughly pertains to sites west of the line Anemorion-Ankyra, thereby excluding the south coast (cf. infra, 3.2.7). Within the small total the majority of the finds belong to ESB, with only a token presence of ESA and ESC, which suggests that these sites had a primarily western orientation as far as imported tablewares are concerned.

For the 1st century developments can mainly be traced through a sequence of well fills from the Staatsmarkt.337 Füllungen B1 and B2, of the early 1st and the second quarter of the 1st century respectively, further document the role of Gassner 1997, 70. Ladstätter 2003, 40, 46-51. 330  Vaag 2002, 37-38; the (surrounding) stratigraphy underwent disturbances during several periods. 331  Tölle-Kastenbein 1974, 159-160, fig. 264. 332  Ladstätter 2003, 42-43. 333  Ladstätter 2010b, 90-91, 99. Terra sigillata fragments make up only 0.7% of the pottery from Bauphase 1. 334  Meriç 2002, see e.g. K129-133, K173, K176-177. 335  Fenn 2008. 336  Zabehlicky-Scheffenegger 1995, 1996. 337  Meriç 2002. 328  329 

Gassner 1997, stratum 7. Ladstätter 2010a, 179-188, Fundkomplex A-BII/6-SR 8. Other classes of ‘Tafelgeschirr’ (43% of 2700 fragments) identified are ESA, ESC, ITS and South Gaulish. Also see Ladstätter 2008, 98. 340  Meriç 2002, i.a. 60-61, 65-66. 341  Gassner 1997, esp. 130-132, 134-135, 243-248. 342  Meriç 2002, 66-67. 343  Aizanoi: Ateş 2001; Amorion: Tomber 1992, 1993b; Antiocheia ad Pisidiam: Arslan 2003; Kozluca: unpublished. 338  339 

41

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:35 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East

Figure 26. Asia Minor-West Central, -East Central and -South Coast: map showing the ware proportions of the different sites and surveys (© Philip Bes/ICRATES Project).

Figure 26 (which also includes the two following regions), however, allows for a cautious and tentative geographical distinction. A single ESA fragment has been recognised at the southernmost site, Kozluca, where locally-/regionallymade tablewares catered for the region’s needs, also during the Roman period.344 To this can be added a handful of imported fragments from Sagalassos, including a wellpreserved skyphos in Pergamene Appliqué Ware found in an Augustan deposit.345 The small quantities at the other three sites practically all originated in western Asia Minor, suggesting a more western orientation regarding the importation of tablewares.

little can be added. In a sondage at Aizanoi, dated to the mid-1st century, the only imported tableware is ESB.346 3.2.7 Asia Minor-South Coast This region stretches from the Datça/Knidian peninsula eastwards to include Pamphylia; Cilicia is defined separately (cf. infra, 3.2.9). Also for this region only a very small quantity of evidence was collected, from two sites. In consideration of the number of harbours, both large and small, this is particularly regrettable, as these surely played an important role as ports of call and centres for the further distribution and redistribution of goods.

Given the small size of the collected evidence, a line chart sketching the chronological development of the wares and any statements resulting from it are deemed highly unreliable, and also from an archaeological point of view

The collected data, 45 entries in total, derives from Xanthos and Perge (Figure 26):347 ESA (n=17) dominates, followed by ESB (n=7), ESC (n=6), ITS (n=5), ARSW (n=6) and ESD (n=4). Whilst ESA is thus the most common class of imported tableware, one might expect a higher quantity

Poblome 1999. From the Augustan period onwards, the local workshops at Sagalassos manufacturing SRSW basically monopolised the market for centuries to come. A similar model applies to the preAugustan period: Van der Enden et al. 2014, in press. 345  Van der Enden et al. 2014, 83, 88, table 1, fig. 6.

Ateş 2001, 326-332. The presence of ESB form 80 (or a forerunner?) suggests the deposit date is not fixed. 347  Perge: Atık 1995; Fırat 2003; Xanthos: Armstrong 2005; Pellegrino 2003. Terra sigillata finds now published from Limyra show that ESA is most common (43.9%), followed by ESB (25.5%): Yener-Marksteiner 2012.

344 

346 

42

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:35 UTC All use subject to https://about.jstor.org/terms

The Late Hellenistic and Early Roman Periods: Discussion and Chronological Phasing of ESD given the proximity, notably of Pamphylia, to Cyprus; this does appear to be the case during the late Roman period (chapters 5-6). Also here the archaeological background is meagre: a deposit from Perge, dated to the Augustan period, contains four ESA fragments.348 3.2.8 Asia Minor-East Central Again only a relatively small quantity of evidence could be collected, from Arsameia am Nymphaios, Karamildan, Kululu, Lidar Höyük, Porsuk and Samosata, totalling 277 finds.349 These sites cover a large region, with the sites dispersed throughout, although the collected data presents a uniform picture, and conforms well with regions discussed below (cf. infra, 3.2.10-13). Figure 27 illustrates the proportions: with c. 96% of the finds, ESA forms the majority, followed by ITS (3.6%), ESB (c. 0.4%) and ESD (0.4%). Figure 26 shows that ESB is found only at the westernmost site in this region, Porsuk; eastwards, ESA reaches a nearmonopoly with the exception of Lidar Höyük, where nearly a dozen fragments of ITS are identified.350 Evidence from Tavium351 suggests that this site may fall within the eastern reaches of the inland distribution of ESB — whereas the other sites belonged to the core zone of ESA (cf. infra, 3.2.10-13, and eastern Cyprus) — and links up quite well with some sites in Asia Minor-West Central (where ESB dominated).

Figure 27. Asia Minor-East Central: pie chart showing the proportions of the different wares (n=277) (© Philip Bes/ICRATES Project).

ESA appears to have been most common during the 1st century BC, and began to dwindle gradually during the Augustan period; 2nd-century forms are attested, though the evidence suggests that ESA played but a minor role then. That this region could have belonged to the core zone of ESA is suggested by the occurrence of early forms, which are not uncommon. Unfortunately, the absence of good archaeological deposits makes it impossible to investigate whether early forms flowed in soon after ESA began to be manufactured. 3.2.9 Asia Minor-Cilicia Asia Minor-Cilicia roughly covers the ancient provincia of Cilicia. After the three previous regions, the quantity of the collected evidence again allows a detailed discussion of the distribution and proportions of the different classes of tablewares. This is also the first of four regions that are characterised by the preponderance of ESA.

Figure 28. Asia Minor-Cilicia: pie chart showing the proportions of the different wares (n=813) (© Philip Bes/ICRATES Project).

Important evidence was gathered from: Anemorion, Diokaesareia, Domuztepe, Küçük Burnaz and Tarsos.352 The total amounts to 813 entries, with proportions shown

in Figure 28. What catches the eye most is the prevalence of ESA, with c. 82%. ITS come second with nearly 8%, followed by ESD with just over 6%. The combined percentages for the other three wares of this period, ESB, ESC and ARSW, is just less than 4%. Given the assumed areas of manufacture, one would indeed expect ESA and ESD to supply the bulk for the regional consumption of tablewares.

Atık 1995, 68-69, 71, sondage VI, schicht 8. Arsameia am Nymphaios: Jones and Goell 1963; Karamildan: Sevin and Derin 1989; Kululu: Jones 1969 [1971]; Lidar Höyük: Kazenwadel 1995; Porsuk: Abadie-Reynal 1989b; Samosata: Zoroğlu 1989. Some ESA is now reported from Tyana: Mazzocchin 2008. 350  The ITS from Lidar Höyük includes some unusual profiles. 351  Weber-Hiden 2003. 352  Anemorion: Williams 1989; Diokaesareia: Kramer 2005; Domuztepe: Rossiter and Freed 1991; Küçük Burnaz: Tobin 2004; Tarsos: Jones 1950. 348  349 

43

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:35 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East

Figure 29. Asia Minor-Cilicia: map showing the ware proportions of the different sites and surveys (© Philip Bes/ICRATES Project).

The pottery from the survey at Diokaesareia was fully published recently. It seems that ESA (n=279) represents c. 90% of the Hellenistic and (early) imperial terra sigillata, whilst ESB (n=4), ESC (n=4?), ESD (n=20) and ITS (?, n=1) roughly make up the remaining 10%. This is in general accordance with the pattern for (eastern) Cilicia.353

altogether different sphere of exchange as far as terra sigillata is concerned. The chronological evidence (Figure 30) further illustrates these different patterns. From the mid-2nd century BC onwards, ESA increasingly put its stamp on the tableware market, and experienced a very strong phase of growth during the early Roman period, and was most common during the third quarter/second half of the 1st century. Afterwards, quantities dwindled and ESA is thought to disappear from supraregional exchange patterns around c. 200.355 The discussion of the collected evidence from the previous regions indicates that ESA reached its peak during the reigns of Augustus and Tiberius; here, this peak differs with about 1-2 generations. Causes may be sought in the nature of the published evidence, that is, sites or excavations (or indeed the selection of material for publication) that are chronologically limited, as opposed to sites or excavations that were active for the entire period of manufacture and distribution of ESA. Alternatively, it results from the increasing inflow of ESD from the reign of Augustus onwards. In fact, Anemorion closely

The site-based evidence for this region (Figure 29) reveals a difference otherwise not apparent. The westernmost pie chart, documenting the proportions for Anemorion, is dominated by ESD, although ESA is about as common. Only at Tarsos, one of Cilicia’s major urban centres, do we detect a modest presence of other wares. However, from there on a considerably different pattern is observed that should be seen as part of the core zone of ESA, in which other wares did penetrate but generally in (very) small numbers.354 ESA, as seen, does occur in previous regions, yet it never attained such high percentages as here. Moreover, ESD, though present in considerable quantities at Anemorion, was rarely identified in the Aegean regions. It appears that Asia Minor-Cilicia thus belonged to an

353  354 

Kramer 2012, 13-17. Meyza 1995, 179.

Reynolds 2010, 90 for the continuation of ESA into the 3rd century (in Beirut); its supraregional exchange probably petered out by c. 200.

355 

44

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:35 UTC All use subject to https://about.jstor.org/terms

The Late Hellenistic and Early Roman Periods: Discussion and Chronological Phasing

Figure 30. Asia Minor-Cilicia: line chart showing the development, in absolute numbers, of the different wares (n=457) (© Philip Bes/ICRATES Project).

matches western Cyprus, the supposed origin of ESD; as was pointed out in connection to Figure 29, this pattern between ESA and ESD only applies to Anemorion. ITS occurred in small quantities mostly during the first half of the 1st century, which corresponds to the general pattern. ESB, ESC and ARSW made only sporadic appearances. Despite the small quantity, it is noteworthy that ESB was marginally more frequent in the second half of the 1st century, and as such accords with its distribution outside the Aegean-East.

Hellenistic Unit, dated to the mid-2nd century BC, also contains a few fragments of ESA.359 The large HellenisticRoman Unit, whose date range was for the most part based on stamped amphora handles and coins to between the mid-2nd and mid-1st century BC, contains only ESA.360 ESA clearly maintains its major position in four deposits dated to the late 1st century BC and the first half of the 1st century, although other wares begin to appear in small quantities;361 this situation did not alter by the late 1st, early 2nd century.362 Last, the tablewares catalogued from the so-called Complex and Ovens deposits (dated to the 2nd and early 3rd century) no longer need to be considered residual.363

In spite of the considerable body of evidence and the interesting aspects that it presents, the archaeological evidence emerges somewhat meagre. For Diokaesareia (also see the final publication),356 Küçük Burnaz and Domuztepe no archaeological contexts are available. The deposits published from Anemorion are generally dated to a range of 100 years or more, and hence do not allow investigation into the development of the different wares in any great detail. This leaves Tarsos that, admittedly, is also not an ideal candidate to approach the appearance and development of the different wares: ‘[i]n the lists of objects associated with the different units, it will be noticed that there is almost always some obviously later material. In trenches which are so riddled with intrusions as the Tarsus ones, such infiltrations are unavoidable and one must make allowance for these or else run into serious difficulties’.357 Still, a single fragment of ESA form 2 appears in a deposit dated to c. 300-175 BC,358 and the Late

3.2.10 Levant-Coastal North This region delineates the coastal zone between modernday Iskenderun and Beirut, and extends inland to include Epiphaneia and Apamea. Beirut, centred on the Levantine coast, is thus included here and marks the boundary between this region and the Levant-Coastal South (cf. infra, 3.2.12). The evidence was collected from ‘Ain Dara, Antioch, Apamea, Beirut, Byblos, Epiphaneia, Gindaros, Leukos Limen, Mutatio Heldua and Tell ‘Arqa.364 The implies that the ESA fragment is intrusive. 359  Goldman 1950, 14-15, 31. 360  Jones 1950, esp. 231-238. 361  ‘Circuit Wall Deposit (North)’; ‘Circuit Wall Deposit (South)’; ‘Fill under Grave 38-1’; ‘Roman Fill’. 362  The ‘Trench 6’ and ‘Concrete Wall Chambers’ deposits. 363  See, again, Reynolds 2010, 90 for the continuation of ESA into the 3rd century. 364  ‘Ain Dara: McClellan 1999; Antioch: Comfort 1948; Waagé 1934, 1948; Apamea: Vanderhoeven 1989; Beirut: Reynolds 1997-1998,

Kramer 2012. Goldman 1950, 29, in particular the ‘Hellenistic-Roman Unit’ discussed on 15, 31-32, 36-37. 358  Jones 1950, 231, ‘Middle Hellenistic Unit’. If its dating is correct it 356  357 

45

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:35 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East period towards the mid-1st century BC, ESA maintained a gradual increase right up to the Augustan-Tiberian period, when it was most common. A mostly uninterrupted dwindling can be noted from the mid-1st century onwards, until the current typo-chronological framework ends by the later 2nd century. ITS was mostly common during the Augustan-Tiberian period, and thus conforms to the general picture. ESB, ESC, ESD and ARSW occurred only sporadically, and played no quantitatively significant role. As for the previous region, little archaeological background is available. On the other hand, the paucity of most other wares leaves little room for discussing their quantitative significance. Lund’s re-examination of the Epiphaneia material adjusted some of the original building phases, and drew attention to a particular aspect in the development of imported tablewares by noting that after c. 100 ‘the number of fine ware decreased dramatically. This was not only the case with Eastern Sigillata A but also with Cypriot Sigillata’.367 Whether or not this resulted from a natural disaster in the early 2nd century, it is more important to see any wider implications of Lund’s observation on the quantitative distribution of tablewares, particularly for the eastern Mediterranean on the whole. In two occupation levels identified at Tell ‘Arqa, disturbed though to a certain extent by later activities, ESA appears in niveau 7.368 Small quantities of ESA were noted in couches 7 and 7B, dated to c. 125-1 BC and 125-50 BC respectively. ESA continued to appear in couches 7A and 7A’, both dated to the later 1st century BC; here ITS appears with single specimens of Consp18 and 22.369 Finally, a deposit found at Beirut clarifies some of the quantitative developments. This cistern fill, dated to c. 65-75,370 includes a large quantity of slipped tablewares (657 RBH) of which c. 80% (by vessel count, including closed vessels), is ESA, about 12% ITS, and c. 8% ESD (of vessel count). Despite a considerable number of restorable vessels, Reynolds argues in favour of secondary deposition, material initially discarded nearby after use and thrown into the cistern only at a later stage. This explains why the material is chronologically not homogeneous: the majority belongs to the second quarter of the 1st century, whereas a second component is of Neronian date. This does not mean that during Nero’s reign ITS still arrived at Beirut in considerable quantities; the quantity of ITS is nonetheless considerable in comparison to other sites.

Figure 31. Levant-Coastal North: pie chart showing the proportions of the different wares (n=3336) (© Philip Bes/ICRATES Project).

proportions are shown in Figure 31, based on a total of 3336 fragments. Here, with nearly 93% ESA is even more common in comparison to Asia Minor-Cilicia. ITS is the most common other class with nearly 6%, whereas the combined evidence for ESB, ESC, ESD and ARSW constitutes c. 1.5%. The site-based map (Figure 32), therefore, does not come as a great surprise. Whereas the predominance of ESA recurs at all sites, ITS is relatively more common at Antioch and Beirut than at any of the other sites: ITS also occurs at Apamea, Byblos, Epiphaneia and Tell ‘Arqa. At Beirut, ESD also reaches a considerable percentage, notably in the cistern published by Reynolds.365 This ware is otherwise rare, occurring only in (very) small quantities at Antioch, Byblos, Epiphaneia, Gindaros and Kinet Höyük.366 Interestingly, the Aegean tablewares ESB and ESC are thus far only identified at Antioch and Gindaros, the latter located in Antioch’s hinterland. ARSW is represented by a few early Roman forms at Antioch. The chronological development further illustrates the ESA-dominated pattern (Figure 33), and basically conforms to that observed for Asia Minor-Cilicia. From the later 2nd century BC onwards, with a brief, wavering

3.2.11 Levant-Interior North This region roughly covers the tract of the Euphrates on which Zeugma is located, to continue downstream to Dura Europos, the easternmost site discussed here. Most sites are in fact located along, on or close to the Euphrates; to this were added Sultantepe and Tell

2004a; Byblos: Lauffray 1940; Montet 1928; Epiphaneia: Johansen 1971; Christensen 1986; Ploug 1985; Lund 1995; Gindaros: Kramer 2004; Leukos Limen: Stucky 1983; Mutatio Heldua: Gebara 1982; Tell ‘Arqa: Thalmann 1978. 365  Reynolds 2004a. 366  Vanderhoeven 1989, 154, no. 900, is close to ESD form P22A. Several other drawings are also not immediately reminiscent of the standard repertoire of ESA, which of course does not mean these are not ESA. At Kinet Höyük (the Hellenistic pottery is studied by the author and Jeroen Poblome), ESD is rare in extremis. ESA is near-dominant, though colourcoated predecessors may have continued in production/use alongside ESA for some time.

Lund 1995, 145. Thalmann 1978, 51-53, 58-59, 69-70; couches 7 and 8 were identified on the absence of ESA in the latter. 369  Thalmann 1978, 59, 126-129, figs 41-42; the profiles are slightly unusual for these forms. 370  For the dating, Reynolds 2004a, 121-122, 125, 129-130. 367  368 

46

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:35 UTC All use subject to https://about.jstor.org/terms

The Late Hellenistic and Early Roman Periods: Discussion and Chronological Phasing

Figure 32. Levant-Coastal North: map showing the ware proportions of the different sites (© Philip Bes/ICRATES Project).

Fakhariyah, the latter possibly to be identified with the colonia of Resaina. A dividing line was drawn between this region and Asia Minor-East Central; however, the evidence from both regions throws an interesting light on the distribution of tablewares (ESA in particular) into landlocked areas. Besides Dura Europos, evidence has been included from Athis/Neokaesareia (Athis hereafter), Doliche, Qal’at Gabbar, Qusair as-Saila/Tetrapyrgium, Resafa, Soura, Sultantepe, Tall Šēh Hamad/Magdala, Tell Fakhariyah, Tell Rifa’at and surroundings, Umm el-Tlel and Zeugma.371 The proportions shown in Figure 34 are based on the accumulated total (n=307); also here ESA is by far the most common tableware. The small numbers of ITS, ESB and ESD should not be seen entirely like a fish out of water: these also occur at for instance Antioch. They are, especially in such inland areas, probably to be seen as occasional arrivals. In fact, the Euphrates made the dispersion of goods relatively easy. Figure 35 also shows the dominant position of ESA, which conforms to the previous regions.

The recently fully published finds from Jebel Khalid, that was abandoned in the late 70s or early 60s BC, also shows a dominance of ESA during the late Hellenistic period. A very small quantity of ESD and a single ITS fragment were also identified, the latter obviously not related to the Hellenistic phase.372 A scatter of ESA and, possibly, a few ESB fragments from Tell Barri (Kahat) in northeast Syria extend the distribution pattern of ESA further east.373 The chronological-quantitative developments are illustrated in Figure 36.374 ESA basically shows a similar pattern as that observed for the Levant-Coastal North. The question remains, of course, whether one can make comparisons considering the quantitative differences. Interestingly, the single sherd reported from Tall Šēh Hamad, on the river Habur, can be placed in a broader context. In a series of surveys and excavations at a number of tells along this river, ESA was identified at four out of the 128 sites. 372  Clarke 2011, xxxi; Jackson and Tidmarsh 2011, 280, 326-356, 357359. Only one mouldmade bowl fragment in ESA was noted; the analyses (cf. supra, 2.2.1) and the observation, however, that the majority of these fragments belong to Fabrics 2 and 3 points to a similar source for most mouldmade bowls as well as the ESA (359-360, with ns). 373  Parmegiani 1987. Also see the recently published results of a survey carried out west of the Euphrates, in northernmost Syria: Newson 2014, esp. 5. 374  The line chart contains nearly only ESA; fragments of other wares, except for a single 1st-century BC ESD, could not be typo-chronologically identified.

371  Athis: Harper 1980; Doliche: Höpken et al. 2008; Dura Europos: Cox 1949; Jebel Khalid: Tidmarsh 2005 (also see n. 372); Qal’at Gabbar: Konrad 1996; Qusair as-Saila (Tetrapyrgium): Konrad 1996; Resafa: Konrad 1992, 1996; Soura: Konrad 1996; Sultantepe: Lloyd 1954; Tall Šēh Hamad: Römer-Strehl 2000; Tell Fakhariyah: Kantor 1958; Tell Rifa’at survey: Kenrick 1978, 1981; Umm el-Tlel: Majcherek and Taha 2004; Zeugma: Gschwind 2002.

47

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:35 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East

Figure 33. Levant-Coastal North: line chart showing the development, in absolute numbers, of the different wares (n=1508) (© Philip Bes/ICRATES Project).

basic dating evidence for these ‘Schichten’ is unclear, though the forms offer some clues. ESA was classified according to the Samaria-Sebaste typology, and whilst no quantitative information is available, a concordance to Hayes’ typo-chronology confirms a more than haphazard occurrence of 1st-century ESA forms, in addition to some earlier forms.376 These sites possibly formed part of another stage in the distribution east of the Euphratian sites. The archaeological context again cannot make any important contributions. Dura Europos has a broad typological spectrum, with forms of both Hayes’ late Hellenistic series as well as those of the 1st century.377 This not only indicates that Dura Europos was already in existence prior to the Roman imperial period, its strategic location along the Euphrates surely attracted economic activity of some volume and significance. At Athis, on the other hand, the forms identified are all post-Augustan,378 and the identified forms from Qusair as-Saila echo this situation.379 Also, at Sultantepe, among the identified forms more than half belong to the 1st century. Furthermore, the evidence from four tells along the river Habur partly matches the evidence from Athis and Resafa. All ESA fragments from the survey at and around Tell Rifa’at

Figure 34. Levant-Interior North: pie chart showing the proportions of the different wares (n=307) (© Philip Bes/ICRATES Project).

Although this in itself is not significant, it does add to the understanding of the geographical distribution of ESA eastwards; that ‘[d]ie Mehrzahl der Funde stammt aber aus Schichten des ersten nachchristlichen Jahrhunderts’375 serves the present purpose. Whether ESA provided the 375 

The author lists (Samaria) forms 1 (Hayes forms 2-4), 2 (Hayes form 5), 14 (Hayes form 36-37), 17 (Hayes form 18), 18 (Hayes form ?), 19 (Hayes form 19), 22 (Hayes form 42) and 23 (Hayes form 45/47). 377  Cox 1949. 378  Harper 1980, 329-331: Hayes forms 36, 40A, 47, 49, 51, 56, 57/60 and tarda-a. 379  Konrad 1996, 2001, 165-168, 178-179: Hayes forms 35/37, 37, 46/47 and 51. 376 

Römer 1995, 353.

48

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:35 UTC All use subject to https://about.jstor.org/terms

The Late Hellenistic and Early Roman Periods: Discussion and Chronological Phasing

Figure 35. Levant-Interior North: map showing the ware proportions of the different sites (© Philip Bes/ICRATES Project).

Figure 36. Levant-Interior North: line chart showing the development, in absolute numbers, of the different wares (n=213) (© Philip Bes/ICRATES Project).

49

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:35 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East ‘are attributable to types of the first century B.C. and the first A.D., and none to the second century A.D.’;380 two fragments of ITS and a possible fragment of ESB were also found.381 A small deposit has been published from Resafa dated to c. 70-80;382 the few forms identified comply with Athis and Qusair as-Saila, and a deposit sealed by a mosaic floor at Zeugma, dated to the second half of the 1st century, also includes only post-Augustan forms.383 Apart from Doliche, Dura Europos and the Tell Rifa’at survey, virtually all other sites present post-Augustan forms. The location of these sites along the Euphrates, at that time part of the eastern fringe of the Empire, suggests that these were of certain strategic/military significance, and were (partly) supported through specific exchange mechanisms. The near-homogeneous pattern, particularly with regard to ESA, differs to some extent from the southern Levant. 3.2.12 Levant-Coastal South This region runs south from Beirut to Sinai, and the eastern boundary runs roughly N-S somewhat west of the river Jordan. Data was collected from Ashdod, Ashkelon, Caesarea Maritima, Dor, Horvat ‘Eleq, Jalame, Oumm ‘el-Amed, Porphyreon, Samaria-Sebaste, Tel Anafa and Tel Mevorakh,384 and totals 771 finds (Figure 37). ESA maintains its prime position with c. 88%.385 ITS and ESD are the best represented of the other wares, with c. 3% and c. 6% respectively, whilst ESB, ESC and ARSW combined barely reach 3%.

Figure 37. Levant-Coastal South: pie chart showing the proportions of the different wares (n=775) (© Philip Bes/ICRATES Project).

Roman period. Actually, the same holds true for recently published finds from Caesarea Maritima, published by the same author. ESA and ESD are the most popular tablewares in early imperial Palestine; other wares, such as ITS, ESB, ESC and early forms of ARSW occur but not regularly.388 To the Caesarea Maritima evidence can be added an interesting collection from Ramat Hanadiv — ESA is most common, followed by ESD and a handful of ITS fragments, thus mirroring the general trends of this area — that sheds light on the non-urban distribution of several contemporary wares, thus including ITS.389

If the recently published typological catalogue from Ashkelon is in any way representative of more general proportions of imported tablewares386 at the site, it roughly mirrors the proportions illustrated in Figure 37. ESA in fact predominates, probably followed by ESD, both represented by a broad typo-chronological range; ESD, however, might be more common than the 6% in Figure 37 (‘found [...] in large amounts’). ITS (with a limited yet interesting range of forms that includes several chalices), ESB and early ARSW are attested yet do not appear to be very common.387 However, given the typological design of this catalogue, any interpretation based on it is likely to be unreliable, both for this as well as the mid- and late

The map (Figure 38) also bears out these proportions. Samaria-Sebaste and Tel Anafa stand out quantitatively; the full quantification from Tel Anafa shows that ESA comprises about 97-99% of the tableware finds, and Tel Anafa forms another important landmark in mapping the distribution of ESA. Such full quantifications are particularly helpful in the interpretation of the distribution of classes of tablewares.

Kenrick 1981, 439. Kenrick 1981, 446. 382  Konrad 1992, 355-356: Hayes forms 37A, 51 and 53. 383  Gschwind 2002, 335, 354: Hayes forms 37A-B, 47-48, 51 and 60B. 384  Ashdod: Dothan and Freedman 1967; Ashkelon: Carmi et al. 1994; Caesarea Maritima: Berlin 1992; Riley 1975; Dor: RosenthalHeginbottom 1995; Horvat ‘Eleq: Berlin 2000; Jalame: Johnson 1988; Oumm ‘el-Amed: Dunand and Duru 1962; Porphyreon: Domżalski 2002; Samaria-Sebaste: Kenyon 1957; Crowfoot 1957; Tel Anafa: Slane 1997; Tel Mevorakh: Rosenthal 1978. 385  Note that ESA, possibly with a geographically distinct pattern, was noted in a survey of the Upper Galilee, yet almost no further information is provided (though see table 3.5): Frankel et al. 2001, 63-64, fig. 3.10, pl. 32. 386  Johnson 2008a. Among the ‘Various Early Roman Fine Wares’ (chapter 3), in addition to the identified SRSW (33, nos 105-106; morphologically these are not convincing), is a possible early imperial SRSW form (23, no. 75), that very well matches SRSW form 1B191 (Poblome 1999, 370-371, fig. 33). 387  Johnson 2008a, 5-18, 25-30, 34-41, also the concluding remarks on 197. 380  381 

Chronologically, ESA mainly pertained to the late Hellenistic period, with a peak during the AugustanTiberian period (Figure 39). Whereas one could argue that the evidence from Tel Anafa, a site where occupation is believed to have ended by the mid-1st century, distorts these results (especially if it were fully included), it does comply with the general pattern for this region: quantities of ESA sharply diminished after the Augustan-Tiberian period. This is in contrast to the regions previously discussed, where ESA dwindled more gradually and/ Johnson 2008b, 21-22, 25-26, 30-32, 35-38, 58-59, 116. Silberstein 2000, 454, 456-461, 464-465, pls XVII-XX, figs 20, 24; for a summary, 465-466. 388  389 

50

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:35 UTC All use subject to https://about.jstor.org/terms

The Late Hellenistic and Early Roman Periods: Discussion and Chronological Phasing

Figure 38. Levant-Coastal South: map showing the ware proportions of the different sites (© Philip Bes/ICRATES Project).

Figure 39. Levant-Coastal South: line chart showing the development, in absolute numbers, of the different wares (n=577) (© Philip Bes/ICRATES Project).

51

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:35 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East or later, and where quantities remained high(er) in the post-Augustan period. Notably the sharp drop-off after the Augustan period suggests that ESA forms inspired by ITS (directly or indirectly) were comparatively scarce. These do, however, occur at Porphyreon (Jiyé, in southern Lebanon).390 Whereas the absence of ESA postdating the mid-1st century at Tel Anafa reflects the end of occupation, their scarcity in the Levant-Coastal South more generally should be sought in other factors. More importantly, whether this pattern stands alone also remains to be seen, as does the effect of newly-published evidence. ESA nevertheless clearly overshadowed the other wares.

to Hayes’ late Hellenistic series,395 significant in mapping the extent of the distribution of these early forms. ESA appears in the pre-Herodian phase at SamariaSebaste, of which the chronology is not entirely fixed.396 The rebuilding of Samaria by Gabinius during 57-55 BC provides a terminus ante quem397 when ESA was already in use, represented by Hayes forms 3 and 22A. A second phase (Roman I) is dated between the reconstruction around 57-55 BC, with the Herodian reconstruction of the town c. 30-20 BC as a terminus ante quem. This phase actually consists of a number of single deposits, several of which are discussed in more detail. Besides ESA, no other wares were identified in this Roman I-phase.398 The subsequent phase, Roman Ia, is dated to the late 1st century BC and early 1st century based on parallels from Haltern and Oberaden. Although ESA still dominates, a fragment of ITS reflects the arrival of other wares by this time.399 However, the attribution by the authors of, for example, Hayes forms 36 and 37, dated c. 60-100, to this phase, appears problematic.400

In so far archaeological contexts have something to offer, a small quantity of tablewares was found in a well at Ashkelon, accompanied by Pompeian Red Ware. The authors’ main aim was to investigate the chronological relation between the wooden logs used in the well’s construction, and the pottery found in it.391 The pottery (or at least the tablewares) seems to fall into a fairly narrow time-frame, whereas the log was dated to c. 110 BC by dendrochronological methods. However, the authors overlooked the fact that a date obtained by dendrochronological method pinpoints (within certain limits) the date of felling the tree, which means that it can have had a long life-span (elsewhere), as seems to be the case here.

Finally, a range of new deposits or chronological groups comes from Tel Anafa, where throughout the second half of the 2nd century BC up to the end of occupation around the mid-1st century, ESA was just about the only tableware in use. Although some of the deposits do concern sealed primary deposits, dating evidence other than tablewares is not as rich as one might wish. It proved possible nevertheless to construct a sequence of deposits; for the present purpose it seems important to discuss in which chronological phases tablewares other than ESA occur. The earliest deposits contain several specimens of Campana Ware in the phases HELL-2A and HELL-2B/ C,401 when early products from the region of Pergamon also found their way to the site.402 Occasional specimens of both wares turn up, perhaps residual, in phases of the later 1st century BC and the early 1st century. It is also in these phases, ROM 1A-1B, as well as ROM-1C of the early to mid-1st century, that ESD is securely attested.403 Very small quantities of ITS and ESB are also found during these phases.404 Interestingly, Italian Thin-Walled Wares, represented by a few dozen fragments, are in fact more common than ITS. The chronological and quantitative discrepancy of these two wares is food for thought concerning the traditional focus on the economic and functional role of terra sigillata. Both Tel Anafa and Tel Kedesh (cf. supra, 2.2.1) are significant with regard to the study of BSP and (early) ESA.

A handful of deposits, invariably dated to between c. 15050 BC, from Ashdod contain small quantities of a restricted number of ESA forms: 3, 4A and 24.392 A deposit from Caesarea Maritima exemplifies the general pattern for this region. Riley’s level H4B, excavated within the city’s hippodrome, is dated to the first half of the 2nd century on the absence of tablewares post-dating the mid-2nd century.393 His quantified overview of level H4B (n=174) points out that ESA is most common (c. 80%), followed by ITS (almost 6%), ESD (about 3%), and miscellaneous c. 11% (including ESB, and Pompeian Red Ware). Indeed, identified forms of ESA and other wares are datable to the 1st century or later, which contrasts somewhat with the chronological development as discussed above. A second deposit from this harbour-city highlights the older end of the terra sigillata spectrum. A levelling fill in the Temple area contains pottery that mostly (c. 95%) dates to the mid- to late 2nd century BC, the remainder is of the late 1st century BC. The actual date of deposition is placed in the later 1st century BC: most of the forms represented and their fragmentation point to secondary deposition.394 The pottery could have originated in houses of 2nd-century BC date near the Temple platform, and the ESA forms belong

Probably including BSP, 113-116, e.g. fig. 52.3. Note the presence of ESA form 3, associated with floor levels of houses thought to have been destroyed during the Herodian reconstruction, which in fact comprises the Roman I phase, 285, 291-293. 397  Kenyon 1957, 284, 288-289. 398  See Crowfoot 1957, for instance 292-295, 310-315, 325-328. 399  See fig. 68.5 for ITS, OCK2109, dated c. 20 BC-5. 400  Kenyon and Crowfoot agreed on the date of this phase, as the latter attributed the pottery in fig. 68 to the Augustan period, 289-290 (Kenyon), 307-308 (Crowfoot). 401  Slane 1997, FW455-457, FW459-460. 402  Slane 1997, FW495-459, FW508. 403  See e.g. FW576-582, FW584-585. 404  For example FW595-596 (ITS), FW597-599 (ESB). 395  396 

Domżalski 2002. Carmi et al. 1994. 392  Dothan and Freedman 1967. 393  Riley (1975, 43) pointed out that the date of the deposit enters the 2nd century because of an ESA fragment with barbotine decoration. Note, however, that the date range of several forms actually also runs into the 2nd century, or falls into that century: Hayes’ form tarda-a was noted twice, and single specimens of forms 60A and 61. 394  Berlin 1992, 112-119, esp. 112. 390  391 

52

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:35 UTC All use subject to https://about.jstor.org/terms

The Late Hellenistic and Early Roman Periods: Discussion and Chronological Phasing 3.2.13 Levant-Interior South This region encompasses the Sinai and the areas on either side of the Dead Sea, continuing north to the Sea of Galilee. A total of 997 finds was collected (Figure 40) from: Damascus, Gadara, Gerasa, Hammath Tiberias, Hippos(-Sussita), Jericho, Jerusalem, Kallirhoe, Kanatha, Mampsis, Nessana, Oboda, Pella, Petra, Philadelphia, Rehovot-in-the-Negev and Scythopolis.405 Proportions are not very different compared to the previous four regions, and also here ESA is the most regularly imported tableware, accounting for c. 78% of the finds, considerably less than for the whole northern Levant yet still dominant; ITS and ESD comprise c. 11% and 9% respectively. The recent and thorough study of the pottery from Gamla, which includes a discussion of the functional-spatial differentiation in the distribution of the pottery within the excavated areas, identifies ESA as being ‘very well represented’.406 The finds from Nessana, Oboda, Mampsis, and Petra in particular, illustrate how far south the quantitative superiority of ESA continued (Figure 41). That ESD accounts for c. 9% of the finds, together with ESB, ESC and early ARSW (together c. 2%), generally agrees with what has been observed for the previous four regions. Interestingly, ITS accounts for a quite remarkable 11%; higher percentages were noted only much further west, in Africa-Cyrenaica, Crete and the Aegean-Southwest. Jerusalem, Oboda and Petra in general have considerable quantities of imported wares other than ESA, notably ITS, which accounts for c. 24% and 17% respectively at the latter two sites. Whilst this might reflect a preference in selecting ITS for publication, the fact that ITS is rather common at Oboda and Petra, located well in the south of the Levant, need not be coincidental, and it will be argued in chapter 4 that the marked presence of ITS here could very well be economically significant.

Figure 40. Levant-Interior South: pie chart showing the proportions of the different wares (n=997) (© Philip Bes/ICRATES Project).

impact of ESB, ESC and early ARSW was negligible, although illustrate the geographical reach and economic integration of these areas concerning these wares. Last, the archaeological background provides several relevant additions. The first concerns a fill in or around a foundation trench along the city wall at Gadara. It includes small quantities of ESA, and the destruction of the wall is attributed to Alexander Jannaeus around 100 BC.407 An older deposit at Hippos, in a sondage in shops surrounding the Forum and thought to date to the first half of the 2nd century BC, contains a fragment of BSP; a second fragment was found in a deposit dated to the 2nd century BC.408 The upper part of this sondage, dated to the early 1st century BC, contains a few fragments of ESA. This picture continues with a less homogeneous deposit, dated to the 1st century BC and into the early 1st century, in which ESA dominates.409 A broadly-dated deposit at Hammath Tiberias (c. 20 BC and 135) contains two fragments of ESA and one of ESD.410 Another deposit from Hippos, the so-called North Wall Assemblage that dates to the first half of the 1st century, contains only ESA.411 More recently published reports on the pottery from Hippos roughly show the same trends: a dominance of ESA, with a token presence of ESD and BSP.412 Jerusalem, where the picture appears mixed yet intriguing, offers more evidence. A sizeable quantity

The chronological development (Figure 42) shows that ESA most frequently occurred during the 1st centuries BC and AD. It attained a peak during the late AugustanTiberian period, after which quantities began to dwindle, yet in comparison to the Levant-Coastal South, quantities here maintained a higher level, and began to withdraw only towards the later 1st century. This, in fact, matches the general eastern Mediterranean pattern of ESA a bit more closely. Both ESD and ITS were most common during the Augustan-Tiberian period. ITS, however, was comparatively more common, and even appeared in some quantity around the mid-1st century. The quantitative 405  Damascus: Damascus 2004; Gadara: Kenrick 2000; Gerasa: Welles 1938; Fisher 1938; Hammath Tiberias: Dothan 1983; Hippos: Mlynarczyk 2001, 2002, 2003, 2004, 2005a; Jericho: Pritchard 1952-1954 [1958]; Jerusalem: Hayes 1985b (now also see Hayes 2008b); RosenthalHeginbottom 2005; Kallirhoe: Clamer 1997; Kanatha: Henrich 2002; Mampsis: Negev and Sivan 1977; Nessana: Baly 1962; Oboda: Negev 1972, 1974; Pella: Tidmarsh 1990; Petra: Horsfield and Horsfield 1942; Khairy 1990; Schneider (C.) 1996; Sieler 2004; Philadelphia: Koutsoukou and Najjar 1997; Rehovot-in-the-Negev: Rosenthal-Heginbottom 1988; Scythopolis: Waagé 1934; Comfort and Waagé 1936. 406  Berlin 2006, 14, more generally 13-14, 21-23.

Kenrick 2000, 265. Mlynarczyk 2003, 52, 54. BSP hints at a terminus post quem of c. 180170 BC, when it is thought to have appeared. Possibly, these deposits predate c. 100 BC, when it is thought to have given way completely to ESA. 409  Mlynarczyk 2003, locus 907. A single ESD fragment was noted. 410  Dothan 1983, 11, 18. 411  Mlynarczyk 2001. 412  Mlynarczyk 2006, 2007, 2008. 407  408 

53

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:35 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East

Figure 41. Levant-Interior South: map showing the ware proportions of the different sites (© Philip Bes/ICRATES Project).

Figure 42. Levant-Interior South: line chart showing the development, in absolute numbers, of the different wares (n=841) (© Philip Bes/ICRATES Project).

54

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:35 UTC All use subject to https://about.jstor.org/terms

The Late Hellenistic and Early Roman Periods: Discussion and Chronological Phasing of ESA was found in the Armenian Garden excavations; in spite of the poor preservation of the deposits, the bulk of the ESA belongs to the interval between the late 1st century BC and the First Jewish Revolt of 66-70, when considerable parts of Jerusalem were destroyed.413 ITS, that comes second, occurs mainly during the late Augustan and Tiberian periods. ESA was, however, not found in houses in the Jewish Quarter that were destroyed during the First Jewish Revolt.414 Interestingly, concerning the consumption and use of tablewares, this could point to settlement diversification. As a matter of fact, Berlin’s recent observations on the fact that ESA was no longer being imported into 1st-century Gamla, despite continued occupation, inquire deeper into people’s motivations and choices regarding the socio-cultural context of their material culture, and the changes that these entail.415 Finally, in an early 1st-century BC destruction deposit at Pella, thought to result from Alexander Jannaeus’ sack in 83 or 82 BC, ‘[e]arly examples of Eastern Sigillata A ware included a deep bowl (9) and a fragmentary plate’.416

Figure 43. Africa-Egypt: pie chart showing the proportions of the different wares (n=284) (© Philip Bes/ICRATES Project).

The archaeological evidence, though not very rich and varied, not only shows that ESA had already penetrated the southern fringes of the Levant around the turn of the 2nd and 1st century BC, but also that different consumption patterns could also have existed within one city, here illustrated through Jerusalem. If so, surely Jerusalem was not an isolated case.

common in the southern Levant. ESD, on the other hand, was quite common in the western part of Cilicia Tracheia. The general outline of the distribution of ESD bears great resemblance to that of LRD in the late Roman period (chapters 5-6). Figure 44 provides a better insight into the geographical distribution. Small quantities are seen at Marina el-Alamein, where ESD was the most common class of terra sigillata during the 1st century, at a time when other wares dominated Alexandria.419 The archaeological context of some of the Marina el-Alamein finds, a necropolis, is nevertheless noteworthy. What catches the eye most, however, is the common occurrence, of ITS (next to Alexandria) at Myos Hormos and Berenike; their proportions are exceptional within the eastern Mediterranean.

3.2.14 Africa-Egypt The focus here on Alexandria and two of the Red Sea harbours is aimed not only at investigating the role Alexandria played in interregional exchange, but also its link with the Red Sea harbours. This system formed one end of the trading links that connected the Roman Empire, the Arabian Peninsula and India (and lands further eastwards). The proportions of the evidence gathered, from Alexandria, Berenike, Marina el-Alamein, Myos Hormos, Pelusium and Tell Atrib,417 are shown in Figure 43, and are markedly different from those elsewhere. ESA is the most common — a handful of fragments of ESA and ITS were attested in the Adulis survey further south418 — with c. 41% of the finds (much less than for previous regions), yet is fairly closely followed by ITS with a noteworthy c. 33%. Also ESB is markedly more common (c. 11.6%), whereas ESC is scarce. ESD attains about 12.3% of the finds, rather low in consideration of the distance between Cyprus and Egypt, although the ware was in fact also not

The chronological development is illustrated in Figure 45. Both ESA and ITS were most common during the Augustan-Tiberian period, something which is not seen elsewhere, although ESA is also represented by pre- and post-Augustan forms. ESB is mostly represented by forms of the first half of the 1st century, which is surprising since the distribution of ESB outside the Aegean-East (cf. supra, 3.2.5) mainly belongs to Hayes’ ESB II (cf. infra, Figure 56). ESC was uncommon, restricted to forms of the late Hellenistic period and the first half of the 1st century. ESD was also not common, mainly represented by forms of the 1st and early 2nd century.

Hayes 1985b. Avigad 1976, 21. 415  Berlin 2006, 151-152, particularly concerning table- and cooking wares. 416  Tidmarsh 1990, 73, 75, fig. 10.9; the bowl is similis to Hayes form 5 but most akin to TA type 15. 417  Alexandria: Hayes and Harlaut 2002; Meyza 2005; Élaigne 1998, 2000; Berenike: Hayes 1996b (also see, for instance, Tomber 1998, 1999b); Marina el-Alamein: Daszewski 1990; Myos Hormos: Johnson 1979; Whitcomb 1982; Pelusium: Rodziewicz 1994; Tell Atrib: Mieliwodzki 1995. 418  Peacock 2007, 82-83, fig. 8.4. 413  414 

Archaeologically, ESA is the main terra sigillata in a pre-Augustan deposit at Alexandria, which matches the Levantine pattern. Because of the absence of ITS the terminus ante quem could fall either c. 30 or 10 BC.420 Four 419  420 

55

Daszewski 1995, 31. Élaigne 2004, 133-135, US 3655 (sigillata: MNI=49). Élaigne draws

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:35 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East

Figure 44. Africa-Egypt: map showing the ware proportions of the different sites (© Philip Bes/ICRATES Project).

Figure 45. Africa-Egypt: line chart showing the development, in absolute numbers, of the different wares (n=230) (© Philip Bes/ICRATES Project).

56

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:35 UTC All use subject to https://about.jstor.org/terms

The Late Hellenistic and Early Roman Periods: Discussion and Chronological Phasing Augustan deposits from Alexandria are characterised by the presence of (mainly) ITS and ESB, and with a terminus post quem of c. 30 or 10 BC, one could have a closing date of 30. With small quantities of ESC and ESD, these four deposits represent the line chart quite well.421 In the Tiberian-Claudian period, ITS comprises about a third of the terra sigillata finds in several deposits. ESA remains common, yet ESB and ESD are again slightly more common; ESC is nearabsent.422 Further south, one deposit from Berenike overlaps with the evidence from Alexandria, whilst two others continue the chronological picture. Three loci dated to the second quarter of the 1st century contain almost exclusively ITS, and one ESB fragment.423 Also from Berenike, two loci with a terminus post quem of 50 and with variable termini ante quem yet none later than 100, contain ITS, ESA and ESB; ITS, notwithstanding the small quantities, prevails.424 A deposit of Neronian date from Berenike contains, among the very small quantity of ceramic material, two fragments of ITS and a single fragment of ESB.425 To this can be added material from Koptos: first, a collection of surface pottery whose form-repertoire generally matches Tiberian- and Claudian-period ITS from Alexandria. The main difference, however, is that ITS is much more common than ESA and ESB which, according to the author, ‘semble lié au caractère militaire et commercial du site où étaient installées des garnisons romaines’;426 secondly, from excavated deposits from Koptos comes ITS and ESA, in particular the Roman 1b assemblage in which ‘[a]lmost all of the imported vessels [...] are from Italy.’; the ‘[a]rretine vessels’ date to c. 10 BC, the ESA is considered to come slightly after c. 10 BC.427

Figure 46. Cyprus: pie chart showing the proportions of the different wares (n=1093) (© Philip Bes/ICRATES Project).

The proportions, based on a collected total of 1093 finds, are shown in Figure 46. The dominance of ESA (c. 43%) is somewhat unexpected, since ESD is thought to originate from (western) Cyprus. ESD represents c. 38.6%, thus also accounts for a large share of the evidence. Consequently, ESB (c. 4.6%) and ESC (c. 1.4%) are uncommon, their percentages roughly matching those seen east of the line Lycia-Sinai. ITS (c. 12%), however, is more common here than elsewhere,429 though in no order when compared to Africa-Cyrenaica, Crete and the Aegean-Southwest;430 Alexandria431 and sites along the Red Sea are exceptions to the rule. Last, early specimens of ARSW account for 0.4%. The geographical distribution is illustrated in Figure 47, showing that the western and southern coastal zones are predominantly represented: ‘[o]ur knowledge of the pottery of Hellenistic and Roman Cyprus is lopsided and fragmented. We are best informed about the south western part of the island, thanks to J.W. Hayes’ seminal treatment of the ceramic finds from the House of Dionysos at Nea Paphos, and to publications by Polish and Italian scholars working in the same city. The available evidence from the rest of the island is less complete, and at times chronologically restricted’.432 A range of sites occupied the coastal zones around the island, and recalling the high

Despite the fragmentary evidence, similarities between Alexandria, Myos Hormos and Berenike (and Koptos) can be noted. Also, the few deposits from Alexandria illustrate the development of the different terrae sigillatae within a brief time span. Moreover, with the exception of Africa-Cyrenaica, Crete and the Aegean-Southwest, these deposits also contain relatively more ITS than elsewhere. 3.2.15 Cyprus The archaeological record for late Hellenistic and Roman pottery from Cyprus is largely restricted to its western and southern (coastal) zones: Amathous, Ayios Philon, Kition, Kopetra, Kourion, Limassol, Maroni Petrera, Palaipaphos, Panayia Ematousa, Paphos and Salamis.428 attention to the issue of distinguishing between late Hellenistic and early Roman, when the first arrival of ITS is difficult to define. This can have implications for the interpretation of exchange patterns. 421  Élaigne 2004, 135, US 3669, 4278, 4373, 4502 (sigillata: MNI=29). 422  Élaigne 2004, 135-136, US 3745/3779, 4317/4387 (sigillata: MNI=98). 423  Hayes 1996b, 155, 166, 169, loci 081, 085, 088. 424  Hayes 1996b, 162-163, 168-169, loci 045, 078. 425  Hayes 1996b, 155, 164-165, a specific part of locus 045. 426  Élaigne 2004, 139. 427  Herbert and Berlin 2003, 12, 101-103, figs 79-80. Not all typological identifications (e.g. R1.3, R1.5-6) conform to this (early) chronology. 428  Amathous: Abadie-Reynal 1987; Aupert 1976; Aupert and Hermary 1980; Burkhalter 1987; Ayios Philon: Du Plat Taylor 1980; Du Plat Taylor and Megaw 1981; Kition-Bamboula: Marquié 2004; Nicolaou 1976; Salles 1993, 1995; Kopetra: Rautman 2003; Kourion: Neuru and Soren 1987; Limassol: Karageorghis 1986; Maroni Petrera: Tomber 2002; Palaipaphos: Goethert 1985; Lund 1993; Panayia Ematousa: Lund

1996a, 1998; Paphos: Guidice et al. 1996, 1999; Guidice and Narbone 2000, 2001; Hayes 1991, 2003; Malfitana 1992, 1993, 1994; Malfitana et al. 2004; Meyza and Papuci-Wladyka 1999; Salamis: Diederichs 1980; Munro and Tubbs 1891. 429  But see Malfitana 2004, 112: ‘John W. Hayes has justly expressed the presence of the sparse number of Italian sigillata as “…rather unexpected in Eastern Mediterranean terms…”’. 430  Berenice, Gortyna, Knossos and Corinth strongly influence the ITS percentages for Africa-Cyrenaica, Crete and the Aegean-Southwest. Also consider the distribution of ITS at sites east of the Roman Empire (cf. infra). 431  Lund 2006b, 47, with references to Élaigne 1998, 2000; Malfitana 2002. 432  Lund 2006b, 31, n. 92.

57

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:35 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East

Figure 47. Cyprus: map showing the ware proportions of the different sites and surveys (© Philip Bes/ICRATES Project).

Figure 48. Cyprus: line chart showing the development, in absolute numbers, of the different wares (n=915) (© Philip Bes/ICRATES Project).

58

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:35 UTC All use subject to https://about.jstor.org/terms

The Late Hellenistic and Early Roman Periods: Discussion and Chronological Phasing percentages of ESA and ESD, their distribution reveals a second geographically distinct pattern (cf. supra, 3.2.9). ESD is the major tableware at Kourion and Palaipaphos; at Paphos it stands roughly equal against ESA. These sites (Figure 47) are located along the western and southwestern coast. At Dhiorios, one of the few sites along the north coast from where evidence is known, six out of seven fragments belonged to ESD, one to ESA. The evidence from the Akamas survey agrees with these proportions.433 In addition, recent excavations on Geronisos Island off the western coast of Cyprus, not only showed the dominance of a locally- or regionally-made tableware,434 but somewhat surprisingly that ESA predominates over ESD.435 On the other hand, at Kition and Panayia Ematousa along the south coast, ESA accounts for about three-quarters of the terra sigillata during late Hellenistic and early Roman times. ESA is also most common among a very small collection of published finds from Salamis, on the east coast. At Ayios Philon, on the northeast peninsula, ESA appears to be dominant (yet here quantities are also small), therefore in keeping with this proposed pattern. However, Amathous, Kopetra and Maroni Petrera are not. ESA clearly dominated at Amathous, which is not the case at Maroni Petrera where ESD ‘is well represented’.436 At Kopetra only a few fragments were found: three out of five belonged to ESD,437 and ESD accounts for about half of the finds in the survey of central Cyprus.438

of regionalism. At the same time it furnishes important evidence for the interpretation of exchange patterns within the eastern Mediterranean. Asia Minor-Cilicia also presented a division (cf. supra, 3.2.9) which, in fact, connects well with that for Cyprus. The chronological development is illustrated in Figure 48. ESA had a considerable impact on Cyprus from the later 2nd century BC onwards, which continued during the 1st century BC. ESA witnessed a second phase of growth during the later 1st century BC, and reached a peak in the Augustan-Tiberian period. After a substantial drop off towards the mid-1st century, growth is again noted, but by the end of that century ESA was in decline. ESD, thought to have appeared around 100 BC, remained of secondary importance during the 1st century BC. It was only towards the end of that century that quantities rapidly augmented when, next to ESA, ESD became important during the Augustan-Tiberian period. This was followed by a brief low, yet quantities rose again from the mid-1st century onwards, and ESD remained in fashion till the mid-2nd century, when it rather swiftly dwindled to near-absence. During the second half of the 1st century, though more clearly for the first half of the 2nd century, ESD took over the role that ESA had held as the major terra sigillata. These developments generally comply with the evidence outside Cyprus, even if that distribution was concentrated in a few zones. The fact that ESA was common on Cyprus shortly after its production began, as well as the geographical division, signify that Cyprus belonged to the koinè of ESA.

The evidence from western and eastern Cyprus on the whole mirrors different spheres of interaction regarding the distribution and consumption of tablewares, and probably goods more generally. Given the high occurrence of ESA, eastern Cyprus most likely had stronger ties (economically, culturally, etc.) with the (northern) Levant, whereas western Cyprus was connected to supply patterns that probably centred on Paphos; Paphos’ varied evidence could be explained by its location, facing west.439 Although this idea is not new,440 it offers clues regarding the concept

Both ESC and ARSW were uncommon. ESB, of minor quantity, was mainly represented by forms of Hayes’ ESB II, and fits the pattern outside the Aegean-East. ITS was somewhat more common than in a number of other regions; a second noteworthy aspect indicates that ITS arrived, as elsewhere, during the late Augustan-Tiberian period yet more or less maintained that level towards the end of the 1st century.441

433  Lund 1999, 7: ‘[t]he overall frequency of the fine wares from the 1st c. BC to the 2nd c. AD, Fig. 6, differs markedly from that of Panayia Ematousa. In the Akamas, Cypriot Sigillata constitutes 85% of the material, followed by Eastern Sigillata A, which merely accounts for 12%. The remaining wares - Eastern Sigillata B, Fig. 8.3, and C, Italiantype sigillata etc. - only make up 3% of the Total. Eastern Sigillata A predominated in the 1st c. BC, Fig. 7, but the situation was reversed at the beginning of the 1st c. AD, and the ware was hardly represented after about AD 50, when Cypriot Sigillata had a virtual monopoly on the market.’. 434  Mlynarczyk 2005b. 435  Connelly 2005, 166, 168: ‘while the rest are mostly of imported Eastern Sigillata A fabric’; for a preliminary discussion, see Mlynarczyk 2010. 436  Tomber 2002, 41. Although absolute numbers are not available, and a detailed understanding of the late Hellenistic and early Roman sigillata cannot be reached, the percentages provide certain hints: ITS is represented by one fragment, ESA comprises less than 1% by weight (about two fragments?), ESB less than 2% by weight, whereas ESD comprises ‘c. 15% of all finewares by weight’. 437  Kopetra and Maroni Petrera had substantial late Roman occupation, and the terra sigillata fragments on the whole were residual in post-early Roman levels. The exception is a 2nd-century deposit at Maroni Petrera, apparently not related to the church construction (Tomber 2002, 43-44). 438  Given and Knapp 2003, 60-201; ware counts are provided in tables. 439  See Lund 2006b, who focuses not merely on tablewares. 440  See esp. Lund 1999, 5-9, fig. 9 (19); 2006b, for example 36-37, 41, figs 5, 14.

The division regarding the circulation of ESA and ESD on western and eastern Cyprus also has a chronological dimension (Figure 49). It points out that in western Cyprus (which includes Paphos, Palaipaphos442 and Kourion) ESA dwindled after the Augustan period, at a time when ESD became the major terra sigillata. Eastern Cyprus (including Kition, Panayia Ematousa,443 Salamis and Ayios Philon) basically shows a reversed pattern: although ESD increased from the late 1st century BC onwards, ESA generally maintained its position as the major terra sigillata. The archaeological background for Cyprus is quite rich, and sheds further light on matters discussed thus far. Deposits from the House of Dionysos at Paphos span a broad chronological range. Early occurrences of ESA are attested in several deposits that pre-date 100 BC. The first, the lower Malfitana 2004, 110, 112-113. Lund 1993, 92-93. 443  Lund 2006a, 217. 441  442 

59

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:35 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East

Figure 49. Line chart showing the development, in absolute numbers, of ESA and ESD on western and eastern Cyprus (© Philip Bes/ICRATES Project).

part of a well fill dated to the mid-2nd century BC, contains an ESA jug fragment that is, however, possibly intrusive from the fill above.444 In another well, whose upper fill dates to c. 150-130 BC, a possible fragment of BSP was found.445 Two deposits of the late 2nd century BC from Paphos, one from the House of Dionysos, the other from the Saranda Kolones site, contain a fragment of ESD form P20 and ESA form 2B respectively.446 A cistern fill from Paphos, which was excavated by the Polish Mission, is datable to the late 2nd and very early 1st century BC, although the stratigraphy of the cistern clearly represents more than one phase. In any case, ESA dominates the picture.447 Remarkably, a trio of contemporary deposits from the House of Dionysos, though small in quantity, did not contain ESA,448 which is also the case with a slightly more broadly dated deposit.449

changed only slightly during the first half of the 1st century BC. ESA maintains its dominant position in a deposit of c. 110-60 BC, although small quantities of ESD now appear.452 No change is observed until the reign of Augustus: ESA remains the most common terra sigillata in a deposit of 50-30 BC.453 Both ESA and Italian Thin-Walled Wares are identified in a small deposit of the second half of the 1st century BC.454 Things started to change during the Augustan period. ESD is a little more common than ESA in four layers datable to c. 40-10 BC, whilst a contemporary well fill did not contain ESD.455 However, the profound changes of the later Augustan period are echoed in well 18, dated to 1-20, wherein ESD is much more common than ESA, and ITS is now a relatively common class of terra sigillata.456 A deposit of mid-1st century date continues this pattern, as next to ESC and ITS, ESD is more common than ESA.457 Well 4, that contains pottery similar in composition to destruction deposits from Pompeii, contains fragments of all five major terrae sigillatae: ESD remains most common, yet ESB makes its appearance in considerable quantities and ITS is also not uncommon,458 which compares with a small quantity of pottery from three deposits.459 This

Next, a deposit dated to c. 110-90 BC contains ESA, ESC and ESD; the first is most common.450 This pattern is even more obvious in the so-called Quarry Pit, dated to about the last decade of the 2nd century BC. This contains several fragments similar in fabric to ESD; ESA is overwhelmingly present, next to a small quantity of BSP.451 This situation 444  Hayes 1991, 111, a well in room ΓΞ; this particular fill was located between 5/4.5 and 4m, and possibly contained more intrusive pottery. 445  Hayes 1991, 156, describes it as black-slipped. The date apparently pertains to the upper 3m of the fill, and a coin of Ptolemy VIII possibly from this fill could provide a somewhat later terminus post quem. 446  Hayes 1991, 116, yet could be intrusive; Hayes 2003, 471. 447  Meyza and Papuci-Wladyka 1999, 80-83, 90-92. The bottom level, which concerns us here, was superimposed by a level in which a possible coin of Cleopatra VII was found. This, in turn, was covered by debris (?) shovelled in towards the mid-2nd century. 448  Hayes 1991, room ΓY, between 2-1.5m, well 20 (a coin of 44-30 BC is considered intrusive) and room ZT. 449  Hayes 1991, cistern ZP, 16-17. 450  Hayes 1991, room ΓE (3). 451  Hayes 1991, 132-134, 137-138, in room AΛ (published in Hayes

1967). A total of 1553 fragments of ESA was found, and 33 black-slipped ESA fragments, presumably BSP. 452  Hayes 1991, room BZ. 453  Hayes 1991, room E; note that the numismatic evidence provides the main dating evidence. 454  Hayes 2003, deposit 6-LH1. 455  Hayes 1991, room Λ, layers 7-10, and well 6. Hayes points to their contemporaneity. 456  Hayes 1991, well 18. A similar pattern is seen in the ‘Fish-tank’ in room ΓH (dated by the pottery as coins were lacking), and a small deposit under a mosaic (200-201). 457  Hayes 2003, deposit 3C. 458  Hayes 1991, well 4, datable to c. 60-80? 459  Hayes 1991, the later contents of well 3, ΔΙ, and the contents in room

60

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:35 UTC All use subject to https://about.jstor.org/terms

The Late Hellenistic and Early Roman Periods: Discussion and Chronological Phasing pattern persists in several deposits of the first half of the 2nd century, where also considerable quantities of ITS and ESB are noted, and ARSW emerges.460

ideally serve as a sequence of images that illustrate the geographical, chronological and quantitative developments of these tablewares. One note concerns these distribution maps: if at a site/location two or more classes are represented by numbers that fall within the same class, only one class is plotted/visualised, since these are linked to one particular geographic location. This is not necessarily a problematic matter. The distribution maps are not intended as strict analytical tools — to that end, quantities for all classes, phase by phase, are presented in appendix 4a-b — but rather as visual aids in support of the text.

Finally, Hayes presents a large number of very small deposits with variable termini ante and post quem dates but all within the 2nd century. Although it is difficult to detect the proportions, it can be noted that all five major tablewares continue to appear. Some are possibly residual by now, yet Hayes already noted the dissimilar development of ITS on Cyprus compared to most of the eastern Mediterranean, which is further discussed here in chapter 4. ARSW occasionally appears in deposits of the 2nd century, though it remains scarce.461

3.3.1 Phase 1 (c. 150-30 BC)

3.3 The Distribution of Late Hellenistic and Early Roman Terra Sigillata in the Eastern Mediterranean

This phase marks the very starting point of the tradition of terra sigillata. ESA fairly quickly attained a wide distribution, reaching the central Mediterranean by the later 2nd century BC. Not long thereafter, workshops in and around Pergamon and on Cyprus also shifted toward the manufacture of red slip tablewares. Their impact, however, remained mostly regional.

The purpose of this section is to move beyond the regional description, and discuss the geographical and quantitative distribution of the six tablewares within a chronological framework (cf. supra, 1.3.5). Both the chronological and the geographical framework have an obvious artificial element, yet allow for a structured discussion. New evidence, which emerges on a regular basis, is likely to alter (some of the) conclusions reached here. Phase 1 starts with the beginning of production of ESA around the mid2nd century BC, for which the evidence suggests a most likely source in the area between Tarsos and Laodikeia.462 ESA was already distributed on a supraregional scale during this phase, although the earliest forms circulated in a geographically limited zone (cf. infra). Although ESC and ESD accompanied ESA during this phase, these do not occur in great quantities until later. The distribution of the earliest forms of ESA already covers a large, yet still limited zone. This can be captured by dividing phase 1: phase 1a, that captures the distribution of the earliest forms, and phase 1b, illustrating the distribution of forms whose date range spans the later 2nd century BC till the Augustan period. The purpose here is to demonstrate the development in the distribution of ESA during the late Hellenistic period. With regard to the chronology of forms that are assigned to either 1a or 1b, those with a terminus ante quem of c. 75 BC are attributed to phase 1a, those of the later 2nd century BC till the Augustan period to phase 1b. The consequence of this is a certain degree of overlap, not only for the period around 100 BC, but for the Augustan period as well (e.g. Hayes form 4A), which already concerns phase 2. Still it is felt that this (slightly unbalanced) approach can reveal basic changes in the direction and intensity of distribution, and subsequently to explain these changes. In support, the regional evidence discussed in section 3.2 (cf. supra) serves as a framework, to pin down (if available) the presence/ arrival of particular wares and forms. Moreover, each of the four phase-distribution maps captures a particular segment of the late Hellenistic to early Roman periods, and they

Phase 1a (Figure 50) thus shows the distribution of ESA, ESC and ESD forms with a terminus ante quem around 75 BC. Even if it cannot be very precise chronologically, it illustrates the geographical and chronological extent, in particular of ESA. ESC and ESD are quite rare, with no more than a handful of finds. ESC already occurs at Assos and Pergamon, and has a thin distribution beyond that area; worth mentioning are the finds at Alexandria (two of the three fragments were residual in later deposits)463 and Tel Anafa (a single fragment was attributed to phase Hell 2B-2C).464 Finally, ESD is found on Cyprus,465 and a scatter beyond, including Jalame in northern Israel.466 ESA presents an altogether different pattern. During phase 1a it reached a considerable distribution that includes the Levantine coast, and smaller quantities are noted at considerable distances inland. It is unnecessary to look at the archaeological background once more, yet some benchmarks should be highlighted. Early forms of ESA are found in late 2nd century BC (into early 1st) deposits at Paphos, Tel Anafa, Hippos and Tarsos,467 although early forms can be identified at many more sites (Figure 50) for which no contextual information is available. These 463  Élaigne 1998, 78-79, 104; single fragments of forms N1 and S2 were found in a deposit of the first half of the 2nd century BC. A second fragment of form S2 was found in a deposit of the second half of the 1st century BC, which basically still falls within the form’s date range; Élaigne 2000, 23-24. 464  Slane 1997, 358, of the late 2nd and early 1st century BC. 465  At Paphos, eight specimens of forms P15 and P20 were identified: Hayes 1991. Four were found in deposits dated to the decades around 100 BC. 466  Johnson 1988, 141, ESD form P20. 467  Paphos: Hayes 1991: a considerable number of early forms was recovered from the so-called ‘Quarry Pit’ in room AΛ (c. 110-100 BC); also see the cistern deposit published by Meyza and Papuci-Wladyka 1999; Tel Anafa: Slane 1997, in phases HELL 2A-2C (between 125-75 BC); Hippos: Mlynarczyk 2003, in the upper section (c. 100-75 BC) of a sondage carried out in the later shops in the forum; Tarsos: Jones 1950, in the ‘Middle Hellenistic Unit’ (c. 300-175 BC), which appears very (too?) early.

ΓΣ, between 0.5 and 1m. 460  Hayes 1991, e.g. the main and upper fills of cistern 5. 461  Hayes 1991, noting single specimens of forms 8A and 26. 462  Cf. supra, 2.2.1.

61

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:35 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East

Figure 50. Map for phase 1a, showing the distribution, in absolute numbers, of ESA, ESC and ESD forms that predate c. 75 BC (© Philip Bes/ICRATES Project).

deposits support the idea that ESA was not uncommon in the Levant and Cyprus by the later 2nd, and (very) early 1st century BC. In addition, single or a few fragments turn up west of Cyprus, at Knossos, Gortyna, Phalasarna, Corinth, Kenchreai, Berenice, Samos, the Troad, and possibly even Stobi; Argos can now be added to this list.468 Noteworthy are specimens of the rare form 16, attested at Corinth (1), Berenice (2) and perhaps Kenchreai (1). At Corinth, it was found in part of a manhole fill, thought to result from clearing debris following Mummius’ destruction of Corinth in 146 BC.469 This deposit appears to provide the soundest chronological indicator for the (occasional) earliest ESA in the central Mediterranean and, based on the archaeological record presently available, remains the only indicator.470

popularity. These characterise the distribution of ESA throughout the 1st century BC, shown on Figure 51 that represents phase 1b (c. 125/100-30/25 BC). Although the date range of several forms (3-4A, 5A-B, 7-8, 11, 22, 24-25, 101-102) runs into the Augustan period (phase 2), these are included here, since the larger share of their date range falls within phase 1b. The same applies to certain ESD forms (P1-P2, P14, P16, P18, P34, P49). Attributing ESC forms to either phase 1b or 2 is less straightforward, and in line with the chronological boundary (c. 30 BC) between phases 1b and 2, those forms were selected whose date range roughly falls into phase 1b. During phase 1a ESA mainly circulates in the Levant, and is identified in a late 2nd-century BC deposit at Paphos, the by then western-most site where it arrived rather commonly; it made only rare appearances further west. This changed fundamentally from the later 2nd century BC onwards, its distribution now involving the entire eastern Mediterranean (Figure 51), and the ware now regularly appears in the Aegean. Though forms of phase 1a continue into the early 1st century BC, this expanding distribution pattern is characterised by essentially only a few forms (mostly 3, 4A, 22). Innovative yet not too elaborate, highly functional,

Things were about to change by the late 2nd century BC. The ESA workshops began manufacturing several distinctive forms which fairly quickly gained great Abadie-Reynal 2007, 87. Robinson 1969, 27-28. 470  Only poor or no archaeological information is available for the specimens of form 16 at Berenice and Kenchreai, which is also the case for the single example found at Dor in Israel. 468  469 

62

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:35 UTC All use subject to https://about.jstor.org/terms

The Late Hellenistic and Early Roman Periods: Discussion and Chronological Phasing

Figure 51. Map showing the distribution, in absolute numbers, of ESA, ESC and ESD during phase 1b (c. 150-30 BC) (© Philip Bes/ICRATES Project).

these typify late Hellenistic ESA par excellence. A number of deposits characterise the distribution during phase 1b:471 two specimens of ESA form 3 were identified in a deposit of c. 130-110 BC at Alexandria;472 an ESA form 4A was found in a late 2nd century BC deposit at Berenice;473 ESA forms 4A and 22A were found at Halikarnassos, in a deposit of the very early 1st century BC;474 a deposit at Ephesos, of the last third of the 2nd century BC, contains a fragment of ESA form 4A.475 Specimens of forms 3, 4A and 22 were found in a slightly earlier deposit, of c. 110-90 BC;476 fragments of ESA forms 4A and 22A were found among the remains

of an olive press installation at Delos, thought to have operated between the Second and Third Mithridatic Wars; these deposits possibly result from the eventual destruction in 69 BC;477 a cistern fill at Athens, associated with cleanup activities following the sack by Sulla in 86 BC, contains four ESA vessels, forms 3 and 4A.478 In another cistern fill, also thought to contain debris from the events in 86 BC, two specimens of ESA form 3 are noted.479 A third cistern fill, of the early 1st century BC, contained an ESA form 22A;480 ESA constitutes about half of the tablewares in contexts of the early 1st century BC at Argos, if not more according to recently published evidence.481 The form range attested at late Hellenistic Knidos includes 3, 4A and 22A.482 Finally, a lot dated to the first half of the 1st century BC from Stobi contains a fragment of ESA form 4A, illustrating the geographical extent ESA reaches by this time.

Since a number of the deposits presented here are dated to the late 2nd and early 1st century BC, these could distort the view concerning phase 1a, because of chronological overlap. However, phase 1a is mostly characterised by forms whose distribution rarely went beyond the core region of ESA. These new forms typify the expanding distribution of ESA, observable already by the late 2nd century BC. This suggests that certain early forms (e.g. 17B, 20) whose production continued after 125 BC were not exported in large numbers, whereas the new forms (thought to start around 125 BC) were. A cistern fill from Paphos (Meyza and Papuci-Wladyka 1999) illustrates this point: although the stratigraphy consists of several layers, it contains examples of both sets of forms. 472  Hayes and Harlaut 2002, 101, 116. 473  Kenrick 1985, 226. 474  Vaag 2002, 37-38, 210. 475  Gassner 1997, 70. 476  Ladstätter 2003, 50-51. 471 

Brun and Brunet 1997, 592. Vogeikoff-Brogan 2000, 297-298. It is interesting to note the absence of form 22A, one of the more common forms, from a functional point of view: Attic mouldmade bowls offer tempting evidence to explain this. 479  Thompson 1934, 421-422, 433. 480  Robinson 1959, group G, layer I, 23. 481  Abadie-Reynal 2005a, 39, 2007, 88. 482  Kögler 2010, 480-482, Fundkomplex E. 477  478 

63

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:35 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East Whereas this list can be supplemented with more examples, the point should be clear. Two deposits illustrate the distribution of ESA towards the end of phase 1b. A deposit from Paphos datable to c. 50-30 BC, contains specimens of forms 3, 4A and 22A,483 and a ship wrecked off the coast near ancient Phaselis around 50-25 BC, includes forms 3, 4A and 22A.484

überprüft werden’.489 Interestingly, Lund forwarded a similar explanation for the late Hellenistic distribution of ESA and ESD, to suggest that these predominantly took place within the Seleucid and Ptolemaic kingdoms respectively (cf. infra, 4.3.1). The evidence for ESD shows no real difference compared to phase 1a. An increase at Paphos can be observed and it is present at Palaipaphos and Panayia Ematousa. Several deposits from Paphos, notably layers 7-10 in room Λ dated to 40-10 BC, contain a fair number of forms.490 Forms of phase 1b are also attested in older deposits, especially at the House of Dionysos. Outside Cyprus ESD is attested at Anemorion, Tel Anafa and Myos Hormos, and only Tel Anafa provides some additional archaeological evidence. Most fragments are found in loci of phase ROM-1B of the early 1st century, yet an earlier phase (HELL-2C) that contains two fragments is ‘contaminated’.491 This evidence suggests that ESD remained of regional significance during most of the 1st century BC. Considerable change takes place during phase 2 (the early Roman period), when two new classes of terra sigillata appear: ITS and ESB.

Both ESC and ESD display a slight increase in quantity and geographical scope. Single or a few specimens of ESC are now found at Athens, Delos, Tenos, Samos, Ephesos, further east at Alexandria and Tel Anafa, to which can be added recent evidence from Knidos.485 Its distribution in the (eastern) Aegean thus intensified during phase 1b; recent evidence from Sardis at least points out that late Hellenistic ESC was probably not uncommon at the site.486 Occasional specimens are found at destinations further east. Chronologically, a fragment at Athens was found in the first cistern fill referred to earlier, yet the other Aegean examples are all found in later deposits; either these still comply with the date range of the forms, or possibly concern residual pieces, likewise for the fragment at Alexandria, found in a deposit of the second half of the 1st century BC.487 At Tel Anafa, the two fragments of ESC form S3 were found in loci of HELL 2C+-ROM 1A (c. 75 BC-25) and ROM-1A (25 BC-25).488

3.3.2 Phase 2 (c. 30 BC-25/30) After ESC and ESD, but above all ESA experienced quantitative and geographical expansion in the course of the 1st century BC — developments that continued during phase 2 — the later 1st century BC sees two new types of terra sigillata appear on eastern markets. ITS reached distant sources throughout the east, even if its distribution betrays geographical and quantitative particularities, whilst ESB remains mostly of regional significance.

This evidence suggests that ESC increased marginally in the Aegean by the later 2nd century BC, whereas the few examples recorded further east indicate that ESC travelled to those regions sporadically only after the earlier 1st century BC. In this context, the recently published surface pottery from Pisidian Pednelissos is a remarkable case. After SRSW and ESD it is ESC that is the most common class of terra sigillata during the 1st century BC and early 1st century, when ESA is absent. One does expect a common presence of SRSW and ESD, yet ESC is rarely found in the published evidence (otherwise poorly known in general) of this region. Zelle wonders whether ‘[e]ine Erklärungsmöglichkeit für dieses Phänomen könnte vielleicht in alten politisch-kulturellen Verbindungen liegen. Seit dem Frieden von Apameia (188 v. Chr.) gehörte Pisidien zum pergamenischen Königreich, und eine entsprechende Einbindung in das ökonomische System und die kulturelle Einflusssphäre dieses Reiches ist zu erwarten. So wird beispielsweise diskutiert, inwieweit pergamenische Architektur die pisidische beeinflusst hat. Könnte es nicht sein, dass die verstärkte Präsenz Pergamenischer Sigillata selbst in einer Zeit nach der Eingliederung Pergamons in das Imperium Romanum ein Reflex solcher Verbindungen ist? Diese Frage lässt sich derzeit nicht verlässlich klären, doch sollte sie in Zukunft

Phase 2 roughly spans the reigns of the emperors Augustus and Tiberius, the start of a lengthy period of relative calm and prosperity throughout the Empire that has become known as the Pax Romana. This period saw the inception of two new classes of terra sigillata:492 the manufacture of ITS and ESB is thought to have begun around 40 and 25/20 BC respectively, and are both yardsticks in defining the lower chronological limit of this phase. It was argued that the distribution pattern of ESA greatly expands during phase 1b. Although the published archaeological background hinders a detailed account, ESA is common by the late 2nd century BC in the Levant and at Paphos, and by the mid-1st century BC achieves a formidable distribution in the eastern Mediterranean. ESC and ESD remain mostly regionally significant, yet attain a thin, geographically wide, distribution during phase 1, which forms the starting point of phase 2. ESA

Hayes 1991, room E. Mitsopoulos-Leon 1975, 102-105, 109-110. Some 300 vessels are mentioned, of which only part was brought to the surface; whether these are all ESA is unclear. Some (broken) vessels that were not retrieved reportedly have diametres of about 80cm; see also Parker 1992, 109, wreck ‘Cape Gelidonya B’. 485  Kögler 2010, 369-370. 486  Rotroff and Oliver 2003. 487  Élaigne 2000, deposit 1. 488  Slane 1997, FW501-502. 483 

ESA maintains a very wide distribution (Figure 52), much like that during phase 1b, and that it reaches a peak during this period seems generally agreed upon. Although the line

484 

Zelle 2007, 197. Hayes 1991, 150. 491  Slane 1997, FW590-591. 492  Cf. supra, 2.2.2 (ESB), 2.2.5 (ITS). 489  490 

64

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:35 UTC All use subject to https://about.jstor.org/terms

The Late Hellenistic and Early Roman Periods: Discussion and Chronological Phasing

Figure 52. Map showing the distribution, in absolute numbers, of ITS, ESA, ESB, ESC and ESD during phase 2(a) (c. 30 BC-25/30) (© Philip Bes/ICRATES Project).

chart (Figure 9) indeed reflects this, a side-effect of the method of quantification used should be noted. Because several common forms of the late Hellenistic period, such as 4A and 22, have a terminus of 10-20, there is considerable overlap with forms that are in vogue from the mid-Augustan period onwards, such as forms 28-30, 42 and 46. The result is a quantitative overlap that could affect this and other charts to some extent.

highly problematic, reflecting the more complex ‘waves of fashion’ of this period.494 Although the typo-chronological evidence indicates that small quantities of ITS were shipped to the east prior to 1, was this enough to trigger a fundamental change in the morphological repertoire of the ESA workshops? In fact, since this is based on typochronological dates, it is not clear when this ITS exactly arrives in the east. In consideration of the technical and morphological properties, a more likely explanation is that the inspiration for ITS and ESA derives from costly metal prototypes.495 As metal vessels (mainly gold and silver) were used by the elite of society, who are thought to have possessed the financial and infrastructural resources as well as the motivation to set up (a) tableware workshop(s), it were members of the elite who provided the main inspiration in the decision-making process as to what shapes were to be made. In reality, shapes in other material categories could also have provided inspiration. Nonetheless, in this model, tablewares in silver and gold (from Italy?), now largely disappeared archaeologically, circulated in the east and

Both groups of forms also draw attention to a socio-cultural phenomenon. The popular forms of phase 1b (3, 4A, 22) clearly belong to a Hellenistic tradition, and are in sharp contrast with the new forms that were manufactured from about c. 10 BC onwards (28-30, 42, 46). These have more pronounced angular features, which become even more manifest towards the end of phase 2 and during phase 3 (cf. infra, 4.3.3). These bear great resemblance to several distinct forms of ITS. This implies that two distinct ceramic traditions coexisted for several decades.493 However, as ITS began to arrive in the east more generally only during the late Augustan-Tiberian period, a direct, linear association of the morphological influence of ITS on ESA remains

494  Élaigne 2004, 133, 135, especially in the context of Alexandria. More generally, see Slane 1997, 273-274. For a late Hellenistic/Republicanearly Roman socio-economic context, see Wallace-Hadrill 2008, 356440. 495  Poblome and Zelle 2002, 278-279, 281 (fig. 1).

Obviously based on the current typo-chronologies; Bes and Poblome 2006, 158-159; Poblome and Zelle 2002; Wilson 2009, 239.

493 

65

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:35 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East made sufficient impact to fuel change in eastern (ceramic) tableware morphology. The fact that ESA forms bearing similarities to ITS may have become more popular only after the Augustan period (cf. infra, 4.3.3) might suggest that the process of change was relatively slow, and possibly reflects preferences or otherwise continued sentiments on the part of eastern populations regarding a material culture they were (too) well acquainted with.496

from Ephesos, Knossos and Perge; Ephesos and Knossos, however, reflect the changing tableware landscape, as these deposits contain significant quantities of ITS, ESB and ESC (and ESD at Knossos).500 To this can now be added further, recent, evidence from Ephesos: the proportions of terra sigillata from Bauphase 1 (c. 20-1 BC), relating to the city’s prytaneion, reflects the last phase of predominance of ESA.501 A fill at Caesarea Maritima of the late 1st century BC contains only (residual?) ESA,502 and two contemporary deposits at Knossos still contain ESA in some quantity,503 yet these already mirror the changing tableware scene. A similar situation is attested at Paphos.504 Deposits of slightly broader date are more diverse. At Paphos, ESA gives way to ESD in a supplement to well 4, still ESA dominates at Tel Anafa and Samaria-Sebaste.505 In the Aegean, ESA still forms a considerable share in two deposits dated to c. 10 BC-10.506 At Argos, the trend that began in the 1st century BC continued into this period.507 Whereas ESD clearly starts to make its mark at Paphos, a dump in Ephesos containing large quantities of burnt ESB characterises the changing tableware situation at that site.508

A more fundamental question entails the direction of influence. As it is very difficult to determine when such and such occurred archaeologically, arguments can favour both a west-east and an east-west direction. Moreover, this direction of influence could have changed at different stages of the production, and the stamps provide an important clue. All stamps on ESA are found on italicised forms, and stamps become more common after the Augustan period, at a time when stamps on ITS had existed already for about a generation or so. That both ITS and ESA vessels were stamped possibly favours a direct inspiration regarding the practice of stamping, whilst metal prototypes provided the morphological inspiration. The general circumstances are more intricate. ‘Roman’ influence intensified in the east from the late Hellenistic period onwards, and Italic material culture could follow in its wake. This process continued during the early Empire, when Roman hegemony encompassed the entire Mediterranean. The influence of Hellenistic lifestyles and material culture in Italy on the other hand, from the late 3rd century BC onward, likely included Hellenistic ceramic tablewares.497

ESD Whilst during phase 1b ESD is mainly identified at Paphos, Palaipaphos and Panayia Ematousa, its distribution expands considerably during phase 2, and now also occurs at other sites along Cyprus’ south coast. Outside Cyprus its distribution merely includes the (coastal) areas closest to Cyprus, particularly the (southern) Levant as far south as Oboda and Petra. Further west, small numbers can be identified at Berenice, Corinth, Tenos, Gortyna and especially Knossos. This geographical expansion can comfortably be compared with Figure 9, which shows a considerable increase towards the late 1st century BC, and is further substantiated through archaeological evidence.509

The distribution of ESA seems to continue along the lines established during phase 1b. Outside the Levant, which remains its core area, substantial quantities are found in Cyprus and parts of the Aegean, and at sites such as Anemorion, Alexandria and Berenice. Illustrative is the fact that rare specimens now reached such distant sites as Carthage and Meroë, and a few fragments were recently identified at Arikamedu in southeast India.498 At least one change seems to have taken place, however. Some quantities were identified at sites in eastern Anatolia (Asia Minor-East Central) and along the Euphrates during phase 1; except for Dura Europos, Lidar Höyük and a single fragment as far inland at Sultantepe this is now no longer attested. On Cyprus, ESA appears to make way for ESD (cf. infra), though its manufacture probably continued on an exceptional scale. These changes, however, will become more visible in phase 3.

ITS The distribution of ITS in the east, where it played but a minor role quantitatively, signifies a profound change on several levels.510 First, forms and stamps with a terminus ante quem around 1 are attested mainly at large(r) urban centres that had easy and often direct access to central and Ephesos: Meriç 2002, Füllung A; Knossos: Sackett 1992, group A2; Perge: Atık 1995, sondage II, schicht 7. 501  Cf. supra, n. 339. 502  Berlin 1992, fill A. 503  Sackett 1992, group A1; Hayes 1971, the lower fills in wells MW/58-9 and RR/K/60. 504  Hayes 1991, well 10. 505  Paphos: Hayes 1991, well 4 supplement; Samaria-Sebaste: Kenyon 1957, Roman Ia phase; Tel Anafa: Slane 1997, groups ROM-1A-B. 506  Slane 1986, locus 543; Williams II and Zervos 1982, manhole 1981-3. 507  Abadie-Reynal 2007, 88. 508  Zabehlicky-Scheffenegger 1996. 509  See e.g. Paphos: Hayes 1991, layers 7-10 in room Λ, the earlier part of well 3, the supplement to well 4, and well 18; Knossos: Sackett 1992, groups A1-2; Hayes 1971, the lower part of well MW/58-9. Deposits dated to the early 1st century could characterise the greatest extent in the distribution of ESD, as it is now found at Tel Anafa (Slane 1997, ROM1B) and Ephesos (Meriç 2002, Füllung B1). 510  For a synthesis, see Poblome 2004b. 500 

Although it is unnecessary to fully repeat the evidence, several deposits highlight the geographical extent that ESA now reaches. At Paphos, a deposit dated to c. 40-10 BC contains ESA and ESD in roughly similar quantities, yet only ESA was found in well 6 of similar date.499 ESA maintains a considerable share in three Augustan deposits Van der Enden et al. in press for a case study. See e.g. Malfitana et al. 2005. 498  Fulford and Timby 1994, phase 3.3; Freed 1998, second street level; Dunham 1957; Slane 1996b. 499  Hayes 1991, layers 7-10 in room Λ, and well 6. 496  497 

66

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:35 UTC All use subject to https://about.jstor.org/terms

The Late Hellenistic and Early Roman Periods: Discussion and Chronological Phasing

Figure 53. The distribution, in absolute numbers, of ITS form and stamp entries that predate 1 (© Philip Bes/ICRATES Project).

eastern Mediterranean waters; there is, again, no certainty when these arrived in the east, though it is not unthinkable that this distribution pattern is a pale reflection of major shipping routes (Figure 53): one would expect a new product to spread initially along the major routes that connect the chief harbour-cities.511 The distribution of ITS prior to c. 1 could reflect this skeletal layout: ITS made a considerable impact in the east during phase 2 in general (Figure 52), and is now (most) common at those sites where it is already attested prior to c. 1. Simultaneously, ITS enters the sphere of secondary towns and settlements more regularly, illustrated by the few fragments identified at Tel Anafa.512 Figure 9 demonstrates that ITS is most common during the later Augustan and Tiberian periods, which implies that ITS achieves its widest geographical and quantitative distribution later in phase 2, and probably into phase 3.

Secondly, the influence ITS had on vessel morphology. In addition to ESB (cf. infra), which is a somewhat specific case, vessel morphology of ITS manifests itself most prominently in ESA (cf. supra): ‘western’ influences are certainly also seen in ESC, yet are particularly slight in ESD forms, whose popularity was also restricted.513 The Augustan period defines a period when a Hellenistic and a ‘Roman’ tradition coexisted, and the new ESA forms, such as 28-30 and 42, are culturally and technologically different, yet the classic late Hellenistic ESA forms 4A and 22 continue into the late Augustan period (cf. supra). ITS is attested in a number of deposits that are dated within this phase. To some extent these chart the developing distribution, and supplement Figure 53 that shows forms and stamps that predate 1.514 Though its quantitative Italian morphology is only thinly observed in ESD: Lund 2002, 197201, 206. Whereas western shapes inspired the manufacturers of ESC (although the direction of influence is still debated), interestingly these forms (e.g. N39a-d and T31a-e) mainly have a regional distribution. That lettered stamps on ESC and especially ESD are rare, interesting in itself, additionally also speaks in favour of a west-east influence. On the other hand, vessel morphology and the custom of lettered stamps need not have belonged to the same set of traits. 514  The Augustan group A2 at Knossos (Sackett 1992); a similarly-dated deposit (38) at Berenice (Kenrick 1985); from a late 1st-century BC 513 

511  At Alexandria, the first imports of ITS are tentatively placed in the middle of Augustus’ reign: Élaigne 2004, 139. Typologically early ITS, probably Augustan, is found among recently published finds from Ashkelon: Johnson 2008a, 25-29, nos 79, 87-88. See Ettlinger 1987 for instructive observations concerning early Arretine ITS in the west; esp. 10, seeing ITS in the east (and Africa) as part of return cargoes. 512  Slane 1997, FW595-596.

67

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:35 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East impact developed relatively slowly, ITS spread widely in the east, and (directly or indirectly) laid the basis for new ceramic traditions. Yet below the surface, the distribution of ITS entails some specific elements, and Corinth appears most exceptional. Here, early ITS is not uncommon, and its import continued well into the 2nd century. By the second quarter of the 1st century it comprises c. 63% of the imported terra sigillata, at a time ‘when it first appeared at other Greek sites’:515 pre-25 ITS does occur in Greece, but indeed is rather rare. For example, at Argos Augustan forms and stamps do occur, yet ITS only becomes significant by the reign of Tiberius.516 Early ITS also occurs at Knossos, and accounts for c. 29% in Tiberian and c. 43% during Claudian times. Early ITS appears to have arrived at Corinth in larger numbers than other Italian products, and set against other central-Greek sites ‘clearly demonstrates that the distribution of Italian sigillata in the East depends on factors other than geographic proximity to Italy. Corinth’s Italian sigillata begins half a century earlier than the bulk of Italian sigillata imports to Achaia and the eastern Mediterranean as a whole’.517 These matters will be investigated in more detail in chapter 4. Slane, however, is convinced that a range of factors influenced the distribution of ITS in the east, or economic traffic more generally.

ESC During the Augustan-Tiberian period the distribution of ESC increases in both quantity and geographical scope, possibly related to the start of production much closer to the coast at Çandarlı (ancient Pitane) (Figure 52).521 Whereas ESC is only sporadically identified outside the area of Pergamon during phase 1(b), by the early Roman period it continued to dominate at Assos, Troia, Smintheion and Pergamon proper. Generally it makes a more significant impact in the Aegean, including Crete, yet its absence in the north(west) Aegean probably reflects the published record rather than anything else. Even though quantities remain persistently small, ESC is found at more distant sites now, including Amorion, Paphos, Tarsos, Gindaros and again Alexandria, Tel Anafa and Pednelissos, and indicates that ESC also experienced growth. Archaeologically, ESC is identified in Füllungen A (Augustan) and B1 (early 1st century) of a well at Ephesos, at Troia in a late Augustan well fill, at Athens in an early 1st-century layer which is part of a cistern fill, at Sparta in an Augustan deposit, and at Knossos in a deposit that tentatively dates to the Tiberian period.522 Beyond, fragments are known from only a few narrowly-dated deposits at Alexandria, Paphos and Tel Anafa.523

ESB

3.3.3 Phase 3 (c. 30-60/70)

Based on morphological similarities, potters’ stamps and the circumstances of the time, Italian potters or owners of workshops had a direct or indirect hand in the inception of ESB.518 Although its heyday came later, the initial quantitative impact of ESB seems to be confined to Ephesos and surrounding sites (even if evidence is slim), and elsewhere in the (southern) Aegean: Crete and central Greece; at Argos, however, the earliest forms remain ‘anecdotique’, yet nonetheless gradually increased in quantity.519 Outside this area (mostly) single specimens have turned up as far afield as Alexandria, Myos Hormos, Petra, Jerusalem, Tel Anafa, Antioch and Tarsos.520

The mid-1st century witnesses the further contraction of ESA, and ITS also generally was past its main export phase to the east. ESB is now found more commonly outside its region of origin. ESD basically continues on the same footing as before, whilst ESC experiences a quantitative decrease. The date ranges of some common forms prohibit making a clear distinction between phases 2 and 3, although the upper limit of 60-70 suits the evidence best, and allows a better definition of the effects of the distribution of ITS and ESB. All five terrae sigillatae witness quantitative and geographical growth during phase 2, and phase 3 further documents this. ESA ESA still figures as a major class of terra sigillata throughout the eastern Mediterranean, and acquired an extensive distribution pattern during phases 1 and 2. After the Tiberian period, however, certain changes are noted. Figure 54 clearly shows that ESA is considerably less common than previously, and now also attested at lesser sites. Although ESA remains a widely-used terra

deposit at Tell ‘Arqa come two ITS vessels (Thalmann 1978). 515  Slane 2004, 32. 516  Abadie-Reynal 2004, 59, 63, 2007, 63-64. 517  Slane 2004, 40. 518  Zabehlicky-Scheffenegger 1995, 1996. 519  Abadie-Reynal 2007, 112. 520  It is found in the Augustan Füllung A at Ephesos (Meriç 2002), with ESA, ESC and ITS, yet is the most common terra sigillata in the early 1st-century Füllung B1. ESB forms the overwhelming majority in a large mid- to late Augustan pottery dump from the Tetragonos Agora (Zabehlicky-Scheffenegger 1996), which contains only some (residual?) ESA and ITS. Northwards, a few fragments are attested in a well fill at Troia (Hayes 1995a). At Athens a fragment is identified in the early 1st-century group F (Robinson 1959); see now also Hayes 2008a, for instance 140-141, nos 185, 187. Early forms also turn up at Corinth, in shop XXI (Williams II 1980), and in the Tiberian floor deposit where it is a minor ware (Wright 1980). A few fragments are present in Sackett’s group A2 from Knossos, but it is slightly more common in group B1 of about 10-40, when it comprises c. 16% of the imported tablewares. A distant find is from Tel Anafa, in ROM-1A-1B (c. 25 BC-25).

Loeschke 1912. Ephesos: Meriç 2002; Troia: Hayes 1995a, main fill of well Bv; Athens: Robinson 1959, group G, layer II-a; Sparta: Bailey 1993, RSC 1N 4168; Knossos: Sackett 1992, group B2. 523  Alexandria: Élaigne 2000, a fragment in deposit 2; Tel Anafa: Slane 1997, FW502, in phase ROM-1A, and unidentified ESC-fragments in the early 1st century ROM-1B; Paphos: Hayes 1991, in the early 1st-century fill of well 18. 521  522 

68

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:35 UTC All use subject to https://about.jstor.org/terms

The Late Hellenistic and Early Roman Periods: Discussion and Chronological Phasing

Figure 54. Map showing the distribution, in absolute numbers, of ITS, ESA, ESB, ESC and ESD during phase 3 (c. 30-60/70) (© Philip Bes/ICRATES Project).

sigillata during phase 3, the contraction of its geographical distribution is corroborated by Figure 9,524 as well as the line charts for e.g. Crete, the Aegean-Southwest and -East, and Africa-Cyrenaica and -Egypt which also receive smaller quantities. The reduction in quantity marginally surfaces also in the Levant and Cyprus. ESA nonetheless remains very common in the Levant-Coastal North, and is now found at new sites as Athis along the Euphrates. The phase map shows that ESA is now more common at Tarsos (Figure 52); in fact, in Asia Minor-Cilicia, ESA actually reached its peak during this phase (Figure 30, mostly reflecting the Tarsian data).525 Conversely, ESA decreases at Anemorion (cf. infra, ESD). Although these developments could result from publication-specific factors in some, or the (temporary) end of occupation at a(n excavated) site in other cases, the pattern is too widespread to be argued

away as a coincidence, and quite probably signifies major changes in the production and distribution of ESA. Furthermore, this pattern can to some extent be substantiated through the archaeological evidence. Slightly problematic is the fact that a deposit that is roughly contemporary with phase 3 could contain forms that because of their date range fall within phase 2. A late Tiberian deposit from Ephesos can serve as an example: it contains ESA forms 3, 7, 22 and 42, all attributable to either phase 1 or 2.526 This material could be residual, the arrival at Ephesos could have been late in their date range, or their life-span extended well beyond it. In a deposit of c. 25-40 from Alexandria the forms better converge with the proposed deposit date.527 ESA accounts for a little under one-fourth of the catalogue but comprises c. 33% of the finds. In two earlier deposits from Alexandria, dated to the second half of the 1st century BC and the late Augustan period, ESA accounts for c. 98% and 52% respectively.528 Füllungen B2 and B3 from the deep well on the Tetragonos Agora at Ephesos still contain some

524  Whereas this line chart confirms that lesser quantities were available than during phase 2, it also presents a brief resurgence in quantity roughly during the second half of the 1st century. This could result from the choice of the upper limit of phase 3, now around the time when a small series of popular plates of ESA appear (36, 37A-B), whose date range spans 60-100. 525  Asia Minor-Cilicia and Cyprus (Figure 49) show an increase during the second half of the 1st century. Whereas this can be explained by the upper limit of phase 3, geographical proximity could offer a better explanation.

Gassner 1997, stratum 7. Élaigne 2000, deposit 3. 528  Élaigne 2000, deposits 1 and 2. One aim was to capture the changing proportions of the different tablewares. 526  527 

69

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:35 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East ESA, most of which apparently belonged typologically to the Augustan-Tiberian period.529 ESA was notably present in the late Tiberian stratum 7 from the Südtor excavations at Ephesos, yet was only marginally present in stratum 8, apparently as residuals.530 This general picture is also seen at Berenice,531 where ESA seems only marginally represented in deposits of the mid-1st century.532 ESA still appears in groups C1 (late Tiberian-Claudian), C2 and N1-3 (ClaudianNeronian)533 at the Unexplored Mansion at Knossos, though is most common in the Augustan (c. 68%) and Tiberian (c. 45%) periods; during the Claudian period it comprises only c. 24% of the imported tablewares (including likely residual material).534 Only a few and small deposits of the mid-1st century are available for Corinth, from the sanctuary of Demeter and Kore, among which ESA was not recognised,535 and the same applies for deposits of the third quarter of the 1st century.536 ESA is uncommon in the well-known South Stoa deposits,537 and at Athens ESA is most common during the Augustan period.538 At Argos, recently published evidence points out that ESA began to dwindle roughly from the reign of Tiberius onward.539 Further east, deposits 3B-F from the Saranda Kolones site at Paphos contain little ESA, and corroborate Hayes’ observation that ESA is dominant during the 1st century BC, being supplanted by (mainly) ESD during and after the Augustan period (also see Figure 49).540 Well 4 from the House of Dionysos, dated to c. 60-80, contained only a few fragments of ESA.541 Information for Levantine sites seems scarce, yet a cistern fill from Beirut points out that at least here ESA remains the major class of terra sigillata.542

after the mid-1st century and now is the commonest sigillata on the western part of the island, gaining ground at the cost of ESA, which remains common at Panayia Ematousa on east Cyprus (cf. supra; Figure 49). Except for Beirut, no specimens are identified for the Levant-Coastal North.543 Except for the more distant regions where it is attested, tracing the archaeological presence of ESD sheds hardly more light on its distribution. At Paphos, it prevails over the other classes in deposits 3B-F that contain only small quantities of tablewares (cf. supra);544 it is the most common among a wide range of tablewares in well 4 from the House of Dionysos (cf. supra, ESA).545 ESD apparently is not the main class of terra sigillata in the excavations in the sanctuary of Apollo Hylates near Kourion.546 ITS ITS fares essentially the same geographically as during phase 2 (Figure 54), though one change stands out. The overall chart in Figure 9 shows that the peak in the distribution of ITS coincides with the late Augustan-Tiberian period, which reflects the communis opinio, after which quantities slowly but surely dwindle. However, several of the common forms assigned to phase 3 (e.g. Consp20.4, 23, 34) have considerably long date ranges that extend beyond the Tiberian period.547 The geographical as well as the archaeological background nonetheless illustrates this change. Although Figure 54 shows that ITS peaks in virtually all regions during phase 2, it witnesses certain changes in the Levant and surrounding regions during phase 3. It is much less common in the Levant-Coastal North, whereas in the southern Levant and Asia Minor-Cilicia it seems to maintain the level of phase 2. Interestingly, Cyprus presents an increase (Figure 48): during phase 2 small quantities are recognised in the (south)west, but now ITS occurs at virtually each site from where terra sigillata is published, most notably Paphos. This rather divergent pattern continues into phase 4, and need not be isolated (cf. infra, chapter 4):548 the region-based charts illustrate the uninterrupted exceptional position of ITS in the AegeanSouthwest, on Crete, and at Berenice.

ESD No extraordinary changes are observed in the quantitative distribution of ESD, which, despite a brief and minor drop around the mid-1st century, is also what the overall chart (Figure 9) indicates. The region-based evidence shows that nearly all regions now receive ESD, though only in three does it play a role of importance: Cyprus, Asia MinorCilicia (particularly Tracheia) and the Levant-Coastal South. This indicates that ESD continues to form part of inter-regional exchange, mostly between Cyprus and its two nearest coastal zones. ESD continues to appear at sites in the Aegean and Africa-Cyrenaica, usually single or a few fragments at most. On Cyprus it is slightly more common

Quantified evidence from Knossos points out that ITS flourishes during the Claudian and Neronian periods (roughly during phase 3) when it accounts for c. 43% and 38% respectively of the imported terra sigillata.549 A study of the pottery excavated at Olympia up to the 1980s

Meriç 2002. Gassner 1997. Kenrick 1985, deposits 44, 48. 532  Kenrick 1985, deposits 55, 73. 533  Sackett 1992. 534  Sackett 1992. Several well fills dated to the very mid-1st century contain little ESA (Hayes 1971): the late fill of well KW/51/13, and the abandonment features of area RT/60. 535  Slane 1990, loci 97, 103, 252. 536  Slane 1986, locus 117; Williams II and Fisher 1976, the fill of the south wall of the ‘Long Rectangular Building’. 537  Hayes 1973a, although the date for these deposits does not receive general agreement. 538  Hayes 1996a, 10. 539  Abadie-Reynal 2007, 89-90. 540  Hayes 2003. 541  Hayes 1991. 542  Reynolds 2004a, cistern fill 006.12300/12237, dated to c. 65-75. 529  530  531 

Reynolds 2004a. Hayes 2003, from the Saranda Kolones site. 545  Hayes 1991, well 4, datable to c. 60-80? 546  Neuru and Soren 1987. Note that different eastern sigillatae seem to have been confused. 547  Consp20.4 (c. 25/30-100), Consp23 (c. 25-75), Consp34 (c. 25/30100). 548  Malfitana 2004, with further references. 549  Sackett 1992, 153, 163, fig. 2. It is important to see whether these and other deposits contain residual material, and how much, as at Argos, see Abadie-Reynal 2004, 64-65; for ITS at Knossos also see Eiring 2004. 543  544 

70

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:35 UTC All use subject to https://about.jstor.org/terms

The Late Hellenistic and Early Roman Periods: Discussion and Chronological Phasing indicates that ITS peaks only after 50.550 Also at Argos ITS becomes most important around the mid-1st century, based on the identification of forms, stamps as well as archaeological deposits.551 This forms a considerable contrast with Ephesos, where the majority arrived during the late Augustan-(early) Tiberian period,552 a pattern also observed at Jerusalem and Beirut.553 Most of the ITS from Epiphaneia is dated to the first half of the 1st century.554 This suggests that some regions or sites are already beyond their peak of ITS by the mid-1st century, whereas others see ITS mostly coming in during phase 3, which continues into phase 4. These divergent patterns in the import of ITS are further discussed below (4.3.4).

pattern continues in Füllung B3 of the third quarter of the 1st century.559 ESB is also the most common terra sigillata in a Neronian deposit excavated at the Südtor in Ephesos.560 Although quantities from these three deposits are small, they seem to document a development that starts in phase 2. ESB is the commonest terra sigillata in the late Tiberian-Claudian group C2 from the Unexplored Mansion at Knossos.561 Whereas ESB comprises c. 16% in the Tiberian period, it grows to about 27% in the Claudian period, and to c. 42% of the imported terra sigillata in Neronian times.562 At Athens, ESB becomes the major terra sigillata after the third quarter of the 1st century. At Corinth, however, ITS prevails yet ESB comprises roughly 11% both in the early 1st century and in the later Julio-Claudian and early Flavian periods, which conforms to the chart for the Aegean-Southwest. Recently published evidence from Argos points out that ESB was rather common by the third quarter of the 1st century, yet as at Corinth it was still overshadowed by ITS.563 If ESB remained of minor importance at Corinth, with little to no increase during phase 4, this strongly suggests that sites in the Aegean-Southwest other than Corinth (and Argos?) conform to the normal pattern for ESB; ITS then sustained its prolific role at Corinth. This latter point is indeed borne out by pottery studies at Corinth, where proportions of ITS remain unparalleled during the later 1st century and the first half of the 2nd.564 Finally, two deposits at Alexandria indicate a development of ESB outside the Aegean. ESB accounts for about 10% of the imported terra sigillata in a late Augustan deposit, and increases to c. 31% in a deposit of 25-40.565 Though this shows less clearly on the map for phase 3, it is of interest (cf. infra, chapter 4).

ESB ESB is not uncommon during phases 1 and 2, when it was mostly attested at Ephesos, yet with a wide but thin distribution in the east. During phase 3 ESB displays substantial quantitative and geographical growth. Figure 9 illustrates that quantities of ESB roughly doubled, and the greater role of ESB is also documented in Figure 54, especially in some Aegean regions, on Crete, in Asia MinorWest Central and at Berenice. Elsewhere, ESB maintains its thin distribution. The only exception is Paphos and several other sites along the south coast of Cyprus, where ESB is now recognised. Specimens turn up at distant sites in Ethiopia,555 the Arabian peninsula556 the Red Sea ports of Berenike and Myos Hormos,557 and India.558 The distribution of ESB now clearly goes well beyond the main urban (coastal) centres, although ESB is occasionally found already in non-urban contexts during phase 2. The regional evidence supports this pattern of quantitative and geographical expansion, and the developments are best observed in the Aegean-East, -Southwest and Crete, to a lesser extent in the Aegean-North, Africa-Egypt and Cyprus. The Aegean-East, home to the ESB workshops, based on the available typo-chronological framework presents a fundamentally different pattern. The contrast is particularly prominent during phase 4: in the Aegean-East, the highest quantity of ESB is noted towards the mid-1st century, after which gradual decrease set in, at a time when ESB continued to rise in the east more generally (Figure 56).

ESC The character of the distribution of ESC changes little, particularly in comparison to phase 2. ESC continues to be the major terra sigillata at sites in parts of the Aegean-North and -East, though at Troia quantities drop slightly. Other classes of imported terrae sigillatae were only sporadically attested for phase 2 and this did not really change during phase 3. The bottom line is that ESC remained the most commonly used terra sigillata. Figures 51-52 present the distribution of ESC during phases 1b and 2 and point out, with the exception of the area of Pergamon and Ephesos, that numbers usually do not exceed a handful of finds. ESC is identified at lesser sites during phase 3 (Figure 54, 19 against 25 in phase 2; appendix 4a), and the quantitative distribution also shrinks somewhat (Figure 9): this is the case in the other Aegean regions, where ESC is now attested at fewer sites. Further east, ESC is identified only at Jerusalem and Paphos,

These changes can also be tracked archaeologically, although the evidence pertains to only a handful of sites. At Ephesos, ESB is the main terra sigillata in Füllung B2 (c. 25-50) in the well on the Tetragonos Agora; this Martin 1997b, 2004. Abadie-Reynal 2004, 2005a, 41, 2007, 64. 552  Martin 2003, concerning differences between Ephesos, Corinth and Olympia; also Meriç 2002, 44-45. 553  Jerusalem: Hayes 1985b, 184; Beirut: Reynolds 2004a, 125. 554  Lund 1995, 137, although the quantity is small. 555  Axum: De Contenson 1963; Karanòg: Woolley and Randall-MacIver 1910; Kush: Dunham 1957. 556  Timna’: Comfort 1958; Khor Rori: Comfort 1960. 557  Berenike: Hayes 1996b; Myos Hormos: Whitcomb 1982. 558  Arikamedu: Slane 1991a, 1996b. 550  551 

Meriç 2002. Gassner 1997, stratum 8. 561  Sackett 1992. 562  Sackett 1992, yet with an increasing chance of residual material. 563  Abadie-Reynal 2007, 112. 564  Slane 1989, 2000, 2004. 565  Élaigne 2000, deposits 2-3. 559  560 

71

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:35 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East whereas it was noted at Alexandria, Anemorion, Paphos, Gindaros, Kourion and Tel Anafa during phase 2.

ESA Although the scarce distribution of ESA west of the line Lycia-Alexandria during phase 3 already signifies an important change, the later part of phase 4 marks the final stage of the interregional distribution of ESA.

Archaeologically, ESC is attested at Ephesos, Knossos, Corinth, Athens and Paphos in deposits that date to the mid1st century;566 ESC is absent in a contemporary group from Argos, and as a matter of fact, ESC is rare in 1st-century Argos.567 At sites in the core zone of distribution, i.e. Assos, Smintheion and Troia, the presence of ESC during phase 3 cannot be supported by archaeological deposits except for Pergamon. As a result, its development is based on typochronological identification but the Troad and the area around Pergamon continue to form the main market for ESC, as well as that beyond this zone no profound change occurred: ESC was shipped across considerable distances, even if only occasionally. Quantitatively and geographically, however, ESC increases again during phase 4.

Small quantities of ESA in the Aegean and AfricaCyrenaica show that its widespread distribution survived into phase 4 (Figure 55). Further east, ESA persists in the Levant as a major class of terra sigillata, and also on (notably eastern) Cyprus and in Cilicia Pedias it remains common. These regions, particularly the Levant, continue to form the core zone of distribution. However, the length of this phase masks the fact that a considerable number of fragments found in the Aegean belongs to forms 36 and 37A-B (dated c. 60-100). This strongly suggests that ESA appears even more rarely west of the line LyciaAlexandria after 100, and that its distribution pattern continued to recede eastwards. This finds further support from ESA forms that post-date 150, thus including Hayes’ tarda-series:570 these are virtually not distributed beyond the core zone.

3.3.4 Phase 4 (c. 70/75-200) The late 1st and 2nd century witness the further dwindling of both ESA and ITS. The latter, however, remained fairly common in pockets in Cyrenaica, southern and central Greece and western Cyprus in the decades around 100. Both ESB and ESC see quantities increase around the Aegean, the former mostly during the late 1st and early 2nd century, the latter continuing into the 3rd century. ESD fared well still mostly on (western) Cyprus, even if (coastal) Egypt and the southern Levant continued to be supplied. In the course of the 2nd century small quantities of ARSW began to arrive at sites throughout the east.

ESD ESD maintains roughly the same level as during phase 3, when it experienced a period of growth, and the basic outlines of the circulation of ESD also remain unchanged. Yet, higher quantities are noted for Cyprus (including its eastern part), Anemorion in Cilicia, and Alexandria and Marina el-Alamein in Africa-Egypt. Examples also still appear quite regularly at sites in the southern Levant. Elsewhere virtually nothing changes, and where ESD is recognised beyond its core zone, numbers usually do not exceed one or a few specimens at most; it strongly resembles phase 3, though ESD is marginally more common at Berenice and in the Aegean-Southwest. The overall chart (Figure 9) shows that the most intensive period in the quantitative distribution of ESD falls in the first half of the 2nd century; during the second half of that century typo-chronologically it is virtually absent.

This comparatively long phase is characterised by several profound changes. Tunisian workshops began producing ARSW in the (second half of the) 1st century. The second half of the 2nd century marks a halt for all but ESC and ASRW, given the existing typo-chronologies,568 or at least their interregional distribution, which suggests a fundamental change in the productive and/or trading frameworks. The production and distribution of some of these classes continue, however, yet presumably on a muchreduced scale, now catering largely for a regional demand. As such, the later 2nd and 3rd centuries are rather bleak. With regard to tableware distribution, this phase and phase 5 (cf. infra) can be seen as a period in its own right.569

ITS Distinguishing ITS forms between phases 3 and 4 is less obvious, since several common forms that are attributed to phase 3 continue into phase 4 (Consp 20.4, 34). Figures 54-55 nevertheless illustrate the marked changes in the distribution of ITS during the later 1st and first half of the 2nd century. Understanding of the later stages in the distribution is quite recent and mainly focuses on Paphos, Argos, Olympia and notably Corinth.571 Figure 55 shows the final stage in the distribution of ITS, and indeed greatly contrasts with phases 2 and 3. ITS is now identified at far fewer sites: considerable quantities are

566  Ephesos: Meriç 2002, Füllungen B2 (c. 25-50), B3 (50-75); Gassner 1997, stratum 8 (Neronian); Knossos: Sackett 1992, groups C2 (2560), N2 (50-75); Corinth: Slane 1990, locus 103 (25-75); Hayes 1973a, groups A, AB, AT, BW, CD, CT (Neronian, not generally accepted); Athens: Robinson 1959, group M, layer I (25-75); Paphos: Hayes 2003, deposit 3F (25-75), possibly Hayes 1991, well 4 (c. 60-80). 567  Abadie-Reynal 2007, 128-129. 568  Again note that the production of ESA and ESB continues into the 3rd century. 569  Lund 1999, who includes the first half of the 4th century. See also Poblome 2006a, 191-192, with bibliography.

570  Hayes 1985a, 42; these are generally dated to the later 2nd century; see now also Reynolds 2010, 90. 571  Paphos: Malfitana 2004; Argos: Abadie-Reynal 2004, 2007; Olympia: Martin 2004, 2006; Corinth: Slane 2004.

72

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:35 UTC All use subject to https://about.jstor.org/terms

The Late Hellenistic and Early Roman Periods: Discussion and Chronological Phasing

Figure 55. Map showing the distribution, in absolute numbers, of ITS, ESA, ESB, ESC, ESD and ARSW during phase 4 (c. 70/75-200) (© Philip Bes/ICRATES Project).

still found in a few regional pockets (Aegean-Southwest, Crete, Africa-Cyrenaica),572 as well as a handful of often single recordings (e.g. Stobi, Tarsos, Antioch, Scythopolis (?), Petra, Berenike). Paphos is the only site east of the line Aegean-Crete-Cyrenaica where ITS of this period is identified in some quantity. At Knidos, ITS is scarcely attested after the mid-1st century.573

century.576 At Kition, a fragment is found in a deposit of the late 1st century;577 the fragment is unidentified, but tentatively suggests that late ITS was not only found at Paphos. At Berenice, a fair quantity of ITS is found in several deposits of similar date.578 Further information about late 1st-, early 2nd-century deposits comes from Tarsos and Stobi.579 At Corinth, several fragments are found in a deposit of c. 80-110, though the typo-chronology of some suggests these could be residual.580 Also from Corinth are two sherds, identified in a late 1st centuryHadrianic deposit.581 Deposits of the early 2nd century that contain ITS are published for Berenice, Ephesos (residual, though), Troia and Paphos, where it occurs in Hadrianic deposits.582 Slightly broader dated deposits (c.

The distribution of ITS during this period is more based on stamps than on forms, though the long date ranges of several of these stamps makes a very precise definition of the distribution of ITS somewhat problematic. As the archaeological information can only partly resolve this, more revealing are the sites and regions where late ITS is identified (cf. infra, chapter 4). Forms of the later 1st and first half of the 2nd century are indeed attested in certain deposits. At Athens, in a late 1st-century layer of a cistern fill, two ITS fragments are found.574 ITS is not uncommon in groups F1-F2 at Knossos (c. 80-100).575 At Paphos, a few sherds turn up in deposits of the late 1st and early 2nd

Hayes 1991, the later part of well 3, space ΔΙ, cistern 5. Marquié 2004, 255. 578  Kenrick 1985, deposits 63, 69-70, contain examples of Consp20.4, which is attributed to phase 3 because of its date range; these and deposits at other sites strongly suggest that this form is still distributed by this time. 579  Stobi, possible residual fragments in lots 18, 385 (Anderson-Stojanović 1992); Tarsos: Jones 1950, trench 6 from where a few fragments (including residual, or wrongly dated?) are catalogued. At Carthage ITS is rather common in similarly-dated deposits, perhaps partly residual (Fulford and Timby 1994, phases 4.6, 4.7c-d, 4.8). 580  Williams II and Zervos 1989, deposit 1988-50. 581  Slane 1990, deposit 73-117. 582  Berenice: Kenrick 1985, deposits 72, 78; Ephesos: Meriç 2002, Füllung D2; Troia: Hayes 1995a, well C29; Paphos: Hayes 1991, deposits ΘΣ, ZY. Also see ITS stamps 1, 23, 25, 32. 576  577 

572  Though see Reynolds et al. 2008, 71-72, for Butrint, considering that the export of ITS to Cyprus also came to a (temporary) halt after Nero; the collected evidence here does not immediately corroborate this. 573  Kögler 2010, 370. 574  Robinson 1959, group G, layer II-b, including a form here attributed to phase 3, Consp20.4. 575  Sackett 1992, but includes residual fragments.

73

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:35 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East 100-150) that contain ITS come from Alexandria (possibly residual), Berenice, Caesarea Maritima, Paphos, and Stobi.583 Finally, a deposit dated to c. 125-150 is known from Knossos,584 and deposits of mid-2nd century date from Knossos and Stobi.585 To this can be added recently published information from Argos and Corinth: around 125-150 ITS regains an important position at Corinth, and to a lesser degree at Argos,586 where ITS does not occur in a group datable to 170-180.587 Despite their uneven quality, these deposits suggest that the distribution of ITS continues, yet now mostly concerns a restricted number of regions and sites in the general Aegean area.

a well at Ephesos, and groups F1-F2 at Knossos (which, if not (largely) residual, seemingly still received sizeable quantities of ESA).591 Interestingly, two-thirds of the catalogued pieces of Füllung D1 are ESB, the remainder ESC, which shows the interplay between ESB and ESC at Ephesos (and other sites in the Aegean-East) from the late 1st century onwards. ESB appears to be the major terra sigillata in layer III of a cistern at Athens,592 and figures prominently with ESC in groups T1-T3 (c. 90-120) at the Unexplored Mansion in Knossos; considerable quantities are found in groups of the late 2nd and early 3rd century.593 Given the new evidence on the continuation of ESB into the 3rd century we cannot simply ignore this and consider this residual. ESB is also common in a more broadly dated deposit from Athens (c. 75-150).594 At Argos, ESB II accounts for c. 64% of the ‘céramique fine’ in a Flavian group, a trend which continued into the 2nd century, to which period c. 56% of the ESB finds can be attributed. The import of ESB to Argos is thought to have halted around the mid-2nd century.595

ESB Phase 4 is also the final period for ESB, although recent evidence shows that ESB continues on a more regional level.588 Phase 3 is characterised as a period when ESB begins to dwindle in the Aegean-East after its mid-1st century peak, whereas it increased in most other regions.589 Even if this interesting pattern becomes clearer during phase 4 (Figure 56), recently published evidence from Knidos, however, points out that ESB was the dominant imported terra sigillata during the period c. 50-150. Clearly this pattern, if it is one, requires further study.

ESB is either absent (Kition, Resafa)596 or scarce in deposits of late 1st, (early) 2nd century date at Berenice, Corinth, Stobi, Paphos and Tarsos,597 sites mainly supplied with other classes of terra sigillata. Deposits with a terminus post quem of 100 or later, and variable termini ante quem not later than 150 in which ESB is hardly or not represented are known from Berenice, Corinth, Paphos, Kourion, Beirut, Alexandria and Caesarea Maritima.598 However, ESB was not uncommon in deposits of similar date from Isthmia, Ephesos, Knossos, Troia and Stobi,599 which shows that ESB remains a (fairly) common terra sigillata in several Aegean regions until the mid-2nd century; beyond, it appears to have travelled only sporadically. ESB is notably present at Ephesos, Athens and Knossos, and these could in fact represent some of ESB’s cornerstones during this phase. Interestingly, ESC is found at Knossos only from the Trajanic period onwards and increases to c. 54% of the imported terra sigillata by the late 2nd century; it also increases at Ephesos from phase 3 to 4 (Figures 54-55).

Figure 55 shows that ESB remained common at Ephesos (which one would expect), and grew in popularity at some other sites in the Aegean-East and -North, notably Assos — caused by forms that fall between c. 70-150 — even if the downward trend in the Aegean-East persists during phase 4. Elsewhere, some of the regional trends show that ESB actually increased during the second half/late 1st and early 2nd century: evidence for the Aegean-Southwest points out that ESB reached its peak in the later 1st century, and at Argos the shift is placed in the 70s.590 In some regions, ESB is of minor significance, and quantities are often too small to discuss in detail. This reverse pattern could fuel the idea that quantities were, as a manner of speech, drained from the Aegean-East to be taken elsewhere, even if the map (Figure 55) does not bear this out. If future research is able to corroborate this reverse pattern — including withstanding a better understanding about the typo-cronological framework into the 3rd century — one might think that an important shift took place sometime in the second half/late 1st century.

591  Athens: Robinson 1959, group G, layer IIb (70-100); Ephesos: Meriç 2002, Füllung D1 (75-100); Knossos: Sackett 1992, groups F1-F2 (c. 80-100). 592  Robinson 1959, group G, layer III (c. 75-125). 593  Sackett 1992. This reflects Sackett’s quantitative information: ESB comprises c. 43% and 57% respectively during the Trajanic and Hadrianic periods; Abadie-Reynal 2007, 113. 594  Robinson 1959, group M, layer II. 595  Abadie-Reynal 2007, 113. 596  Kition: Marquié 2004, US 8-10; Resafa: Konrad 1992, phase I. 597  Berenice: Kenrick 1985, deposit 69; Corinth: Williams II and Zervos 1985, deposit 50; Stobi: Anderson-Stojanović 1992, lots 392, 1538, 1957; Paphos: Hayes 1991, i.a. the later part of the fill in well 3; Tarsos: Jones 1950, trench 6 and the Concrete Wall Chambers. 598  Berenice: Kenrick 1985, deposits 72, 76; Corinth: Williams II and Zervos 1985, Oval Furnace; Tenos: Etienne and Braun 1986, group Ga; Paphos: Hayes 1991, the main fill of cistern 5, the upper part of well 21, well 13; Kourion: Neuru and Soren 1987, locus 014; Maroni Petrera: Tomber 2002, loci 25.14/32/33; Beirut: Reynolds 1997-1998, BEY 006.11629; Alexandria: Élaigne 1998, US 1213/1214; Caesarea Maritima: Riley 1975, level H4B. 599  Isthmia: Gregory 1995, deposit 90-1; Ephesos: Meriç 2002, Füllung D2; Knossos: Sackett 1992, groups D1, D3-6; Hayes 1971, the upper part of the well fill RR/K/60; Troia: Tekkök et al. 2001, well C29; Stobi: Anderson-Stojanović 1992, i.a. lots 692, 930.

ESB is the major terra sigillata in deposits of the late 1st century: part of a cistern fill at Athens, in Füllung D1 of 583  Alexandria: Élaigne 1998, US 1213/1214; Berenice: Kenrick 1985, deposit 76; Caesarea Maritima: Riley 1975, level H4B; Paphos: Hayes 1991, esp. ITS stamps 12-14, 18-20, 26; Stobi: Anderson-Stojanović 1992, lots 934, 938, 940 (part of this could be residual). 584  Sackett 1992, deposits D5-D6. 585  Knossos: Hayes 1983, deposit SH6; Stobi: Anderson-Stojanović 1992, lots 1474, 1477. 586  Abadie-Reynal 2004, 65. Since part of this is probably residual it distorts our view: Abadie-Reynal 2007, 65. 587  Abadie-Reynal 2007, 65. 588  See n. 126; also Poblome 2006a, 191, 199-200. 589  Lund 2003, 128-129, discussing the evidence from Ephesos. 590  Abadie-Reynal 2004, 64, 2007, 112-113.

74

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:35 UTC All use subject to https://about.jstor.org/terms

The Late Hellenistic and Early Roman Periods: Discussion and Chronological Phasing

Figure 56. Line chart showing the development, in absolute numbers, of ESB in the Aegean-East and the (combined) evidence for the other regions (© Philip Bes/ICRATES Project).

Knossos, Tenos, Berenice and Alexandria.603 Lastly, mid2nd century and later deposits (with a terminus ante quem of c. 200) are known from Corinth, Argos, Sparta, Tenos, Knossos and Paphos.604 Regardless of a token presence in the (second half) of the 1st century, recently published evidence from Argos shows that ESC occurred in a late 1st, early 2nd century deposit, and remained relatively modest throughout the first half of the 2nd century, when ESB still widely circulated on Aegean markets. The mid2nd century seems to mark a turning point, as ESC is neardominant in groups of the second half of that century.605

ESC ESC remains of great importance in and around Pergamon and the Troad. Its distribution pattern expanded considerably during this phase: substantial quantities are now found at Knossos, Berenice, Ephesos, in the Aegean-Southwest,600 and it is prolific at sites along the Aegean’s north coast.601 ESC remained scarce outside the Aegean: it is identified at Perge, Anemorion and Paphos. Figure 9 shows that ESC remained stable during phase 4, following a low around the second half of the 1st century. The substantial quantities of ESB and ESC at Knossos and Berenice suggest that these were probably distributed along a north-south axis. In fact, the geographical expansion of ESB (not everywhere) and ESC during phase 4, both common now especially at Knossos and Gortyna, could be related: perhaps both wares were shipped in joint cargoes till the mid-2nd century, when ESB disappeared (?) from interregional exchange.

ARSW Pre-200 ARSW forms are attested in small numbers in the east, and its quantitative impact remains very modest (Figure 55; appendix 4a). As a matter of fact, in the east, ARSW becomes important only after the early to mid3rd century onwards, a development that is possibly related (in)directly to the demise of ITS, ESA, ESB and ESD by the later 2nd century. The largest quantity is found at Berenice (17 specimens), followed by smaller quantities at, for instance, Antioch, Anemorion, Delphi,

Archaeologically, ESC (and ESB, cf. supra) is attested in a fair number of deposits. Late 1st and early 2nd century deposits are attested at Athens, Ephesos, Knossos, Troia and Berenice.602 Moreover, deposits of the first half and the second quarter of the 2nd century are attested at

603  Knossos: Sackett 1992, groups D3, D6; Hayes 1983, SH7, baulk 9; Tenos: Etienne and Braun 1986, group Ga; Berenice: Kenrick 1985, deposit 76; Alexandria: Élaigne 1998, US 1213/1214. 604  Corinth: Williams II and Zervos 1983, pit 1982-2; Argos: AbadieReynal 1995, sous-sol II; Sparta: Bailey 1993, deposit RSC 2W 4119; Knossos: Sackett 1992, groups D2-D3, D6, R1-R3; Hayes 1983, i.a. SH4-6; Tenos: Etienne and Braun 1986, group Gb; Paphos: Hayes 1991, room ZO. 605  Abadie-Reynal 2007, 129; between 9-15% up to the mid-2nd century, between 91-96% during c. 150-200.

To which can now be added Delphi: Pétridis 2010, 127-128, pl. 41, fig. 224. 601  Malamidou 2005. 602  Athens: Robinson 1959, group G, layer II-b; Ephesos: Meriç 2002, Füllungen D1-D2; Knossos: Sackett 1992, groups F1-F2, T1, T3, D4; Hayes 1971, the upper fill of well RR/K/60; Hayes 1983, baulk 8; Troia: Tekkök et al. 2001, well C29; Berenice: Kenrick 1985, deposit 69. 600 

75

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:35 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East Ephesos, Koroneia, Palaipaphos, Petra and Tanagra.606 The proximity to Tunisia should not be dismissed as a contributing factor concerning the more common presence of early ARSW at Berenice, which sets it apart from the east more generally. It is noteworthy that early ARSW also appears at Knossos and Paphos, harbour towns, and major urban hubs, such as Antioch that, like Berenice, are well connected to long-distance exchange routes. This skeletal (?) framework very much recalls the distribution of early (i.e. pre-1) ITS (cf. supra, 3.3.2). A variety of forms is found in a mid-1st century deposit at Berenice.607

At Knossos, single specimens are known from a late 1st to early 2nd-century deposit,608 a deposit dated c. 160-180, a mid-2nd century deposit, and an ARSW flagon form 161 from a 2nd century deposit.609 A late 2nd, early 3rd century deposit at Paphos contains an ARSW form 26.610 The collected data offers particular insights into the distribution of tablewares during the late Hellenistic and early Roman period. Next, chapter 4 seeks to set the geographical, quantitative and chronological data against the aims of this study.

606  Antioch: Waagé 1948, e.g. nos 847, 888; Anemorion: Williams 1989, nos 223-224; Delphi: Pétridis 2010, 126-127, pl. 41, fig. 220; Ephesos: Ladstätter 2010a, 178, a (very) late 1st-early 2nd century deposit from Hanghaus 2; Koroneia and Tanagra: Bes forthcoming a; Palaipaphos: Lund 1993, no. C227; Petra: Sieler 2004, no. 109. 607  Kenrick 1985, deposit 73

Sackett 1992, T3. Hayes 1983, SH 4, 5; E23; and J5 respectively. 610  Hayes 1991, deposit EN. 608  609 

76

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:35 UTC All use subject to https://about.jstor.org/terms

Archaeopress

Chapter Title: Framing Late Hellenistic and Early Roman Tablewares Book Title: Once upon a Time in the East Book Subtitle: The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East Book Author(s): Philip Bes Published by: Archaeopress. (2015) Stable URL: https://www.jstor.org/stable/j.ctvr43kch.9 JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected]. Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at https://about.jstor.org/terms

Archaeopress is collaborating with JSTOR to digitize, preserve and extend access to Once upon a Time in the East

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:38 UTC All use subject to https://about.jstor.org/terms

Chapter 4

Framing Late Hellenistic and Early Roman Tablewares and publication of pottery, for it would strengthen the methodological basis for comparison.611

Introduction Each phase and region discussed in the previous chapter has its geographical, chronological and quantitative characteristics; whereas some regions match one another, others do not, and although the quantity and quality of the collected data can definitely be improved, it seems highly undesirable to explain all these differences simply because of quantity and quality. Therefore, these differences suggest the existence of different (economic) exchange dynamics. The purpose of this chapter is to identify geographical, chronological and/or quantitative patterns in the evidence discussed in chapter 3; to seek and determine the nature of these patterns; and what these patterns can reveal about underlying mechanisms of tableware exchange in the eastern Mediterranean.

4.2 Contextualising Tablewares Tablewares obviously made a contribution to exchange patterns. It is nonetheless difficult to substantiate that claim. Their role, not only in quantitative and relative terms but also with regard to their value, is a largely unexplored field. This is not only so because of the state of the evidence, but also because of their role in the theoretical debate concerning the Roman economy. Yet, some basic outlines situate tablewares in Roman economy and society. Recently, Pieri stated that ‘le maintien de la vitalité économique’ is the ‘thèse qui est majoritairement retenue aujourd’hui’ and actually consists of ‘la thèse de la décommercialisation et la théorie du «state administered trade»’ and ‘la thèse de la continuité des échanges et de la production’.612 This, even if it pertains specifically to the late Roman period, introduces the modes of exchange that framed economic traffic, the people behind and within these, and perhaps most fundamentally, the actual goods and products.

4.1 Difficulties of the Evidence Whereas archaeological evidence is innately limited, the following limitations pertain specifically to the period under study: (1) several regions are comparatively wellrepresented, such as the Aegean-Southwest and the southern Levant; also for Crete and Cyprus the evidence can be assessed positively. Yet, lacunae hinder a more detailed analysis: what, for instance, about harbours such as Gaza and Tyre? Asia Minor’s south coast is also relatively unknown with regards to, for example, harbour towns such as Side in Pamphylia, whose intermediary role in seaborne traffic must have existed. Also, evidence often pertains to a specific period or category of terra sigillata; (2) a poor knowledge of the actual production of terra sigillata. That the whereabouts of the workshops of ESA remain unknown, is at the same time tantalising and frustrating; the view is consequently based on its distribution only. Evidence from Tunisia (ARSW) and Sagalassos (SRSW) not only allows us to investigate typo-chronological frameworks in detail, but also to map production and development; and (3) pottery studies still only rarely present a full quantification, and even then the methodology can differ: only rims, for example, a RBH, or a full sherd count (and weight). Other publications merely present a catalogue, with little to no knowledge as to how the catalogue was compiled. These differences prohibit the assessment and interpretation of the evidence on equal terms. Nevertheless, collecting evidence en masse, could to a certain extent, overcome these and other problems. Roman pottery studies are still a long way from a uniform methodological and publication platform, although much progress has been and is being made. Since practical considerations can often prove obstructive, ceramologists might strive for a more uniform approach in the study

4.2.1 Modes of Exchange Peacock and Williams envisaged three mechanisms that drove the distribution of agricultural produce, wood, marble and so forth within the Roman world, though applicable to other periods as well, offering archaeological and other evidence to explain these.613 First, redistribution, which means that goods were collected by a central authority, and subsequently redistributed to appointed people, settlements or areas. Secondly, reciprocity, essentially a system where goods are presented to another person, group or institution, with the social obligation for the latter to ‘return the favour’ at some point in the near future, so as to confirm a relationship of some sort. Thirdly, market exchange, selling goods in order to yield profit. The unravelling of such mechanisms through the use of archaeological evidence alone is not possible in the greatest detail because of the complexity of these systems and the simple fact that a single pottery fragment tells us nothing as to how it got to where it was found.614 However, by studying the evidence from different perspectives and using a broad(er) geographical and chronological basis, See Malfitana et al. 2006. Pieri 2005, 148. 613  Peacock and Williams 1986, 54-66; Wickham 2005, 694-708. 614  Ward-Perkins 2001, 173-174. 611 

612 

77

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:38 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East patterns emerge that can be explained as being the result more of one mode of exchange than another. Also, these modes of exchange will have functioned on different levels: all three modes may have been operational within a region, city, or even a small settlement.

mechanisms by means of local and (supra)regional markets, both physical and abstract.620 This involved basic food requirements, building materials, clothing, agricultural implements; the list is, figurally speaking, endless. A community basically aimed for self-sufficiency as much as possible, and interacted with real — for goods that were not or could not be made or grown within the settlement or area: Egypt, for instance, depended to a certain extent on external supplies of wood and olive oil — and/or more abstract markets: the provincial and imperial administration, which fuelled the circulation of, for example taxes, in either coin or kind. Besides these more official flows, literary sources mention products that were favoured for their quality or status. Whether that was indeed the case for red slip tablewares cannot be said — though note the example of rhosica vasa — the collected data nonetheless testifies that certain classes circulated widely throughout the Mediterranean.

4.2.2 Human and Institutional Factors Secondly, the main human agents actively or passively involved in economic traffic need to be briefly considered. Pieri lists the state, the church and the owners of large estates as the main parties involved in late antique trade; these had managed to increase their significance by possibly (partly) decommercialising trade.615 Peacock and Williams also discussed these parties as important driving forces in exchange patterns throughout the (Roman) period.616 The church, obviously, only became more important from the 3rd-4th century onwards. It should not be doubted that a considerable share of mechanisms in operation and the goods in circulation were the result of activities practised by these parties, whose role need not have been restricted to one mode of exchange. The role played by (independent) middlemen, peddlars, but equally the actual consumers in town and country, obviously cannot be neglected. References to such merchant middlemen occur with some frequency in literary sources of the time; possibly they acted mainly by order of the main parties mentioned.617 Yet merchants must also have worked more or less independently, selling and buying goods at regional fairs and markets to consumers. To make matters more complicated, a consumer could have been an individual, a family, a village, but also the provincial or imperial authorities (and later of course the church).

4.3 Interpretation of the Collected Evidence Black slip tablewares grosso modo characterise the production of pre-Roman tablewares, a tradition that in some areas apparently persisted into early Roman times.621 By the mid-2nd century BC, however, workshops in the area between Tarsos and Syrian Laodikeia fundamentally changed their firing technology, which set the tone for the subsequent millennium. ESA was the first large-scale exponent of this evolution — a formidable trendsetter — for some time accompanied by the transitional BSP. Since BSP has only recently been identified, its implications remain fairly unclear, and the fact that workshops of BSP and ESA remain an archaeological vacuum to date does not help. Nonetheless, it has been argued that ESA appeared c. 145-140 BC, even if opinion has differed in the past.622

4.2.3 Goods and Commodities It is impossible to capture here the myriad of goods and commodities that circulated in exchange patterns. It is more relevant perhaps to draw attention to different levels of circulation and the parties involved, and Davies’ model of the flow of resources and goods is instructive.618 By drawing attention to the multiple levels on which goods, commodities and so forth could have circulated, Davies captures the essence of ancient economic systems: their complexity. The modes of exchange defined by Peacock and Williams form an integral part of his model, and mirror a fundamental element of that complexity: that market exchange, redistribution and reciprocity will have taken place on a number of different geographical and social levels. Although Davies is largely concerned with pre-Roman society, the different levels of the flow of goods can in general terms be extrapolated to the Roman period.619 The intra- and interregional circulation of goods and commodities within the Roman Empire can be seen taking place on different levels and through different

4.3.1 Phase 1: The Late Hellenistic Period till the Reign of Augustus This section, from the late 2nd century BC to the Augustan period, focuses on the wide distribution of ESA in the eastern Mediterranean. A first set of forms (with a terminus ante quem of c. 75 BC) was distributed in considerable numbers throughout the Levant and Cyprus (Figure 50). These include forms 1-2, 16-18, 20, 105 and his form rara-a, and belong to a broader Hellenistic morphological repertoire; these forms are conveniently coined group I here. Considerable quantities were identified as far south as Petra, eastwards at Sultantepe and Dura Europos, while the somewhat enigmatic site of Kululu forms a De Ligt 1993; Bang 2008; Erdkamp 2005; Davies 2005, 143, fig. 6.8. Bailey 1993, 221; Martin 1997b; Zoroğlu 2005, 244; Slane 1991b, 151; Ladstätter 2007; in Egypt (Ballet 2002, 91): ‘leur longévité est intéressante: si elles ne semblent plus attestées au début de l’Empire dans le mobilier alexandrine, elles sont encore présentes aux Ie-IIe s. en province, et en particulier à Tebtynis dans des contexts dates du HautEmpire. Au Bas-Empire, les céramiques noires ont définitivement disparu du repertoire local alors que l’importation des sigillées […] influence la production des céramiques fines égyptiennes à engobe rouge, notamment celle du «Groupe K» (Rodziewicz 1976)’. 622  For a summary, Slane 1997, 257. 620  621 

Pieri 2005, 144. Peacock and Williams 1986, 56-59. 617  See e.g. Whittaker 1983; Pieri 2005, 144. 618  Davies 2005. 619  Davies 2005, 142. 615  616 

78

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:38 UTC All use subject to https://about.jstor.org/terms

Framing Late Hellenistic and Early Roman Tablewares northern limit of a core zone of distribution that also included Cyprus. ESA is rare beyond this, already, large area: further east, a few specimens were identified at Seleukeia ad Tigrim and Uruk,623 while westwards single or a few specimens were identified at Berenice, Gortyna, Knossos, Corinth, Kenchreai, in the Troad, possibly Stobi and now also Athens.624 Its rarity west of Cyprus can be coupled to, first, the notion that this was perhaps still a phase of experimentation and gradual growth for the ESA workshops: they were ‘scanning the market and consumer responses’, so to speak. That many other areas also manufactured (quality) tablewares could have posed an initial threshold, yet that these forms, even if characteristically Hellenistic, with a red slip surface were the ‘new kids in town’ may have increasingly triggered popularity, which could nevertheless be disparate geographically and culturally.625 Secondly, its predominant circulation in the Seleucid kingdom626 and to a lesser extent in the Ptolemaic kingdom, suggests that economic exchange was to some extent determined by political boundaries, at least in so far as ESA is concerned. Thirdly, the general circumstances of that period. Though the Roman Republic became increasingly involved in the political and military arena of the east, at least as early as the early 2nd century BC, it remains somewhat problematic to gauge in what ways (economic) interaction between Italy and the east materialised.627

Figure 51 captures the quantitative and geographical growth of ESA, illustrating the distribution of group II (phase 1b). The overall chart Figure 9 supports this, and actually shows two periods of major growth, the first around 100 BC, the second from the mid-1st century BC onwards. What these do not reveal is that the lengthy period of forms 3, 4A and 22A(-B) (the late 2nd century BC into Augustan times) implies that their distribution and popularity need not have been linear. Fortunately, the archaeological evidence discussed in section 3.2 provides some key points for the distribution of ESA at that time,629 as deposits of the late 2nd, early 1st century BC from sites west of Cyprus did contain specimens of mainly these forms, even if this still does not answer the question whether these forms were common throughout their date range. In light of the distribution of Italian table- and cooking wares, perhaps far more specimens of especially these forms of ESA are later within their date range than can be presently proven.630 Forms of group I rarely figure in deposits datable to phase 1a. Though evidence is meagre, it is not unthinkable that forms 3, 4A and 22A(-B) were, by preference, taken westwards, as the map for phase 1b suggests (Figures 51 and 57),631 and that forms of groups I and II were more mixed in the core zone of ESA, though not necessarily.632 Even if the archaeological evidence is not overwhelmingly rich, it hints at what the distribution of ESA may have been like from the late 2nd century BC on. Forms of group II have been identified in later deposits (including Augustan); those without an archaeological context nevertheless testify to their popularity in the east.

According to current typo-chronological knowledge, a small range of new forms appeared around 125-120 BC; these continued into Augustan times. These, chiefly Hayes forms 3, 4A and 22A(-B), as well as forms 5-11, 15, 19, 23-25 and 101-102, are coined group II. Some of these overlapped with group I for about one to two generations (from 125-120 to c. 75 BC). Forms of group II, however, notably 3, 4A and 22A(-B), became extremely popular in the east; in addition to still being Hellenistic these were innovative shapes, a duality that combined with the new look could well have augmented their popularity.628

The innovative character of these forms was not the only factor that contributed to their popularity. A more satisfactory view is to see that this was an indirect result of growing Roman interference in the east, thereby successfully bringing areas under their political, diplomatic and military wings, and which added to the dwindling power of the Seleucids. By the late 2nd century BC Pamphylia and Cilicia were within the grip of Roman political interference.633 During the 60s and 50s BC, the defeat of piracy, the creation of the Roman provinces of Syria and Cyprus and the end of the Mithridatic threat surely contributed to more peaceful conditions in the east, which possibly stimulated (economic) interchange between Italy, the Aegean and the Levant. It is then tempting to couple the steep (continued?) increase of ESA in Figure 9 around the mid-1st century BC to these developments?

623  ESA fragments have been published from Uruk (Strommenger 1967), Babylon (Deubner 1957) and Seleukeia ad Tigrim (Debevoise 1934). Since these sites were located within the Seleucid kingdom this is not surprising. That, in fact, so little ESA has turned up in these areas is noteworthy, perhaps caused by pottery production in the Seleucid kingdom proper, and/or the novelty of ESA which was perhaps mostly drawn westwards? 624  Hayes 2008a, 26. 625  Abadie-Reynal 2007, 87. 626  Lund 1997, 209: ‘Eastern Sigillata A was certainly the ceramic fine ware par excellence of the Seleucid Kingdom’ (after Lund 1993, 39-40). 627  For (eastern) findspots of Campana Ware, see e.g. Kenrick 1985 (Berenice), 9-29; Bald-Romano 1994 (Corinth), nos 30-36; Rotroff 1997a (Athens), nos 1638-1645; Kenrick 2000 (Gadara), no. 197; Élaigne 2000 (Alexandria), fig. 1.4; Slane 1997 (Tel Anafa), nos FW455-465; Sieler 2004 (Petra), no. 151. For Campana A finds in the southern Levant, see Handberg et al. in press. For the standard work on Campana Wares, see Morel 1981; for a socio-cultural case study, see Roth 2007. 628  Some forms of group II (e.g. form 6) are very similar to shapes in Campana Ware, and an influence should not be doubted. It is more problematic, however, to determine the direction of this influence. Since the workshops of ESA are not located, the matter largely remains theoretical.

629  E.g., a deposit (c. 110-90 BC) in Hanghaus 1 at Ephesos includes Hayes forms 3, 4 and 22 (Ladstätter 2003); at Halikarnassos, specimens of forms 4 and 22 were found in a fill dated to 100-90 BC (Vaag 2002); a well fill at Knossos, dated to c. 100-25 BC, included specimens of Hayes forms 4, 5 and 22 (Hayes 1971). 630  Lund 2004. 631  Hayes (1985a, 23) has suggested that forms 3-4 and 22A(-B) (and 5) formed a service. Even if this was intentional, consumers in, for example, the Aegean may have had (a) different framework(s) regarding services. 632  Jackson and Tidmarsh 2011, 328, table 16; Hayes forms 3, 4A and 22A also dominate amongst the as yet unpublished ESA from Kinet Höyük (abandoned c. 60-50 BC). 633  Sherwin-White 1976.

79

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:38 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East

Figure 57. Column chart comparing the distribution (in percentages) of ESA forms of phase 1 between the north Levant and Cyprus, and the central Mediterranean (© Philip Bes/ICRATES Project).

As economic traffic involved much more than tablewares, one must be extremely cautious in suggesting any direct association between historical dynamics and tableware distribution. The distribution of goods in the east seems to have continued within existing exchange systems.634 Egyptian grain, for one, appears to have been an important commodity within the pre-Augustan east, and this and other sources perhaps followed a system of redistribution that centred on Rhodos;635 after 167 BC the balance must have shifted in favour of Delos. The enormous quantity of Rhodian amphora stamps at Alexandria, totalling about 100,000 examples, could well be a testimony; perhaps this system also triggered the manufacture of red slip tablewares.636

after the demise of Rhodos. In general, however, Italian products appear to have been rather scarce in the east prior to Augustus, with the exception of a few centres such as Delos, where ‘[t]he unparalleled varied spectrum of Italian ceramics’ includes table- and cooking wares that were, to a large extent, thought to have been personal possessions.639 With its relatively large Italian population, as well as eastern families and socio-religious associations, Delos acted as a binding factor in (economic) contacts within the east and between west and east.640 This probably included the channelling and (re)distribution of Egyptian grain and Rhodian amphorae, as well as western (Italian) products; the infamous slave trade, designed to supply Italy with large numbers of slaves, reportedly also went through Delos. Athens was most likely supplied with Italian and other goods via Delos, whereas Alexandria in turn acted as a redistributive centre in the east. This in any case is the likely reflection of Italian interests in the east,641 which focused on major sites in order to exert political-administrative and economic control. Delos was deprived of its role in 87 BC after being destroyed during the conflicts between Mithridates and Rome. A second destruction in 69 BC probably put an end to this for good. The early ESA at Delos, through which it probably arrived at other sites in the Aegean,642 may partly have come via Alexandria within the context of the exchange of grain and amphorae, which would in fact also explain the presence, by about 120 BC, of a range of Italian products at Alexandria.643 Subsequent redistribution from

The evidence for western (i.e. Italic) ceramic products in the east provides further evidence to assess the level of interaction. About the distribution of Italian pottery and economic development during this period, Lund recently emphasised that ‘[w]e are obviously confronting a complex situation which cannot be explained by a single, simplistic hypothesis’.637 He draws attention to the fact that in important cases the distribution of Italian amphorae differs to that of tablewares. Whereas Italian tablewares were quite common between 250-100 BC at Berenice, amphorae were scarce, even if this shows that Berenice was linked to Italy. A reverse scenario seems to be the case at Athens,638 and Alexandria (cf. infra), where requirements for (Italian) wine and olive oil may have been contributing factors, which perhaps even augmented

639  Lund 2004, 12; Peignard 1997, Peignard-Giros 2000 for pottery from the Maison des Sceaux; c. 13% is Italian; also see Morel 1986. 640  Rauh 1993, esp. xviii-xix, 28-58, chapter 2. Also Le Dinahet-Couilloud 1997. 641  Malfitana et al. 2005. 642  Poblome et al. 2001b, 144. 643  Hayes and Harlaut 2002, 99-131, including cooking and thin-walled

See e.g. Lund 1997, 211, 2006b. Rickman 1980, 118-119. 636  Élaigne 1999. 637  Lund 2004, 12. 638  Lund 2004, 12; see also Kenrick 1985, 494. 634  635 

80

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:38 UTC All use subject to https://about.jstor.org/terms

Framing Late Hellenistic and Early Roman Tablewares Alexandria may explain the presence of Italian ceramic products elsewhere in the east.644 If the archaeologically visible Italian pottery evidence is indicative of the level of interaction, except for the major towns this then remained at a fairly low level. In turn, ESA arrived in Italy probably also through Delos.645 The rather common presence of late Hellenistic ESA in Italy contrasts with the current evidence for the central Mediterranean: ESA is best represented in southern Italy and the Campanian sites of Pompeii and Puteoli.646

identified at Pergamon, Assos and to a lesser extent at Troia and Smintheion. Elsewhere, examples were found at a number of Aegean sites, including Athens,652 and further east at Alexandria653 and Tel Anafa. Delos received ESC in considerable quantities, perhaps directly;654 presumably, the distribution of ESC in the east (including the Aegean) also took place via Delos:655 ships leaving Delos contained mixed cargoes that were assembled on the island, and included ESC, Italian pottery,656 Rhodian amphorae and Ionian mouldmade bowls.657

That Italian tablewares were uncommon in the east between c. 50 BC and the arrival of ITS is another interesting point to which Lund draws attention. The Italian amphora evidence ‘supports this observation to a certain extent as far as the Aegean is concerned – but not in the Levant’ as ‘E.L. Will noted a dearth of datable amphorae in Athens and Delos during the second quarter of the 1st c.’.647 The fates of Athens and Delos, as well as other political and military developments, echo Slane’s observation that ‘[i]f it was only in the Tiberian or Claudian period that the range of Italian imports again reached the level they had been about 100 BC, we may begin to comprehend the disastrous impact of the civil wars on the East’.648 This could of course also have affected the distribution of ESA and other wares during phase 1, or our perception of it: the wider developments of the 1st century BC are rather equivocal for our understanding of the distribution of tablewares. On the one hand, one might expect that the destruction of Athens and Delos during the Mithridatic Wars, as well as civil wars later in the 1st century BC, negatively affected exchange patterns. On the other, the defeat of piracy, and the creation of the Roman provinces of Cyprus and Syria, in the 60s BC could have incited more stable conditions in parts of the east, which the Athenian evidence in any case bears out.649 Framing the distribution of tablewares (artefacts) within a changing political and economic landscape is thus not straightforward;650 in fact, distribution need not have (closely) conformed to wider developments at all.

ESA probably also triggered the start of ESD.658 Also Cyprus had a long tradition of tableware manufacture, not least with the Colour-Coated Wares defined by Hayes,659 with which ESD shares macroscopic characteristics. Small numbers of ESD are attested mainly at Paphos and a few other sites on Cyprus during phase 1a (Figure 50, section 3.2.15), including the southern and southeastern coasts. The published record suggests an absence there during phase 1b, though generally ESD mainly catered for western Cyprus; some specimens were found on Crete,660 at Berenice and Tel Anafa (Figures 50-51; appendix 4a). ESA remained the most common tableware on Cyprus, including the western part, during phase 1. Cyprus was thus firmly included in the distribution pattern of ESA, probably receiving ESA directly through coast-to-coast exchange between Cyprus and the northern Levant. ESC and ESD clearly operated on different levels: whereas ESA relatively quickly gained a supraregional distribution, ESC and ESD remained mostly of regional importance. For ESC it was recently stated that ‘it is important to note that the producers of the ware did not use the creation of the province of Asia to their benefit’,661 although Sardis and Ilion (where ESC is attested) fell within the Pergamene sphere of influence. In general this could also apply to ESD, since the creation of the Roman province of Cyprus in 58 BC seems to have affected neither its production nor distribution.662 Exchange between Asia and Delos possibly became more regular after the creation of the province of Asia (133-129 BC); the presently known evidence, however, only suggests this for Delos.663

The area of Pergamon became actively involved in the manufacture of ESC from around the time the production of ESA began; also this area had an existing tradition of tableware manufacture. However, not until the mid-1st century BC did these workshops made the transition to fully red slip tablewares.651 ESC remained of regional importance (Figures 50-51): important quantities were

Hayes 2008a, 50-51; Rotroff 1997a, 1, 222, 408-410. Élaigne 1999, 222, ns 16-17. 654  Bruneau 1991. 655  Evidence for ESC in the east is thin; it is attested at, e.g., Alexandria, Tel Anafa, Sagalassos, Pednelissos, possibly an appliqué at Kinet Höyük. Vessels with appliqués were also manufactured at Knidos: Kögler 2010. 656  Morel 1976, 491. 657  Élaigne 1999, 222, n. 15. 658  Hayes 1991; Lund 1997, 203, 2002, 206: ‘Eastern Sigillata A presumably exerted some influence as well, and the impetus to start a redgloss production probably came from this ware, which was exceedingly popular in Cyprus in the late 2nd and 1st centuries B.C.’. 659  Hayes 1991. 660  Lund 1997, 204-205, 209-210 (after Hölbl 1994): ‘whereas Cypriot Sigillata was mostly distributed in an area that used to be part of what may be called the Ptolemaic Commonwealth’. 661  Poblome et al. 2007, 224; Poblome and Zelle 2002, 275. For epigraphic evidence that pertains to Italian citizens and patroni in the east, see Eilers 2002. 662  Lund 2002, 206. 663  Rauh 1993, 43; Bruneau 1991. 652  653 

wares. Non-Italian products such as Ionian mouldmade bowls could also have come through Delos. 644  Berlin 1993, 1997. 645  Malfitana et al. 2005. 646  Malfitana et al. 2005, 200-201, tables 2, 5, appendix (205). 647  Lund 2004, 12. 648  Slane 2004, 40-41; perhaps also Hayes 2008a, 59, 62, and Rotroff 1997a, 1, 14, 406-407. 649  Hayes 2008a, 20. 650  Lund 2002, 206-207. 651  The black band around the rim of certain forms indicates that an oxygen-rich kiln atmosphere was a familiar technology but that the way of stacking (purposefully?) prevented a fully red slip surface; Élaigne 1999.

81

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:38 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East The supraregional distribution of ESA during phase 1 can also be used to underline certain (morphological) preferences. ESA is characterised by a fairly restricted number of forms during phase 1, and it has been argued that notably Hayes forms 3, 4A and 22A(-B) were widely distributed (beyond the core zone), possibly even as a service. These three forms clearly had commercial potential and appealed to consumers for aesthetic and/ or functional reasons (Figure 57). When forms 24-25 — mouldmade bowls664 — are removed from the data and percentages recalculated, the preference of forms 3 and 4A remains, whilst form 22A(-B) becomes more common in the northern Levant and Cyprus. The chart includes all of Hayes’ late Hellenistic forms, supporting the notion that forms 3, 4A and 22A(-B) indeed dominated the ESA repertoire at the selected central Mediterranean (here: Aegean-Southwest, Crete and Africa-Cyrenaica) sites. More important, however, is the fact that this implies knowledge on the part of shippers, merchants and others of what and what not to take on board, for example mouldmade bowls. Some in the exchange system must have been well aware that there was no point in shipping ESA mouldmade bowls westwards, since production of such vessels still flourished at a number of centres in the central Mediterranean. Two other such examples for ESA are closed forms and the chalice Hayes form 15, both rarely found outside the core zone.665 That said, we need to reckon that there was also a practical and economic rationale behind the (long distance) trade in tablewares: certain shapes were more easily stacked and/or shipped, or took up comparatively less space.

from nowhere.667 SRSW may be seen as a third ware,668 one of surely many classes of red slip tableware whose distribution was essentially regional. Secondly, the morphological change that several classes of eastern sigillatae underwent, which may have been either directly or indirectly inspired by Italian shapes; surely, the reality behind this was complex.669 In this context, ESB was probably an exceptional case, since Italians are thought to have been involved. However, ESA shows strong affiliations with ITS from the Augustan period onwards. The processes behind this were probably multi-layered. As owners of the resources for setting up a workshop or manufactory, the socio-cultural elite is thought to have been instrumental. They owned land and capital, and could not only provide the spatial and financial infrastructure but also clay, fuel, water and so on; more importantly, setting up a workshop probably occurred in addition to more profitable (agro-economic) activities, e.g. cash crops. Potters played only a secondary role, though much of course depended on their technical proficiency. Typological inspiration is thought to have derived from metalware shapes owned by the elite. This phenomenon, skeuomorphism, is not uncommon in the Roman world, and certainly not restricted to metalware and ceramic shapes alone.670 The third party are (potential) customers, as obviously tablewares had to be sold.671 Interestingly, not all major classes absorbed this morphological inspiration. Next to ESA and ESB, certain forms of ESC and ESD also betray an Italian influence. This, however, remained skin-deep, particularly with ESD of which ‘none of the forms in question were apparently produced in great quantities’.672 The influence is slightly more manifest for ESC forms; in fact, these are even considered to predate ITS, implying that ESC inspired ITS.673 The frequency and distribution of italicised ESC forms was restricted to a core zone, which was not the case for ESA. It is noteworthy that those involved in the initiation

4.3.2 Phase 2: The Augustan-Tiberian Period Under Augustus the Mediterranean and parts of the lands beyond became part of one political and military system for the first time in history. It is generally thought that this marked a time of change, affecting different areas of society, though some aspects and customs also saw continuation. It is within this context that the production and distribution of terra sigillata underwent several profound changes.

Ladstätter 2007, 208; Hayes 2008a, 31, ns 5-6. Poblome 1999. SRSW was (near-)monopolistic at Sagalassos and probably to a large extent for Pisidia from the Augustan period onwards; products from elsewhere appeared in minute quantities during late Antiquity, esp. three ARSW vessels — Hayes forms 99A, 104B and 105: Bes forthcoming b. SRSW was also distributed west- and southwards in considerable quantities, including Laodikeia ad Lycum (Duman 2014), Pednelissos (Kenkel 2007) and Perge. The distribution further afield is gradually better understood; very small quantities have recently been attested at Athens (e.g. Hayes 2005, 22, fig. 14), and perhaps Corinth (examined during a visit to Corinth in early November 2004 with Jeroen Poblome, though given the highly micaceous fabric, at least one vessel should be ESB: that ESB forms 62A and 74B bear resemblance to SRSW form 1C191 might cause confusion; I warmly thank Kathleen Slane and the American School for their hospitality). More doubtful, at least from a morphological point of view, are finds at Ashkelon and Caesarea Maritima (Johnson 2008a, b), which as such bring back to mind the case of LRD (Poblome and Fırat 2011). 669  Willet 2012. 670  Vickers 1994. 671  Poblome and Brulet 2005. 672  Lund 2002, 197-201, 206. 673  Meyer-Schlichtmann 1988, 194; although theoretically possible, Slane (1991b, 151-152) has argued against it, drawing attention to the metal prototypes that provided the basic inspiration for at least part of the manufacture of terra sigillata, and perhaps tablewares more generally. 667  668 

First, two major tablewares appeared in the east: ITS and ESB, and it is generally accepted that around 2520 BC certain ITS potters instigated the production and morphological spectrum of ESB, either through a branch workshop set up by Italian workshop owners and/ or potters, or manufacturers closely following ITS (or its metal prototypes) that started to arrive in the east;666 recent evidence, however, shows that ESB did not appear 664  Hayes (1985a, 24-25) originally dated his form 24 to the 1st century BC and continuing into the 1st century, his form 25 to continue even into the 2nd century. In light of the chronological development of the mouldmade bowl more generally, this appears to be rather late, not in the least within the context of the morphological repertoire of ESA and the changes it underwent during the early imperial period. Now indeed see Hayes 2008a, 22, table 2, where a date not later than Augustan is mentioned for both forms. This has considerable implications. 665  See, for instance, Abadie-Reynal 2007, 89. 666  Zahn 1904; Zabehlicky-Scheffenegger 1995, 1996.

82

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:38 UTC All use subject to https://about.jstor.org/terms

Framing Late Hellenistic and Early Roman Tablewares of SRSW made three decisions; the transition to a fully red slip tableware from about 25 BC onwards, yet just about ignored the morphological inspiration on offer (mostly ITS, or its metal prototypes), and reorientated themselves to the exploitation of clays that were now exclusively quarried at Çanaklı, located about 8km southeast of Sagalassos.674 The evidence for stamps supplements this model. Again ESA and ESB picked up this Italic phenomenon, although both wares (especially ESA) interpreted it differently. Stamps are rather uncommon on ESC675 and ESD (and all but absent on SRSW). Although a ratio between stamped and unstamped vessels is difficult to determine, the custom was certainly not as common as on ITS.

resources, and the production of artisanal items.677 Within this context, the evidence for phase 2 strongly suggests that the direction, intensity and scope of the exchange of tablewares to and within the east increased. Recently, the Mediterranean-wide apparent uniformity of terra sigillata, a binding factor that climaxed by this time, was defined as the ‘tableware boom’.678 ITS, since it comes from the west, best allows us to look into underlying aspects of distribution patterns. Figure 53 shows that pre-1 ITS has been identified in the east, and that it was most common at Corinth; elsewhere only one or no more than a handful of fragments has been identified. Corinth is one of the better-studied sites in the east, which probably distorts the published record; ITS was nonetheless much more common in the west. It would not be unexpected that early ITS mainly travelled along the major lines of communication. This is reinforced if we look at the locations where early ITS was identified: Corinth (and some sites along the Corinthian Gulf), Knossos, Gortyna, Ephesos, Berenice, Alexandria, Beirut, Antioch, as well as Jerusalem and Petra; Athens and Argos can now be added.679 Presumably it is no coincidence that a number of these were large(r) urban centres with a role in the provincial administration, and mostly also harbours with easy access to the Mediterranean. The distribution pattern of ITS for phase 2 as a whole suggests that its dispersion in general developed fairly slowly and gradually.680 In fact, the late(r) Augustan and Tiberian periods (perhaps also the Claudian period) are often seen as the period when ITS was typologically and archaeologically most common.681 ITS comprised c. 63% of the imported terra sigillata between 25-50 at Corinth (and in the Aegean and on Crete more generally), a percentage unparalleled in the east except for some Red Sea harbours. ITS now also appeared in the southern Arabian peninsula and India.682

Thirdly, of chief interest here, is the quantitative and geographical change in the distribution of these tablewares during the early Empire (Figure 52), which is in great contrast with phase 1(b). The hegemony of ESA in the Levant came to a halt, and Cyprus also presents a more mixed picture, though ESA and ESD still dominated. ESC maintained its dominant position in and around Pergamon and in the Troad, though small quantities of other wares now came in. The region of Ephesos presents a more mixed picture. Although ESB made a considerable impact from the start, other wares (except for the rare ESD) thus maintained or acquired an importance. The picture is especially mixed for Africa-Cyrenaica, Crete and the Aegean-Southwest, where no major tableware production existed.676 Particularly ITS, but also ESA, ESB and ESC were important, whilst ESD was of some importance only on Crete. Other categories were and mostly remained significant on a regional level: for example, SRSW was of some importance at Perge. The relatively peaceful Pax Romana created favourable circumstances for, amongst others, economic exchange. Part of the (urban) population in Italy and elsewhere acted as a major magnet for goods and wealth from across the Mediterranean, which perhaps reflects Augustus’ words that he found a city of brick, and left a city of marble. The exchange of goods on a regional level must also have continued. Rome had already adapted to its socioeconomic role during the late Republic (cf. supra, 4.3.1), which crystallised in the Pax Romana. By drawing resources from across the Mediterranean (and beyond) in support of military and administrative systems, a desire for semi-luxuries, and a tax system applicable to all, it is thought that regional and supraregional, commercial and non-commercial, production and exchange was thus predominantly stimulated, not created, including the growing of cash crops, the exploitation of natural

Some interesting aspects are revealed by closer study. In a recent paper, based notably on ITS, Slane convincingly argued that geographic proximity was not the only or 677  Scholars hotly debate growth (theory) in the ancient economy. Those in favour, amongst others, draw upon a range of archaeological sources to point out that quantities increased during the late Republic and early Empire. Those against consider the definition of growth, and point to important artefact categories that have since perished. See e.g. Hopkins 1980, 1983; Jongman 2002; De Callataÿ 2005; Hitchner 2005; also Millett 2001. 678  Poblome et al. 2001b, 144; Poblome and Zelle 2002, 277. 679  Hayes 2008a, 42-43, 45, noting a geographically different pattern; Abadie-Reynal 2007, 63. 680  Abadie-Reynal 2007, 64. 681  See e.g. Hayes 1985a, 2008a, 41-44; Kenrick 1985, 493; Slane 2004; Abadie-Reynal 2004, 2007, 62-64; Zabehlicky-Scheffenegger 2004; Élaigne 2004. 682  The largest quantity published to date is that from Arikamedu in India: Wheeler et al. 1946; Casal 1949; Slane 1996b. Renewed investigations by Begley (1993, 1996; Begley and De Puma 1991) between 1989-1992 brought to light further fragments of ITS; a few sherds of ESA and ESB are now recognised (Comfort 1991; Slane 1991a, 1996b); also see n. 708 for Muziris. Small quantities are known from the Arabian peninsula, see e.g. Comfort 1958, 1960; Sedov 1992 (now also 2007); an early ESB vessel from Axum in Ethiopia (De Contenson 1963). These finds chiefly date to the late Augustan-Tiberian period although the occasional later fragment exists.

674  Considering the high number of Augustan coloniae in Pisidia, this challenges expectations (Levick 1967); also, ITS has not been identified at Sagalassos, which seems an important imperative for the association between Italian presence and Italian material culture (Poblome et al. 2001b, 149). The clay quarried at these new sources was used for the body, whereas for the slip a refined clay from the Potters’ Quarter was used. 675  Meyer-Schlichtmann 1988, 182-187; Hayes 1991, 56-57. 676  Manufacture was practised at e.g. Patras (Hübner 2003) and on Crete (see e.g. Marangou-Lerat 2002, 68-69).

83

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:38 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East determinative factor for a site to receive considerable quantities of ITS. In the case of Corinth, its colonial and administrative status as well as its fertile countryside probably contributed to the (exceptional) position of ITS.683 Corinth is a relatively well-published site, so Slane was able to show that Corinth received ITS much earlier, in larger quantities and for much longer compared to other sites in the Aegean-Southwest, possibly even within the east. It was recently argued that Argos received ITS (and other goods) by redistribution from Corinth,684 and a redistributive mechanism could well have worked elsewhere in central Greece: a comparison of ITS stamps suggests that the route through the Corinthian Gulf, via the diolkos across the Isthmus into the Aegean was not exclusive; ships could have sailed around the Peloponnese. In addition, several stamps known for Ephesos were not identified at Corinth, and of these none are known for Knossos.685 ITS is less common at (eastern) Aegean sites: it is attested at Ephesos,686 yet marginally only at Pergamon and Troia;687 the ESB and ESC manufactories offer a probable explanation. ITS is, however, quite common on Crete, comprising c. 43% and 38% in the Claudian and Neronian periods respectively at the Unexplored Mansion at Knossos.688 It was also not uncommon at Gortyna, though few stamps are published.689 The state of the published record does not only explain the high percentage of ITS for the Aegean-Southwest and Crete (cf. supra, 3.2.2-3). Crete was probably a major stop for ships sailing eastwards, and that Knossos, Gortyna and Corinth were coloniae may have been a further contributing factor; Patras can be expected to fit this picture as well. The lower percentages of ITS for Crete could indicate a general falloff effect: the further east, the less available. In general terms, we can observe this also for, for example, ESA.

Although ESA still largely dominated further east, thereby still penetrating beyond the (major) coastal inlets, ESD grew in significance in the southern Levant and on Cyprus, and generally continued to reach Cyprus’ closest coastal zones.693 ESA is still found in considerable quantities in the Aegean-Southwest, Crete, at Ephesos and Berenice into the Augustan period (ESD remains negligible), yet its role in the central Mediterranean began to diminish by or after the early 1st century (Figure 9; also phase 3). That ITS, ESB and ESC were identified at coastal and non-coastal sites in the eastern regions is indicative of a continuation (and intensification?) of exchange between the Aegean and the Levant, though the full quantification at Tel Anafa shows that wares other than ESA played but a marginal role.694 Figure 52 suggests that ITS arrived in the east in relatively large quantities, a picture possibly distorted through a preference to publish ITS, and stamps in particular. It was already argued that Corinth, Knossos, Gortyna and Ephesos were among the major points of exchange involving the Aegean and Africa-Cyrenaica, tablewares being but one aspect of that exchange. Findspots in the east of the earliest ITS, ESB and ESC provide a clue to substantiate this idea; their small quantities mostly did not travel beyond Alexandria, Paphos, Caesarea Maritima, Beirut, Antioch and Tarsos. This receives further support from an observation by Lund, that ‘[t]he importance of urban demand for commodities of all sorts and the pivotal role of Ephesos, Nea Paphos and Antioch in the political, military and economic life of the surrounding provinces need hardly be emphasized. It is surely no coincidence that the ceramic industries in question were located in their vicinity. A comparable situation seems to have existed in North Africa, for the workshops which began making African Red Slip Ware about AD 60 were probably placed in the region of Carthage’.695 Tentative clues are further provided by Africa-Egypt, where ITS was also common at sites studied. Although this observation is mainly based on evidence from Alexandria, Myos Hormos and Berenike, ITS seems to have been common in Alexandria nonetheless. It is already indicated that (the negligible quantities of) ESB and ESC in the Aegean-Southwest and on Crete arrived through regional exchange, and those found at eastern destinations were shipped together with ITS. Therefore, it should not be surprising to find these three wares at Alexandria, though one cannot ignore, for instance, their presence at Tel Anafa. In part of the east, the dominance of ESA probably prevented other tablewares from penetrating into smaller (urban) and rural settlements in very small quantities;696 most were probably absorbed by urban markets.

Small quantities of ESB and ESC at sites in the AegeanSouthwest and Crete provide evidence for more regionallybased exchange. Also here, redistribution through Knossos could explain the ITS and ESC at Kydonia/Akrotiri and Vrokastro;690 perhaps Gortyna functioned similarly with regard to Kommos (ITS, ESB, ESC) (Figure 52), and small sites along the south coast (ITS, ESD).691 Evidence from the area of Pergamon and the Troad shows that ESC had now firmly established itself, even if the significance of ESB and ESC beyond western Asia Minor remained fairly limited.692 Slane 2004, 40-41. Abadie-Reynal 2004, 64-65, 2007, 63 (also lamps, amphorae); Slane 2004, 32; Bes and Poblome 2006, 156-157. 685  Zabehlicky-Scheffenegger 2004: OCK53, 242, 415, 423, 536, 646, 1594, 1808, 2348, 2368 and 2431 are not known from Corinth. Given the large number of stamps known from Corinth this suggests that these reflect other exchange patterns. See Baldwin Bowsky 2011 for an overview of ITS stamps from Knossos and Crete. 686  Zabehlicky-Scheffenegger 2004. 687  Pergamon: Japp 2003, 243-244; Troia: Tekkök 2003, 237-238. 688  Sackett 1992, 153, 163, fig. 2, which may well include residual fragments. 689  Some synthetic data was recently published (Panero 2008), though chronologically not immediately useful. 690  Kydonia/Akrotiri: Raab 2001; Vrokastro: Harrison and Hayden 2005. 691  Kommos: Hayes 2000a; south coast: Blackman and Branigan 1975. 692  For ESB, see Hayes 2008a, 36, with notable exceptions. 683  684 

Lund 1997, 205. Slane 1997. Lund 2003, 131. 696  See e.g. Slane 1997; Silberstein 2000; this suggests that geographical proximity was a factor influencing the proportions for the AegeanSouthwest and Crete. 693  694  695 

84

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:38 UTC All use subject to https://about.jstor.org/terms

Framing Late Hellenistic and Early Roman Tablewares ITS stamps in particular, however, from Berenice and beyond, furnish indirect evidence for the pivotal role of Alexandria. Puteolan stamps (as well as some from other workshops) are found at Oboda, Petra, Myos Hormos, Qana’, Arikamedu, and at Alexandria,697 and Kenrick noted a ‘strong Puteolan element amongst the Italian Sigillata’ at Berenice.698 Moreover, ESB I has been recognised at these sites (except Qana’), and also at Berenike, Khor Rori (Sumhuram), Timna’ and several sites in Ethiopia. ESB is even thought to have attained its peak at Alexandria already during the first half of the 1st century.699 The volume of this eastward flow of Roman tablewares should not be overrated;700 Mediterranean wine amphorae and other agricultural commodities, and items such as bronzes, had far greater economic significance. In any case, the distribution of these wares suggests that Alexandria, or one or more harbours in the southeast Mediterranean, was the most likely point of arrival. From there, goods (including tablewares) found their way to, for instance, Petra and Berenike, that owed part of their significance to their location along the caravan and sea-routes that ultimately connected central Asia, India and lands further east with the Mediterranean.

affected the importance of Puteoli.704 The peaceful conditions of the early Empire thus may have favoured supraregional exchange between the east and Rome. The evidence, however, notably that of ITS, suggests that this may have been a fairly slow process, thus echoing Slane’s words about the impact of the civil wars in the east.705 It is probably not coincidental that roughly during this phase (and into phase 3) Mediterranean tablewares are attested, in small quantities, roughly along the sea (and land?) routes to India (e.g. Myos Hormos, Berenike, Qana’, Shabwa,706 ed-Dur,707 Arikamedu, recently also at Muziris): the Pax Romana had far-reaching effects, even if the exceptional geographical extent of Mediterranean tablewares was short-lived.708 The development of exchange patterns during the late Republican era may in turn have bolstered regional exchange along existing trade routes in the east, as the distribution patterns of ESA and ESD suggest; a comparison with phase 1b supports this view. One further point must be made. Much emphasis is put here on Alexandria, yet that emphasis focuses on official exchange, particularly the grain supply to Rome. To what extent small goods such as tablewares were carried aboard ships carrying official cargoes, or whether these belonged much more to the realm of commercial exchange cannot possibly be answered on the basis of tableware distribution alone.

If Alexandria was indeed the chief source of Puteolan stamps found in the southeast Mediterranean, this possibly reflects the importance of Africa-Egypt to Rome, in which context the grain supply easily comes to mind.701 Egyptian grain had been important within the Hellenistic east, and whilst it had been shipped to Rome since the mid-2nd century BC, its importance seems to have materialised only during the early Empire: ‘[t]oday we saw some boats from Alexandria – the ones they call ‘the mail packets’ – come into view all of a sudden. They were the ones which are normally sent ahead to announce the coming of the fleet that will arrive behind them. The sight of them is always a welcome one to the Campanians. The whole of Puteoli crowded onto the wharves, all picking out the Alexandrian vessels from an immense crowd of other shipping by the actual trim of their sails, these boats being the only vessels allowed to keep their topsails spread’.702 Regardless of the channels through which the grain was commanded, i.e. taxation, rent or private purchase, Puteoli is regarded as the major port for goods to enter and to leave Rome, i.e. Italy: ‘[b]y the time of Augustus Puteoli was a major entrepôt and packet port for goods and passengers travelling between the Tyrrhenian coast of Italy and the ports of the eastern Mediterranean’.703 Both the construction and development of Portus at the mouth of the Tiber, and the eruption of Vesuvius in 79, negatively

4.3.3 Phase 3: The Claudian-Neronian Period During phase 3, the evidence suggests that the geographical distribution of ITS in the eastern Mediterranean remained intact, yet began to diminish quantitatively; the earlier part of this phase, however, still falls within its peak. Notable exceptions are: the Aegean-Southwest,709 in particular Corinth where quantities remain roughly as common as during phase 2; Crete, where quantities remain comparatively high, though Figure 54 and 3.2.2 indicate that the island now received smaller quantities of ITS; Cyprus, where ITS continues at the same level as phase 2, and is now also found at more sites. Last, considerable quantities can still be observed at Tarsos, Beirut and Petra. The distribution of ESA now resembles that of phase 1a, and it no longer played a significant role west of Cyprus. In some eastern regions, Asia Minor-Cilicia, Cyprus, to a lesser degree the Levant-Coastal North and -Interior South, quantities actually increased. That this occurs in several eastern regions (and not at all sites) could be significant. The basic distribution of ESC remained unchanged, though quantities diminished slightly: it is most common in the Aegean-East and -North, and small quantities trickled

Observed by Zabehlicky-Scheffenegger 2004, 78-79. Kenrick 1985, 494. 699  Kadous 1988, 141-142, 154-155; Lund observes (2003, 130, n. 75) that since ESB was most (or only?) common at Myos Hormos and Berenike roughly during the first half of the 1st century, it supports Kadous’ picture. 700  See e.g. Slane 1996b. 701  Wickham 1988, 191-193. 702  Seneca, letter LXXVII, in Rickman 1980, 70-71. 703  Jones 2006, 23. Wax tablets found at Pompeii attest to a family of bankers from Puteoli. 697 

Keay 2012, 41. Slane 2004, 40-41. Badre 1991, 272, 304, figs 20, 33. 707  Rutten 2007. 708  Post-early imperial pottery was found along these routes: Tomber 2008, 161-170, table 3. For Muziris (modern Pattanam): Shajan et al. 2008; Tomber 2007b, 2008, e.g. 154-161, table 2. A larger assemblage of terra sigillata was recovered at Muziris in 2011. I am most grateful to Roberta Tomber for sharing this information. 709  In Athens, the peak spanned c. 10-50; Hayes 2008a, 43.

698 

704  705  706 

85

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:38 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East into the Aegean-Southwest and Crete. Elsewhere it is only attested at Jerusalem and Paphos.

If tablewares are considered to be a proxy of the frequency and intensity of exchange, this implies that after the early 1st century, exchange within the east, as well as between west and east, started to dwindle. As the quantitative distribution of other commodities, amphorae and so on, falls beyond the scope of this study, any conclusive answer can simply not be given. In fact, decline need not have been the case at all. Changes in tastes and fashions, or the production infrastructure, are but a few scenarios one can think of; presumably it was a multitude of factors that underlay these developments.

The evidence indicates that only ESB and ESD increased, the latter following a minor dip around the mid-1st century. Their regional distribution reveals some interesting features. ESB grew continuously from the late 1st century BC onwards, yet the mid-1st century marks a turning point: it increases (sometimes marginally) in (most) regions except for the Aegean-East, where it gradually dwindled from 20-30 onwards, perhaps reflecting the shift in workshops.710 Finally, ESD also became more widely available, though the outlines of its distribution remained similar to those for the previous phases.711 Interestingly, this increase seems restricted to Cyprus and the western part of Asia Minor-Cilicia, and also here it is restricted to a few sites: Anemorion in Cilicia Tracheia, and Paphos, Palaipaphos712 and Panayia Ematousa713 on Cyprus. Thus, phase 3 seems to mark a change regarding western Cyprus:714 although the evidence is rather equivocal, it probably is transitory to phase 4, during which ESD became the most common tableware also on eastern Cyprus, at any rate at Panayia Ematousa, ‘even if the figure declined in absolute terms’.715 Elsewhere, quantities of ESD remained constant or slightly decreased, as in the southern Levant. To sum up: the role of ITS and ESA diminished, that of ESC changed only marginally, and that of both ESB and ESD grew in importance.

At least one aspect appears to have contributed, thus drawing attention to a major lacuna in our knowledge concerning the function and use of tablewares. The appearance of ITS is thought to have influenced, perhaps even prompted, directly or indirectly the choice of shapes the manufactories of (especially) ESA and ESB were making. Notwithstanding the possible socio-cultural implications, a similar model could have influenced manufacturers of more localised importance, both new and existing ones alike. Regrettably, locally-made terra sigillata continues to largely remain terra incognita, due to the paucity of archaeological evidence. Moreover, claims of local production are seldom corroborated through archaeometrical analyses, both of raw clays and sherds, and/or archaeological evidence, even if the recognition that tablewares do not belong to any major category is in itself significant.

What caused these changes? After the first half of the 1st century the total quantity of these five major categories of terra sigillata began to diminish (Figures 9 and 48). This supports the observations above, and suggests that the intensity of distribution started to shrink. Whether or not this is indicative of a decrease in exchange more generally cannot be answered at present; as a matter of fact, it need not necessarily reflect a poorer economic climate. At the same time, however, this can be seen as the very beginnings of what Lund observed for the 3rd and early 4th century.716 ITS and especially ESA decrease noticeably; though their quantities were probably of a different order, ESB and ESD fared fairly better. The evidence indicates that ESD replaced ESA, for example at Anemorion, as the major imported red slip tableware. ESB in turn became somewhat more common at Aegean sites. In general, however, ESB and ESD did not fully take the opportunity to fill the gaps left by ITS and ESA for good-quality terra sigillata. Perhaps the workshops of ESB and ESD were approaching levels of maximum output (the case perhaps with the seemingly peculiar distribution pattern of ESB from the mid-1st century onwards (cf. supra), or were playing safe economically.

At Knossos, locally-made tablewares are more common during the 1st century, comprising c. 35-36% during the Claudian and Neronian periods. At Corinth, by the late Flavian and Hadrianic period no less than 92% of the pottery is regarded as local or regional. Even if this pertains to phase 4 (cf. infra), locally-made tablewares also played a certain role in earlier phases. Fragments of unidentified categories of late Hellenistic to late Roman terra sigillata and red slip ware from Tanagra account for something in the order of 40%, whereas the percentage of locally-made tablewares (most of which hardly classifies as red slip tableware strictu sensu) at Koroneia, also in Boeotia, is c. 95%.717 The evidence for locally- or regionally-made tablewares is poorly understood, hampering a clear understanding of their scale and local/regional impact. Unfortunately, local or regional production tends to be viewed sometimes as a sign of impoverishment, the population lacking the capacity to afford the ‘luxury’ of imported terra sigillata. Conversely, local production can equally signify an economically healthy and enterprising community. Obviously, similar to small(er)-scale manufacture, ESA also catered for a local/regional population, albeit that the manufacture (and distribution) of ESA operated on probably the largest scale conceivable. In brief, the infrastructural and socio-

Hayes 2008a, 32. Lund 1997, 205, 207. 712  Lund 1993, 101. 713  Lund 2006a, 217. 714  Also observed for Kition, through which Panayia Ematousa, located only a few km inland, was most probably supplied (Marquié 2002, 291292, 294-295). 715  Lund 2006b, 36-37, 217. 716  Lund 1992. 710  711 

Pers. obs. This shows the poor knowledge of such wares yet also their (local/regional) significance. Local manufacture at Koroneia is archaeologically attested through wasters, kiln fragments, moulds, potters tools, etc.

717 

86

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:38 UTC All use subject to https://about.jstor.org/terms

Framing Late Hellenistic and Early Roman Tablewares economic background of tableware manufacture in the Roman east is poorly understood, and deserves much more detailed attention.

was noted at several sites elsewhere in the Aegean, AfricaCyrenaica and Crete. Though considerable quantities were recognised at Paphos, further east ITS appeared more rarely in the Levant and Asia Minor-Cilicia.

Nonetheless, the question to what extent the major lines of communication are laid bare through the collected evidence for this phase can be less easily answered than for preceding and following phases. Basically, the outlines that were sketched for phase 2 seem to continue. The AegeanSouthwest, Crete and the site of Berenice still received the largest quantities of ITS, indicative of continuing contacts with Italy; ITS identified further east must have travelled through this zone. The manufactories at Arezzo remained of importance given the distribution of OCK514 (30-70); it is common at Corinth, whilst its presence at Berenice could point to a different supply route. ESB, as was shown, began to play a more prominent role in the Aegean and Africa-Cyrenaica, whereas ESC remained mostly of regional importance as the Aegean-North indicates. That ITS, ESB as well as occasionally ESC are attested in eastern regions, nurtures the idea that these wares travelled along the same lines of exchange. Although the evidence for Africa-Egypt/Alexandria now appears insufficient to argue for a strong Aegean link, the diversified picture for Cyprus and the southern Levant suggests a south/ southwesterly supply route (cf. Alexandria), rather than one from the north. More generally, ESA and ESD remained the most widely traded categories of terra sigillata in the eastern regions, though the overall evidence for ESA points to a contraction of its geographical and quantitative distribution. The distribution of ESD suggests that contacts between Cyprus and its closest zones persisted, although these zones present a rather varied picture: ESD became more common on Cyprus in general, yet is now no longer attested in the northern Levant (except for Beirut), and diminished slightly in the southern Levant.

ESA remained common in Cyprus, but above all in Asia Minor-Cilicia and the Levant; smaller quantities were recognised at Alexandria and Berenice. Yet, ESA now arrived but sporadically on Crete and in the Aegean: single or at most a handful of fragments were identified at e.g. Gortyna, Knossos, Corinth, and Ephesos; at Athens, ESA made way for Aegean products, particularly the ‘numerous examples’ of ESC.722 Although ARSW, a new ware, did not reach its first major peak until the second half of the 4th century (cf. infra, chapters 5-6), early specimens have been identified in the east (Figure 55): the highest concentration is found at Berenice,723 and small numbers are attested at for instance Corinth, Tanagra, Kythera, Gortyna, Knossos, Ephesos, Perge, Paphos, Anemorion, Antioch, Caesarea Maritima, Petra and Koroneia, and now also at Athens, Delphi and Argos.724 Developments during this phase are not matched easily. For one, did a relationship exist between the demise of ITS and ESA on the one hand, and the increase of ESB, ESC and ESD on the other? It is tempting to consider that those involved in the production and distribution of ESB, ESC and ESD noted, and responded to, the ongoing withdrawal of ITS and ESA. Although this could be one explanation for the dwindling quantities of ESB only in Ephesos, catering for the home market would appear to be a no- or low level-risk business. The evidence, however, supports the idea that ESB, ESC and ESD partly filled the gaps left by ITS and ESA. Concerning the underlying mechanisms of patterns of distribution, the evidence now permits us to propose certain patterns with greater clarity. The distribution of stamps on ITS provides a first clue. Those stamps included in the database can be supplemented by the OCK, especially for Athens and Corinth.725 At Berenice and Sabratha all stamps (n=15) are Pisan, notably OCK1212 and 1690.726 At Carthage (n=60) the majority is Pisan, with some from Arezzo; also here OCK1212 and 1690 are well represented, and OCK1342 also comprises a significant share. ARSW was well on its way to monopolising the tableware market in Africa Proconsularis, surely affecting the import of ITS negatively. That ARSW was also quite common at Berenice during the later 1st and 2nd century could hint at redistribution through Ostia-Portus together with (Pisan) ITS, perhaps as part of Rome’s developing reliance on agricultural produce from Africa. Certainly,

4.3.4 Phase 4: The Late 1st and 2nd Century The second half of the 2nd century marks the end of the distribution of most late Hellenistic and early imperial categories of terra sigillata; production and distribution either stopped718 or continued on a reduced scale.719 Prior to this, however, the distribution of ESB, ESC and ESD720 matured during this phase, attaining their widest distribution during the late 1st and 2nd century, although ESC remained widespread during the larger part of the 3rd century (cf. infra) (Figures 48 and 55). ITS and ESA further developed along the lines sketched for phases 2 and 3: ITS remained fairly common in the Aegean-Southwest,721 and Lund (1997, 210) considers that an earthquake caused the end of production of ESD. Lund (2003, 131-132), however, rightly draws attention to the possible effects of local and wider developments for ESB. 719  Poblome 2006a, 191-192, 198-200; see ns 27, 126 for 3rd-century ESA and ESB. 720  Form P42, dated 175-300 [attested at Amathous (n=1), Paphos (n=4), Samaria-Sebaste (n=1) and Tarsos (n=1)], should be attributed to Phase 5b and not, as erronously happened, to Phase 4. The line and pie charts in question are changed accordingly, though the relevant distribution maps, unfortunately, not. 721  Though note Hayes (2008a, 43-44): ‘[t]he later Italian products were 718 

largely eclipsed in Athens (in contrast to sites such as Corinth and Patras) by the ESB and Çandarlı series’. 722  Hayes 2008a, 20, 30, 52. 723  Kenrick 1985, 493-494. 724  Athens: Hayes 2008a, 71, 73-74; Delphi: Pétridis 2010, 126-127, pl. 41, fig. 220; Argos: Abadie-Reynal 2007, 170-171. 725  OCK 2000. 726  Two stamps published by Martelli (1976) also concern OCK1690, from Pisa.

87

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:38 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East since Carthage continued to receive ITS for some time, a direct route along the North African coast that supplied Berenice with ARSW and ITS directly cannot be excluded. In the Aegean-Southwest, however, the pattern is very much different. At Corinth (n=170), where ITS continued to arrive into the 2nd century, most stamps are in fact not Pisan, but from Central Italy. This is also the case at Argos (n=7), Athens (n=13), Delphi (n=4), Olympia (n=61) and Patras (n=14). Through what channels these sites received these vessels is largely conjectural, though redistribution via Corinth in the case of Argos is plausible.727 As for ITS on Crete, two late Pisan stamps are known from Knossos, as are four out of the six late stamps now recorded for Gortyna.728 Although this favours Cyrenaica as a source, Pisan stamps have also been found at Corinth; considering the quantity this must remain hypothetical.729 The opening and development of the harbours at Portus and Ancona now provided Italy with two major foci for maritime exchange; possibly each played an important role in different lines of exchange.730 Therefore, we can envisage at least two routes, one that connected Italy with AfricaCyrenaica (directly or indirectly), the second linking Italy with Corinth and the Aegean(-Southwest) more broadly.

to understand the workings of regional and supraregional exchange patterns, Lund has recently investigated more fully the import and export of goods and commodities to and from Cyprus.734 This apparent floruit of ESD may have occurred in the wake of two types of amphorae defined by Hayes as types I and III that (chiefly) belonged to the late 1st and 2nd century. Both types have been attested on Crete and at Marina el-Alamein;735 type III, with which ESD thus may have travelled, seems to have been rather common at the latter site. On Crete, however, ESD was more common prior to phase 4. ESA still largely dominated the Levant (for instance at Paphos, Anemorion and Alexandria), but it was no longer as plentiful as it had been, notably at certain sites in the southern Levant where local products may now have supplanted ESA.736 Although ESA is attested at inland sites in Asia Minor-East Central and the Levant-Interior North prior to phase 4, the evidence at hand shows a slight increase especially along the Euphrates. Several of these sites had a military character at the time, and whether or not the arrival of ESA (and surely other goods) was the result of their status, it at least shows new directions in the distribution of ESA.

The evidence for other tablewares may help. Both ESB and ESC reached their peak during this phase (ESC continued into the 3rd century), and continued to reach Berenice,731 possibly via Crete. Both the (relative) scarcity of late ITS at Athens,732 Knossos and Gortyna and the increasing impact of ESB and ESC in the Aegean finds support in the evidence for the Aegean-North and -East, regions now mostly devoid of ITS. In fact, quantities of ITS may never have been sufficient to supply considerable parts of the east, and the dwindling quantities again reveal the major lines of exchange, just as during the earlier part of phase 2 (Figure 53). On a similar note, early ARSW on Crete and in the Aegean (and rarer pieces further east) could have travelled along these major exchange lines as well.

It is likely that Alexandria continued to play an important redistributive role, yet we have to deem the tableware evidence insufficient proof. Tablewares for this phase mostly comprise ESA and ESD. The opening of the porphyry and other stone quarries at Mons Porphyrites and Mons Claudianus added to Alexandria’s significance as a major outlet (though the quantity and repertoire of imported pottery is limited, also chronologically: both sites were supplied predominantly by Egyptian sources),737 and the supposedly important grain supply between Italy and Egypt/Alexandria might have resulted in a more varied picture of imported tablewares; the present evidence, however, does not corroborate this. Flourishing regional productions possibly catered for a substantial share of Egyptian households,738 whilst international tablewares merely touched Egypt’s coastal zones, mostly Alexandria. Marina el-Alamein may have received ESD directly — common (also) during this phase — alongside Cypriot amphorae.739

Wares other than ESA and ESD are only of minor importance in the east, except at Paphos, though ITS, ESB, ESC and ARSW are also attested at for example Anemorion and Caesarea Maritima. ESA and ESD still largely dominated distribution patterns further east, though ESA continued to decrease in quantity. ESD on the other hand attained its widest distribution, both quantitatively and geographically,733 strongly reflected in especially Cilicia Tracheia, a pattern that already existed before. In order

Whereas the maps for phases 2 and 3 illustrate a varied picture, as did the Aegean-Southwest, Crete, and the sites of Berenice, Paphos and to a lesser extent Perge for phase 4, most categories of tablewares are now characterised by a more regional fingerprint.

Abadie-Reynal 2004, 64-65, 2007, 63 (also on lamps and amphorae); Bes and Poblome 2006, 156-157. Baldwin Bowsky 2011, 120, table 8.1a, nos 29, 37 for Knossos, no new, late stamps are listed. For Gortyna, see Baldwin Bowsky 2011, 121, table 8.1b, nos 47-49, 51. 729  See Baldwin Bowsky 2011 for a fuller discussion, leaving open the possibilities of different routes of supply with regard to ITS. 730  Keay 2012; Slane 1989, 224, 2008b, 238; perhaps the earthquake thought to have struck Corinth between 70-77 temporarily disrupted supply patterns (Abadie-Reynal 2004, 65). 731  Small quantities of ESB already reached Berenice during phases 2 and 3. 732  Hayes 2008a, 43-44. 733  Lund 1997, 205, 207. 727 

Although tablewares still bridged distances that spanned the entire eastern Mediterranean, the evidence is not

728 

Lund 2006b. Daszewski 1990, 46-48, fig. 12.4. 736  E.g. Jerusalem, see Magness 2005. 737  Tomber 2007a, 179. Also see Tomber 2001. 738  Élaigne 1998; thin-walled wares were manufactured during Tiberian and Claudian times, e.g. at Assouan. 739  Daszewski 1990, 44; Daszewski 1995, 31-32. 734  735 

88

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:38 UTC All use subject to https://about.jstor.org/terms

Framing Late Hellenistic and Early Roman Tablewares plentiful. The role of Alexandria as a major in- and outlet undoubtedly continued for goods that include grain, papyrus, porphyry and other stones from the Eastern Desert. The collected evidence does not allow us to go any deeper into this matter, though the evidence from Paphos remains compelling. Figures 52 and 54 show that, next to ESD, small quantities of ITS, ESA, ESB, ESC and ARSW arrived at Paphos (the latter two rather negligible), probably reflecting that ‘l’ouest de île est plus ouvert aux influences occidentales et égéennes, tandis que la partie orientale est davantage marquée par les importations levantines’.740 The stamps on ITS provide further colourful and convincing evidence. Although small in quantity, the rarity of stamps in the east more generally confirms the fortunate position of Paphos in the wider scheme of exchange patterns.741 All stamps from Paphos (n=27) attributed to this phase are Pisan of the tardo-italico group,742 as such essentially conforming to Berenice rather than Corinth, which implies

a connection with Cyrenaica. Whether or not this was (partly) direct cannot be answered; again Alexandria may be a missing link. Ships on a (return) journey to Egypt called at Berenice, continuing eastwards to Alexandria, from whence goods could be redistributed to, amongst other cities, Paphos. This possibly also explains the small but significant quantity of ARSW at Paphos. Some ESA identified in the Pylos survey in the western Peloponnese belongs here, and it has been suggested that these came via Alexandria along with or in the wake of the grain trade.743 ESC became prolific in the Aegean in the course of the late 1st and 2nd century, when ARSW made its first modest appearance. The distribution of tablewares further changed after the 2nd century. In fact, the second half of the 2nd century rang in the rather different distribution patterns of tablewares during the 3rd century and after. The 3rd to 7th centuries will be covered in chapters 5 and 6.

Marquié 2002, 294. That western and eastern Cyprus belonged to different patterns of exchange has been suggested before, for instance by Lund. Marquié also points out that Kition was increasingly supplied with ESD by the later 1st and 2nd century, and at Panayia Ematousa, a few km inland from Kition, ESD became of growing importance from phase 3 onwards. 741  Lund 1997, 203. 742  Malfitana 2004, 111-113. 740 

743 

89

Berlin and Heath 2005, 201.

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:38 UTC All use subject to https://about.jstor.org/terms

Archaeopress

Chapter Title: The Mid-and Late Roman Periods: Discussion and Chronological Phasing Book Title: Once upon a Time in the East Book Subtitle: The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East Book Author(s): Philip Bes Published by: Archaeopress. (2015) Stable URL: https://www.jstor.org/stable/j.ctvr43kch.10 JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected]. Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at https://about.jstor.org/terms

Archaeopress is collaborating with JSTOR to digitize, preserve and extend access to Once upon a Time in the East

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:39 UTC All use subject to https://about.jstor.org/terms

Chapter 5

The Mid- and Late Roman Periods: Discussion and Chronological Phasing ARSW experienced only marginal growth during the 3rd and first half of the 4th century, yet witnessed a period of strong growth after the mid-4th century, which lasted into the late 4th century, after which ARSW declined considerably, and from the second quarter of the 5th century it continued at a lower level but with no major fluctuations. By the late 5th and early 6th century, however, quantities dropped only to rise again towards the mid-6th century. Roughly by the third quarter of the 6th century we observe another drop, followed by a final peak in the decades around 600. After that, ARSW slowly but surely dwindled to eventually disappear from Mediterranean exchange.

Introduction Combining the mid- and late Roman periods allows us to take into account the evidence for ARSW, a Mediterraneanwide and chronologically continuous class of red slip tableware. ARSW is not eastern in origin, and it thus forms a ceramological bridge between west and east. ARSW became increasingly important in the east from the early 3rd century onwards, which makes it possible to study its development by itself and in relation to contemporary wares. As a result it can contribute to a better understanding of tableware distribution more generally. A second reason to draw attention to ARSW is that since the publication of LRP and A Supplement to LRP, a number of adjustments and alternatives to the typo-chronology have been published.744 The most recent is Bonifay’s Études, which moves beyond the typo-chronological framework of LRP, yet serves as ‘un simple index’ rather than an independent classification.745 Conversely, it includes a selection of the forms found in LRP, the effects of which have been explored elsewhere.746 Still, a wider use of Bonifay’s Études, particularly for the 6th and 7th centuries, holds the potential to evaluate and reinterpret the (last phase of) supraregional distribution of ARSW in the east.

LRC joined ARSW by the mid-4th century, and after a phase of gradual and uninterrupted development this changed into a sharp growth around the mid-5th century, to reach a peak in the early 6th century. An equally steep and somewhat puzzling decline is noted, reducing it to a more modest contributor to the tableware market by the mid-6th century. However, LRC quickly increased again and reached a second, albeit lower, peak in the early 7th century, which suggests a sudden and sharp reduction of its overall role — which in fact appears geographically more specified — in the east. Like ARSW, LRC disappeared from the eastern Mediterranean during the later 7th century.

5.1 The Collected Published Evidence

The start of the distribution of LRD is set in the late 4th century, yet was restricted to one form (Hayes form 1) till the mid-5th century, and remained quantitatively rather modest. A first period of growth came during the second half of the 5th century, to reach a first peak in the early 6th century. This was followed by a steep decrease toward the mid-6th century, followed by a strong increase during the second half of the 6th century, reaching a second, higher peak around 600; a similar pattern characterises LRC between the mid- to late 6th century. LRD remained rather common during most of the 7th century, yet also gradually dwindled, even if manufacture and a restricted distribution may have continued into the early 8th century.750

This section briefly evaluates the collected evidence. Four wares are distinguished: ESC, ARSW, LRD and LRC; as ESD form P42 continues into the 3rd century it is included here. Figure 58 captures the absolute quantitative development of these five wares, from the early 3rd to the late 7th century. Notwithstanding the importance of ARSW in central North Africa (e.g. Sabratha),747 the evidence from Berenice presents different developments when compared to the eastern Mediterranean more generally.748 ESC witnessed its last phase of distribution during the 3rd century ― the Aegean remained its main zone of distribution ― and faded from the tableware market after the early 4th century (Figure 58).749 ESD is represented by a negligible quantity during the 3rd century.

These observations serve as a yardstick for further study, by comparing regional patterns with the more general, in order to understand to what extent these observations comply with the existing literature so as to come to a more balanced appreciation of the ensuing discussion of the regional evidence.

744  See e.g. Hayes 1976a, 1977; Fulford 1984; Tomber 1988; summaries in Abadie-Reynal 2005a; Sieler 2004. 745  Bonifay 2004, 87: ‘[l]a nomenclature proposée en filigrane pour chaque grande catégorie de mobilier, […] n’est pas une classification mais un simple index destiné à faciliter la liaison entre le texte et les figures.’. 746  Bes and Poblome 2009. 747  Fulford and Tomber 1994. 748  Timby 1994; Kenrick 1985. 749  Hayes 1972, 316-317, 417-418.

5.2 Comparing the Collected Evidence with Key Literature Here, the evidence for ARSW, LRD and LRC as discussed above will be evaluated against the key literature available. 750 

90

Armstrong 2007.

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:39 UTC All use subject to https://about.jstor.org/terms

The Mid- and Late Roman Periods: Discussion and Chronological Phasing

Figure 58. Line chart showing the development, in absolute numbers, of ESC, ESD, ARSW, LRD and LRC during the mid- to late Roman period (n=6982) (© Philip Bes/ICRATES Project).

half of the 5th century, apparently prior to the arrival of Vandal tribes in 439, when a decreasing volume is coupled with ‘contracting distribution’.757 Hayes, on the other hand, sees a probable ’connection’ between the start of the Vandal occupation and the contracting distribution of ARSW in the east.758 Bonifay wonders whether the cause(s) for this decrease should be sought in more general economic activities, or the competition posed by LRD and LRC (cf. infra).

ARSW ARSW is, in McCormick’s words, ‘far and away the beststudied ceramic from late antiquity’.751 Bonifay’s Études — and subsequent publications — is the most in-depth study of Tunisian pottery presently available, framing the production, typology and distribution of pottery during the mid- and especially late Roman period.752 In the process he has drawn detailed attention to both artisanal and economic matters, thereby pointing out lacunae and possible misconceptions persistent in research, such as the traditionally held view that the production of a surplus of olive oil formed Africa’s main economic motor.753

The distribution of ARSW increased again towards the late 5th and early 6th century — though see Figure 58;759 Bonifay emphasises that this process was already underway prior to 533, the year in which Justinian reconquered North Africa.760 Hayes points out that a ‘resurgence’ only took place in the mid-6th century, thus after Justinian’s campaigns, yet observes that conditions between Rome and Vandal Africa may well have allowed ARSW and other goods to ‘circulate more freely’ from the late 5th century onward.761 ARSW, and in particular LRC, were both common in the east around the mid-6th century.762 Despite the fact that the continued production

ARSW grew in geographical and quantitative importance during the 3rd century, although western Greece and Cyrenaica were already witnessing ‘substantial importation’ in the early 2nd century.754 This developed further during the second half of the 4th (and the early 5th century?), a period when the distribution of ARSW can be characterised as a Mediterranean-wide koinè.755 Although Egypt may have been an exception, ARSW was much less common in the east roughly during the second half of the 5th century [and into the early 6th century].756 McCormick notes that this development had already started in the first

Hayes 1972, 417-420; McCormick 2002, 55; Fulford 1984, 113. Hayes 1972, 423. 759  Note the difference for the late 5th century between Figure 58 and the line chart published in Bes and Poblome 2009, 84, fig. 3, which is based on a slightly smaller data set. As such, the collected data and published literature on the matter diverge here. 760  Bonifay 2003, 123, 2004, 482, 2005, 569-570; McCormick 2002, 55; Fulford 1984, 114. 761  Hayes 1972, 418-420, 423, specifically stating the ‘resurgence’ of ARSW for ‘Greece and the Aegean’, ‘South Turkey and Syria’ and ‘Palestine’, see also 426. 762  Hayes 1972, 423. 757  758 

McCormick 2002, 53-54. Bonifay 2002, 2003, 2004 (esp. 445-486), 2005; Bonifay et al. 2012. 753  Bonifay 2003, 128. 754  Hayes 1972, 417, 421; Bonifay 2005, 566, 568. 755  Hayes 1972, 417-423; Bonifay 2005, 568; Sodini 2000, 181; Hayes 2001, 279; McCormick 2002, 55; Fulford 1984, 112. 756  Hayes 1972, 421, 423; Bonifay 2003, 122, 2004, 480-481, 2005, 568569; Sodini 2000, 181; McCormick 2002, 55; Fulford 1984, 113; AbadieReynal 1989a, 144-145. 751  752 

91

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:39 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East and distribution of LRD and LRC in the east still posed ‘competition’ to ARSW, Bonifay characterises the late 6th and early 7th century as a period of ‘relatif dynamisme économique’ (Figure 58).763 Perhaps this resulted from a renewed grain supply to Constantinople and elsewhere, in addition to which Bonifay draws attention to the wide distribution of African spatheia during this period.764 Hayes noted that ERSW and LRD may have (largely) replaced ARSW and LRC in Syria, Palestine and Egypt by the early 7th century, whilst the distribution of ARSW and LRC was maintained in the Aegean and Cyrenaica.765 McCormick sees ‘significant quantities’ arrive in the east early in the 6th century, with its distribution continuing ‘for the rest of its production life’. However, both the second peak in Figure 58 and Bonifay’s ‘relatif dynamisme économique’ of c. 575-625 do not recur in McCormick’s discussion, which also seems to diverge from observations made by Hayes. Instead, McCormick says that ‘the long, drawnout contraction in volume and geographic distribution that follows [i.e. the early 6th century] ends in a whisper in the decades after 650’ during which ARSW ’rarely moved inland’.766 Yet the most common of Hayes’ latest forms (e.g. 105, 107, 109) appear in modest quantities both at coastal and inland sites in the Levant, some of which appear to have had a military or administrative function.767 The latest LRC form, 10C (c. 600-650/675), is also not uncommon in the Levant and on Cyprus.

constant level following the Vandal conquest, and some time prior to the reconquest began to increase again. It is tempting to infer that the conquest did impinge on the economic framework of Africa, but that other factors had a (far) greater effect prior to the conquest. One line of thought — admittedly from a tableware perspective alone — is that the manufacturies of LRC and LRD (as well as perhaps other, smaller, categories) began to blossom. LRD Sodini pays virtually no attention to LRD, yet pointed out that it has been under-represented in discussions concerning the late Roman economy.771 Hayes’ LRP remained the basic study concerning LRD for a long time, though recently LRD has received considerable attention,772 including the chronological gap between ESD and LRD.773 The early stage in the distribution of LRD shows it to have been in its infancy.774 However, after the distribution of ARSW dwindled from the early 5th century onwards, LRD may have partly filled that gap, dominating on Cyprus, the area of Alexandria and southern Asia Minor.775 Finds from, for instance, Constantinople and Emporio suggest that LRD even appears to have ‘enjoyed a greater vogue than ever before’ in the first half of the 7th century.776 Shortly after LRP appeared, Rodziewicz discussed to some extent the LRD found at Kôm el-Dikka.777 Based on preliminary results from excavations at Paphos, Rowe claimed that ‘[p]reliminary study of the ceramics from Area Three, particularly the Cypriot red Slip Ware, suggests that the commonly accepted date range for some of Hayes types cannot be sustained’.778 Her conclusions, however, have received fundamental criticism.779

Both the outlines of the quantitative and geographical distribution of ARSW and the typological change of both amphorae and tablewares do not (always?) appear to run parallel chronologically to political and historical events.768 Though the quantitative and geographical changes seem hard to deny, there is little consensus concerning the (relative) impact of the Vandal, and later Justinianic, conquest on the production and circulation of Tunisian pottery (including ARSW), nor the chronological demarcation of these quantitative changes. If indeed — as Figure 58 suggests — the quantitative distribution of ARSW started to decrease already prior to the Vandal conquest, the latter’s impact on the artisanal and economic framework of Africa may have been less profound than once thought, nor would it have been the only disruptive force.769 The fact that the distribution of ARSW was triggered again prior to Justinian’s reconquest is equally noteworthy. This is also suggested by Hayes, though the profound (final) change still came towards the mid-6th century, if not thereafter.770 The collected data does show that the distribution of ARSW remained at a roughly

LRC Specimens of the early forms may have been more common in the Aegean compared to the east in general.780 According to Hayes, the leading phase of LRC in the east spans the second half of the 5th and early 6th century. According to Sodini, LRC reached its ‘apogée’ around the second quarter of the 5th century, a ‘déclin progressif’ nonetheless comprised Hayes form 10 (c. 575-650/675).781 Hayes further noted that several regions differ from the general trends.782 At Alexandria, for example, quantities increased during the 6th century, yet by the end of that

Sodini 2000, 182. Meyza 2007: Poblome and Fırat 2011. 773  Meyza 1995, 179-202; Meyza 2007, 17-18, 43-81. 774  Meyza 2007, 101-102, map 5. 775  Hayes 1972, 423; Meyza 2007, 102-103. 776  Hayes 1972, 424; Meyza 2007, 103-104; Constantinople: Hayes 1992; Emporio: Boardman 1989. 777  Rodziewicz 1976, 42-47. 778  Rowe 1999, 276-277. 779  Lund 2006c, 215, n. 22. See, however, Armstrong 2007. 780  Hayes 1972, 417. 781  Sodini 2000, 182. 782  Hayes 1972, 417-424, esp. 423: ‘with the exception of Cyprus, the Alexandria region and possibly some part of southern Asia Minor, where Cypriot products succeeded the African Wares’. 771  772 

Bonifay 2003, 123-124, 2005, 571-572; Sodini 2000, 181-182; to a lesser extent see Fulford 1984, 114. 764  Bonifay 2003, 127-128. 765  Hayes 1972, 424. 766  McCormick 2002, 55; Fulford 1984, 114. 767  See for example Harper 1980, 327-348; Harper 1995, 21-33 (for redating: Magness 1999, 189-206); Orssaud 1991, 260-275; AbadieReynal 2005b, 523-546. 768  Bonifay 2003, 121-122; McCormick 2002, 55-56; Fulford 1984, 256257. 769  See e.g. Rodziewicz 1976, esp. 29; Peacock et al. 1990, 83. 770  Hayes 1972, 414-424, esp. 423-424; compare map 10 (457) to maps 11-12 (458), echoed in Hayes 2008a, 73. 763 

92

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:39 UTC All use subject to https://about.jstor.org/terms

The Mid- and Late Roman Periods: Discussion and Chronological Phasing century ERSW-A and LRD dominated the market.783 Although ARSW regained some of its former position in the east by the mid-6th century, LRC remained widespread with distribution continuing during the late 6th and 7th century. Whilst ERSW and LRD may have ‘superseded’ ARSW and LRC in the Levant (where Hayes form 10A-C is nevertheless found in substantial quantities at a number of sites) and Egypt, the circulation of both wares continued in the Aegean and Cyrenaica, as many sites attest.784 Only recently has the typo-chronological framework and manufacturing process of LRC been the subject of renewed attention,785 though Hayes’ LRP remains the best available.786 5.3 Regional Developments In this section the mid- to late Roman regional evidence is evaluated, so as to examine to what extent each region compares with the general framework and its neighbouring region(s). Appendix 3b contains an overview of the relative quantities per ware, per region.

Figure 59. Africa-Cyrenaica: pie chart showing the proportions of the different wares (n=311) (© Philip Bes/ICRATES Project).

5.3.1 Africa-Cyrenaica the Sanctuary of Demeter and Persephone, which could explain the chronologically limited span of ARSW.789 The substantial collection of pottery from Apollonia represents tablewares that were found in the destruction debris of the so-called Palace of the Dux, and mostly belonged to the 6th and 7th centuries.790 The finds from Tocra were found during excavations in the Sanctuary of Demeter and Kore, which lost its religious function certainly by the late 4th or early 5th century (note the difference from Cyrene!). Foundation trenches and other signs of occupation of 6th-7th century date indicate that the site was (partially) occupied by domestic structures. Though some earlier finds were catalogued, this scenario also limits the chronological span of Tocra.791 Only at Berenice is a considerable part of Hayes’ typo-chronology represented. The ARSW, spanning a longer period, shows a particular pattern. The highest quantity is predominantly recorded around the mid-3rd century, and the rise in the second half of the 4th century cannot be observed here. In fact, only the second half of the 5th-early 6th century can be characterised as another phase of growth, yet in lesser numbers than in the 3rd century.792 LRC also shows a pattern slightly different from that generally seen elsewhere. At Berenice it also reached its peak by the early 6th century, later forms have been recorded yet in smaller quantities only; it seems as if ARSW had regained its position by now. ESC and LRD

Collected evidence from Africa-Cyrenaica comes exclusively from the Pentapolis, and besides Berenice, which remains one of the major publications for the central Mediterranean, material has been collected from Apollonia, Cyrene and Tocra.787 The proportions of the collected evidence (n=311) are shown in Figure 59. ARSW is the major tableware (c. 71%), which implies a mostly western orientation. LRC accounts for c. 25%. LRD represents c. 3.8%, which roughly complies with the evidence west of the line Egypt-Lycia (cf. infra). Hayes suspected that ESC was quite common, yet the evidence remains insufficient to corroborate this hypothesis:788 ESC accounts for c. 0.3%. Figure 60 captures the site-based proportions, and despite the fact that the material from Apollonia and Tocra is chronologically limited (cf. infra), their proportions present two interesting aspects. Leaving aside Cyrene, the uniformity of ARSW suggests that the inhabitants of Cyrenaica remained important consumers of ARSW. Second, it is interesting to note that Apollonia, the easternmost site here, received proportionally larger quantities of LRD. The three main wares present a chronological pattern that appears more strongly dependent on the nature of the site (Figure 61). The ARSW from Cyrene belonged to the 3rd and early 4th century. Kenrick considered that the pottery as a whole need not postdate 262, the closing of

Kenrick 1987, 11. Hayes 1976b [1978], 264-265, with little or no residual material. Hayes 1973b, 108. 792  Kenrick 1985, 351, 356-357. Kenrick emphasised that ARSW forms of the 3rd century are particularly common at Berenice, especially Hayes forms 31 and 50, and that these and other forms are large and thinwalled, thus liable to produce more sherds. Both forms also have long date ranges, which may cause an over-representation, yet this remains a predominantly theoretical matter. See also Hawthorne 1997, Fentress et al. 2004. 789  790  791 

Hayes 1972, 420-421, 423-424. Hayes 1972, 423-424. 785  Vaag 2003, 2005. 786  Hayes 1972, 323-370, 1980, 525-527, 2008. 787  Apollonia: Hayes 1976b [1978]; Berenice: Kenrick 1985; Cyrene: Kenrick 1987; Tocra: Hayes 1973b. 788  Hayes 1972, 421. Note that ESC, here, was restricted to the 3rd century. 783  784 

93

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:39 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East

Figure 60. Africa-Cyrenaica: map showing the ware proportions of the different sites (© Philip Bes/ICRATES Project).

Figure 61. Africa-Cyrenaica: line chart showing the development, in absolute numbers, of the different wares (n=249) (© Philip Bes/ICRATES Project).

94

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:39 UTC All use subject to https://about.jstor.org/terms

The Mid- and Late Roman Periods: Discussion and Chronological Phasing are both sparsely represented, the latter by forms of its entire period of production. 5.3.2 Crete The location of Crete at a crossroads between east and west, as well as the Aegean and North Africa, is reflected in the diversity and quantity of its pottery: its main urban centres, Knossos and Gortyna, are well published.793 Knossos, located close to the island’s north shore, is extensively published although the late Hellenistic to midRoman periods predominate. Gortyna has been extensively published since 1988, the amphorae and tablewares in particular testify to contacts round the Mediterranean. Furthermore, several extensive surveys shed important light on the distribution of imported tablewares into smaller urban and non-urban settlements.794 The recently published finds from Eleutherna, a city close to Crete’s central northern shore, are a welcome addition to the existing corpus of imported pottery of the late Roman period.795

Figure 62. Crete: pie chart showing the proportions of the different wares (n=805) (© Philip Bes/ICRATES Project).

A total of 805 fragments have been collected; Figure 62 seems to reflect the differences in published finds, especially concerning Knossos and Gortyna. Whereas the published tablewares from Gortyna comprise the entire Roman period, those from Knossos are predominantly late Hellenistic to mid-Roman in date. The quantity of ESC captures the ware’s continuing importance in the 3rd century, but since ARSW grows in importance during the 3rd century it may have affected the quantities of ESC reaching the island. LRC is the most common class (49.3%), being most common at Gortyna and in the Southwest Coast and Vrokastro surveys. ARSW comes second (c. 41%), ESC third (c. 8%), whilst LRD is represented by c. 1.7%.

the 3rd century, eventually becoming the major imported tableware towards the late 3rd century, and reaching a peak in the second half of the 4th century.797 ARSW is joined from about the late 4th century by LRC, which grew rather steeply and reached a peak in the early 6th century.798 ARSW declined in quantity during the entire 5th and early 6th century, yet — despite a little dip around the mid-6th century — started to increase again after the early 6th century and continued to do so, reaching a second peak around 600. LRC sharply decreased after the early 6th century, yet remained the dominant tableware into the early 7th century.799 In the course of the 7th century, when they were already occurring in much smaller numbers, both ARSW and LRC disappeared. The very small numbers of LRD, which reflects a general Aegean pattern, have virtually all been found at Gortyna. Identified forms, however, span the entire typological range, thus corresponding roughly to the overall chart.

Figure 63 illustrates the site-based evidence. ARSW dominated at Knossos — where as said the late Roman period is not well represented in publication, and the evidence is also somewhat influenced by prolific finds of ESC — and Kommos, though quantities at the latter are hardly sufficient to make realistic statements. LRC dominated elsewhere in varying degrees. LRD was generally rare: it is noted at Knossos, at Gortyna and in the survey south of Gortyna.

Site-specific literature substantiates the generalities for Crete. LRC was the dominant tableware from the late 6th to the early 7th century at Eleutherna (77-85%),800 similar to Gortyna, where LRC comprised between 65% and 85% of the imported tablewares during the same period.801 The evidence from Knossos appears to be less reliable for reasons enumerated above.802 That LRC comprised more than 75% of late Roman imported tablewares at Eleutherna is noteworthy in relation to Knossos, given the relative

When looked at in detail (Figure 64), ESC continued to be the major imported tableware during the 3rd century, and at Knossos it was practically the only tableware in deposits of the later 2nd century.796 ARSW steadily increased during Knossos: Hayes 1971, 1983; Sackett 1992; Gortyna: Dello Preite 1997, 2004; Lippolis 2001; Martin 1997a; Monacchi 1988; Rendini 2004; Rizzo 2004. For further data, also see e.g. De Tommaso 2001. 794  South coast: Blackman and Branigan 1975; Southwest coast: Hood 1967; Sphakia: Nixon et al. 1988, 1989, 1990; Vrokastro: Harrison and Hayden 2005. Only Hood 1967 and Harrison and Hayden 2005 are included in the data. 795  Yangaki 2005; Vogt 2004; also see finds from Itanos in northeastern Crete: Xanthopoulou 2004. These could not be studied in detail. 796  Hayes 1972, 417-418; Sackett 1992, 163, fig. 2; Forster 2001, 145. 793 

Hayes 1972, 417; Forster 2001, 147. Hayes 1972, 417-418. 799  As also indicated by Hayes 1972, 418. 800  Vogt 2004, 928. 801  Dello Preite 1984 [1988], tables 2-4, quoted in Vogt 2004, 928. 802  Forster (2001, 147, 149) does not discuss post-4th century ARSW; the ‘relatively small numbers’ of LRC found at Knossos are also not helpful for a quantitative approach. 797  798 

95

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:39 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East

Figure 63. Crete: map showing the ware proportions of the different sites and surveys (© Philip Bes/ICRATES Project).

Figure 64. Crete: line chart showing the development, in absolute numbers, of the different wares (n=598) (© Philip Bes/ICRATES Project).

96

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:39 UTC All use subject to https://about.jstor.org/terms

The Mid- and Late Roman Periods: Discussion and Chronological Phasing proximity of both sites. Clearly, more evidence is required to assess the pattern for Knossos.

5.3.3 Aegean-Southwest812 The Aegean-Southwest embraces the Peloponnese, central Greece and the Cyclades, which, during the late Hellenistic to mid-Roman periods formed the provincia Achaia. Two major urban centres, Corinth and Athens, have a long tradition of archaeological investigation.813 Additionally much archaeological work (both excavation and survey) has been carried out at a fair number of other urban and non-urban settlements (small(er) urban centres such as Eretria, and sanctuaries as Isthmia, Olympia and Delphi),814 as well as in rural areas (Pyrgouthi, Akra Sophia).815 As for Crete, this diversity captures tableware distribution beyond the major centres of consumption and (re)distribution: although larger and middle-sized urban centres in Boeotia, Attica and the Corinthia figure predominantly, archaeological and ceramic studies are geographically well-represented.

The different extensive surveys carried out on Crete not only present important evidence for the intra-island distribution of imported tablewares; also, as many of the settlements in the surveyed areas are non-urban or rural, they throw light on their use beyond the urban. Most of the regions where surveys were carried out present a ceramo-chronological gap in the 3rd, and to a lesser extent, the 4th century.803 The dearth in supraregional tablewares circulating in the eastern Mediterranean during the 3rd and early 4th century remains striking, as only ARSW and ESC are testimonies of supraregional exchange, although ARSW became common only at a later stage, and ESC was mainly an Aegean phenomenon. Causes for the relative scarcity of tablewares in supraregional exchange need not necessarily result from the political and military ‘crises’ of (especially) the mid-3rd century. Recent contributions point out that the evidence should be assessed on a regional basis, and that an explanation will result from archaeologically, geographically and thematically widening the debate.804

The proportions, based on a collected total of 1534 entries, are shown in Figure 65. ESC is the major distributed tableware during the 3rd century and still makes up a considerable proportion. ARSW comprises 67.4%, which is in marked contrast with Crete and especially with the other two Aegean regions (cf. infra). LRC accounts for nearly 27.5%, and, with less than 1%, LRD makes only sporadic appearances (in the Aegean), as already noted for Crete. These proportions, notably those of ARSW and LRC, also surface on the site level (Figure 66). In fact, the collected data allows us to make a twofold division: for Athens, Berbati, Sparta, Hyettos, Koroneia, Thespiae and Tanagra816 the percentage of LRC is about a quarter or less, whereas Argos,817 Corinth, Halieis, Kenchreai,

In the Akrotiri survey in western Crete, late Roman tablewares gained a wider significance compared to the early Roman period, as ‘[t]hese African and eastern imports also reached habitation sites of every size’.805 According to the author (therein following Hayes) late Roman tablewares were less costly to manufacture and intended ‘for mass shipment’, which may partly explain their wider distribution.806 Just over half (56%) of the 50 fragments belong to LRC, 42% to ARSW, whilst one sherd of LRD was identified.807 In the Vrokastro area on eastern Crete, a summary of the surface finds present a highly similar picture: c. 55% of the fragments (n=101) is attributed to LRC, the remainder to ARSW.808 Smaller quantities of imported late Roman tablewares have also been found in the Lasithi survey (ARSW, n=4; LRC, n=3).809 In a survey in the Mesara plain, besides early imperial tablewares ARSW (n=2), LRD (n=1) and LRC (n=7) were found.810 Finally, the Sphakia survey in southwestern Crete did not produce quantifiable information concerning imported tablewares. The pottery, studied by Hayes, indicated an absence of 3rd-4th century ARSW, though ARSW and LRC did arrive at a number of coastal and inland sites in the area during the 5th to 7th centuries.811

See now the two important volumes on mid- to late Roman pottery from Greece, mostly from the Aegean parts: Papanikola-Bakirtzis and Kousoulakou 2010. Both volumes, the second in particular, contain a wealth of data from both larger and smaller (urban) sites, including midto late Roman tablewares. Even if this new body of evidence does not affect the major trends put forward here, it will nevertheless do much to refine our ideas on the more regionalised patterns of distribution of ARSW, LRC, etc. 813  Athens: Waagé 1933; Robinson 1959; Corinth: Wiseman 1972; Williams II and Zervos 1982, 1983, 1985; Biers 1985; Slane 1990, 1994; Sanders 1999; Slane and Sanders 2005. Note now Slane 2008a, which contains additional material from the Sanctuary of Demeter and Kore at Corinth, but more importantly a chronological discussion with (probable) wider implications. 814  A small deposit (not yet unpublished), probably datable to the first half of the 3rd century, with a terminus ante quem of c. 275, is now known from Titanè, in the mountainous hinterland behind Corinth. This suggests that the Sanctuary of Asklepios at Titanè described by Pausanias was possibly still functioning at this time. The deposit contained an ARSW Hayes form 50A, a probably late ESB Hayes form 60, yet no ESC. I warmly thank Kris Tytgat, former director of the Dutch Institute in Athens, for allowing me to study this material. 815  Isthmia: Marty 1993; Gregory 1995; Hayes 1998a; now also Rife 2012, 99-102, for a small collection of fragments that are typologically mostly unidentified, and Gregory 1993, 113-114; Olympia: Martin 1995, 2001; Schauer 2001; Sinn et al. 1993; Delphi: Perdrizet 1908. For the rural dimension of tablewares, see e.g. Pyrgouthi: Hjohlman et al. 2005; Akra Sophia: Gregory 1985, as well as different surveys. The late Roman tableware landscape of, for example, Eretria remains poorly understood; evidence is now available from Delphi (Pétridis 2010), where local manufacture seems prolific. 816  Bes forthcoming a for Hyettos, Koroneia, Tanagra and Thespiae; Berbati: Forsell 1996; Sparta: Bailey 1993; Pickersgill and Roberts 2003. 817  Abadie-Reynal 1995; Aupert 1980, 1986; Oikonomou-Laniado 2003. 812 

Raab 2001, 43, yet noting the continued importation of tablewares at Knossos during the 3rd century. 804  Poblome 2006a; Lund 2006c. 805  Raab 2001, 150, fig. 69. 806  Raab 2001, 150, following Hayes 1997, 59. 807  Raab 2001, 150-151, table 3. 808  Harrison and Hayden 2005, 57-58 (only absolute numbers). 809  Vance Watrous 1982, 82-84. 810  Blackman and Branigan 1975, 31; also Vance Watrous et al. 1993. Also see Sanders 1976. 811  Nixon et al. 1989, 209-210, 1990, 217-218. 803 

97

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:39 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East remaining evidence for the Aegean-Southwest basically conforms to this division. Although numbers usually are small, finds from the Aetolia, Veloukovo and Attica surveys, and Olympia indicate that ARSW was more common.823 Thus, southeastern Greece links up better with Crete (cf. supra), where LRC attained between c. 49% and 54% at Gortyna and the Vrokastro survey respectively, and about three-quarters at Eleutherna. Even if this division is not (yet) clearly defined, the ratio ARSW:LRC roughly shifts in favour of the latter when moving from (north)west to (south)east.824 Figure 67 captures the chronological developments. ARSW gradually increased during the 3rd and early 4th century, when ESC withdrew from the Aegean-Southwest.825 ARSW grew strongly towards the mid-4th century, and reached a peak toward the late 4th century. ARSW decreased in the early 5th century, a process that continued into the early 6th century with the exception of a stable period right around the middle of that century.826 Even if LRC was not particularly common in its early phase, it grew steadily from the late 4th century onwards and continued to do so, reaching a peak around the late 5th, early 6th century when, interestingly, ARSW was at a quantitative low. During c. 500-525, ARSW increased substantially and remained constant till the late 6th century. LRC decreased during the first half of the 6th century, a process that halted by the middle of that century, then quantities increased again, reaching a second, yet more modest peak around 600. After around 600 both wares gradually declined and disappeared from the Aegean-Southwest. LRD is only sporadically attested; identified forms span the earlier and later phases of distribution (yet not Athens)827 — thus conforming to Crete — echoing Hayes’ observation that ‘small quantities of the ware occur throughout the Aegean region from early in the fifth century onwards, and stray pieces have turned up as far afield as Corfu’.828 Although these trends reflect the basic trends in the overall chart, there are obvious quantitative differences. Here, ARSW is much more prolific in the 4th century, and the high peak for LRC of around 600 is absent, and probably to be found elsewhere. Another difference concerns the 6th and 7th centuries. ARSW and LRC run roughly similar in the overall chart, yet ARSW clearly dominated the AegeanSouthwest from the mid-6th century onwards. This is in contrast to Hayes, who also places the change in the mid6th century but stresses that LRC ‘retains the bulk of the market’.829

Figure 65. Aegean-Southwest: pie chart showing the proportions of the different wares (n=1534) (© Philip Bes/ICRATES Project).

Kythera and the Methana and Keos surveys present higher percentages of LRC, between c. 40-61%.818 Pylos and Siphnos are at odds with c. 35% and 31% respectively.819 In consideration of its location, recent evidence from Antikythera (survey), Keos (excavation and survey) and Skyros (survey) fits the latter group very well, as LRC (Antikythera: 58.2%; Keos: 65% for the survey; Skyros: 60.5%) clearly outnumbers ARSW (Antikythera: 18.8%; Keos: 13% for the survey; Skyros: 17.4%),820 whilst recently published evidence from Delphi, where ARSW exerted its influence both directly and indirectly, and Achaïa suggests that these belong to the first group.821 To this can now be added data from a survey in the Corinthia, where LRC (n=122) dominates over ARSW (n=17).822 The See also, overlooked, Ivantchik 2002. 818  Halieis: Rudolph 1979; Kythera: Coldstream 1972; Methana: Mee and Forbes 1997; Keos: Cherry et al. 1991. 819  Berlin and Heath 2005; Brock 1949. 820  Antikythera: Quercia et al. 2011, 55-58; Keos: Zachos 2010, 785, 787, 791, figs 5, 7; Skyros: Karambinis and Bes forthcoming. 821  Pétridis 2007, 50, 2010, 126-129. Amongst small quantities of late Roman Gaulish, painted Attic and central Greek wares, single fragments of LRD, 3rd-century ESC and LRC (no Hayes form 10), ARSW is ‘les plus nombreuses parmi les sigillées’: it is commonly attested in 4thcentury deposits, and some of the latest forms were also identified. Pétridis (2007) focuses on certain morphological and decorative traits that Delphi, besides an important impetus provided by African pottery, shared with a number of regional (urban) centres in central Greece, reflecting both actual exchange as well as a material culture these centres partly had in common. Interestingly, amongst the locally-made Roman-period ceramic repertoire from Koroneia in Boeotia, a variety of morphological and decorative styles is recognised (amongst which are the typical gouged jugs well known from Athens) that can very well be associated with other manufacturing centres in central Greece, including Delphi, possibly extending as far west as Patras and southern Italy. I warmly thank Carlo di Mitri for sharing illustrations of some of the tableware finds from Brindisi; Achaïa: Bes 2010 (ARSW n=37; LRC n=2). 822  Pettegrew 2007, 777, table 13. Even if the aim of this paper is thoughtprovoking with regard to the interpretation and contextualisation of survey data, it appears that more relevant information can be extracted from the EKAS pottery, including valuable chronological and fabric/ware indicators that seemingly already overcome some of the issues addressed by the author.

Aetolia: Bommeljé and Vroom 1993; Veloukovo: Vroom 1993; Attica: Ober 1987; Olympia: Martin 1995. 824  Evidence from several (other) islands throughout southern Greece provide an interesting dimension, helpful as this evidence can be in investigating the finer details of the geographical distribution of ARSW and LRC: Felten et al. 2003 (Aegina); Hood 1970 (Bay of Itea, Rafti Island). 825  Hayes 1972, 417. 826  Hayes (1972, 417): ARSW is ‘the standard fine ware’ ‘[f]rom the middle of the third century until the early fifth century’. 827  Hayes 2008a, 90. 828  Hayes 1972, 418, 2008a, 89-90. 829  Hayes 1972, 417-418. 823 

98

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:39 UTC All use subject to https://about.jstor.org/terms

The Mid- and Late Roman Periods: Discussion and Chronological Phasing

Figure 66. Aegean-Southwest: map showing the ware proportions of the different sites and surveys (© Philip Bes/ICRATES Project).

Figure 67. Aegean-Southwest: line chart showing the development, in absolute numbers, of the different wares (n=936) (© Philip Bes/ICRATES Project).

99

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:39 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East Quantified information from the available literature corroborates the picture for this region. For Argos the relative development of the major imported wares has been discussed in some detail. ESC remained dominant during the 3rd century, yet the mid- or third quarter of the 3rd century seems to have been a turning point in favour of ARSW: by the late 3rd and early 4th century, ARSW already accounted for about half of the imported tableware. It is thought that ESC was not imported into Argos any longer during the 4th century.830 In ‘groupes homogènes’ of the 4th century ARSW comprised between 50% and 80% of the tablewares, now from several sources within Tunisia and with a greater variety of forms, further characterised by ‘une uniformisation du matériel fin du table à travers l’Empire’;831 it made up c. 85% during the second half of the 4th century, whilst LRC represents c. 10%.832 ARSW and LRC occurred in about equal percentages in the early 5th century, though the latter increased to 75-80% in the course of this century. LRC remained dominant during the 6th and early 7th century; ARSW increased to about 40%.833 Here, however, the Argos evidence and that collected here run dissimilar, as it was not LRC but ARSW that was the dominant imported tableware between the mid-6th and -7th century (Figure 67). Admittedly, at the time of writing Abadie-Reynal did not have much information, either from Corinth or from Boeotia at her disposal;834 some modest changes to Abadie-Reynal’s model can now be proposed. The evidence from, among others, Athens, Sparta, Hyettos, Koroneia, Tanagra835 and Thespiae indicate that ARSW is the tableware arriving most commonly during Late Antiquity, which is in marked contrast to other parts of the Aegean. Contrary to what Abadie-Reynal suggested, it appears that it is not LRC but ARSW that is dominant at these sites after the mid-6th century. Indeed, the available evidence indicates that LRC form 10C (c. 600-650/675) appears to be rare in the region (n=10), being only attested on Kythera, at Halieis, and in the Keos and Methana surveys, all located in the eastern and southeastern part of this region.836 The slightly earlier forms 10A and B (c. 575-625) are a bit more common (n=37), and their distribution encompasses a wider area.837 See now also Abadie-Reynal 2007, 129-130, ranging between 40% and 48% in two 3rd-century groups. 831  Abadie-Reynal 2007, 173-174. 832  Abadie-Reynal 1989a, 144, 2007, 178. 833  Abadie-Reynal 1989a, 155-156. 834  Bes et al. 2006, 2011; Bes accepted b, forthcoming a; Peeters et al. in press; Van der Enden and Bes forthcoming. 835  The majority of the LRC fragments from urban Tanagra in Boeotia (to be fully published in Bes forthcoming a), with a few exceptions, belong to Phocaean LRC. 836  Keos: Cherry et al. 1991 (n=3, at three sites); Methana: Mee and Forbes 1997 (n=1, at MS204); Kythera: Coldstream 1972 (n=5, in three deposits); Halieis: Rudolph 1979 (n=1); a single Phocaean LRC form 10C was recently identified among the pottery excavated at the hilltop site of Agios Konstantinos near Tanagra; not a single specimen of form 10A, 10B or 10C has been found by the Boeotia Project: ARSW dominated the market for later 6th-7th century imported tablewares. The excavations were carried out in 2000 and 2001 by the 9th Ephorate of Prehistoric and Classical Antiquities, Thebes. The Roman pottery is studied by Katerina Chamilaki, whom I warmly thank for allowing me to look at the material and referring to it here. 837  Argos: Aupert 1980 (n=13); Corinth: Slane and Sanders 2005 (n=3, in one deposit), Sanders 1999 (n=5, in two deposits), Biers 1985 (n=2, in

Contemporary ARSW forms (i.e. 99C, 104-109) are more common (n=59); for instance, ARSW forms 104 and 105 are fairly common at Tanagra.838 That the slight evidence of the latest LRC form is confirmed at more than one site, seems significant in relation to the reconstruction of patterns of exchange. That ARSW was the prevailing imported tableware in the Aegean-Southwest from the mid-6th century onwards is in marked contrast to other parts of the Aegean, where LRC continued to dominate (cf. infra). Abadie-Reynal originally already suggested that southern Greece was fortunately located along the axis Africa-Constantinople.839 Southern and central Greece continued to profit from exchange along this axis: in fact, contacts between Africa and the capital were renewed or strengthened after Justinian’s reconquest in 533. 5.3.4 Aegean-North840 This region comprises the northern Aegean and part of the late Republican and early imperial provincia of Makedonia. The few sites available for study span a geographically wide area.

Figure 68. Aegean-North: pie chart showing the proportions of the different wares (n=939) (© Philip Bes/ICRATES Project).

830 

A total of 939 fragments were brought together from the following sites: Abdera, Alexandria Troas, Amphipolis, Assos, Constantinople, Kepia, Samothrake, Smintheion, Stobi, Thasos and Troia (Figure 68).841 ESC is most two deposits); Halieis: Rudolph 1979 (n=1); Keos: Cherry et al. 1991 (n=1); Kythera: Coldstream 1972 (n=3, in two deposits); Methana: Mee and Forbes 1997 (n=9, at eight sites). 838  Bes forthcoming a; Peeters et al. in press. 839  Abadie-Reynal 1989a, 159. 840  For mid- to late Roman pottery from Greece: Papanikola-Bakirtzis and Kousoulakou 2010 (also see n. 812). 841  Abdera, Amphipolis, Kepia, Thasos: Malamidou 2005; Alexandria Troas: Japp 2007; Assos: Zelle 2004; Constantinople; Hayes 1992; Samothrake: Love 1969; Smintheion: Akyürek 1992; Stobi: AndersonStojanović 1992; Troia: Kozal 2001; Tekkök et al. 2001 (also see Tekkök 1996).

100

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:39 UTC All use subject to https://about.jstor.org/terms

The Mid- and Late Roman Periods: Discussion and Chronological Phasing

Figure 69. Aegean-North: map showing the ware proportions of the different sites (© Philip Bes/ICRATES Project).

Figure 70. Aegean-North: line chart showing the development, in absolute numbers, of the different wares (n=754) (© Philip Bes/ICRATES Project).

101

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:39 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East common (c. 41%) — the large published quantities from Abdera and Amphipolis distort the quantification — followed by LRC (c. 35.2%); ARSW accounts for nearly 21% of the data, and small quantities reached this area before LRC appeared. LRD is scarcely represented (c. 2%), as elsewhere in the Aegean; virtually all finds derive from Constantinople, albeit single items were found at Stobi and Thasos. At Stobi, ARSW (n=63) predominates over LRC (n=33), and as before Stobi was probably (partly) linked to other exchange patterns, along the Via Egnatia. Recent evidence from Alexandria Troas indicates that, besides ESC, both ARSW and LRC occur, the latter being more common and ‘in most cases’ from Phokaia, yet ‘not attested in a high number of pieces’; only form 3 was identified, variants E and F in particular.842 After 330 Constantinople acquired its place among the main urban centres of the eastern Mediterranean. Therefore, it is most unfortunate that knowledge of the city’s imported pottery is rather scarce; that found during excavations of St. Polyeuktos remains the most important corpus available.843 Further south, Troia, Assos and Smintheion make the connection with the AegeanEast (cf. infra), and their proximity to Constantinople provided these sites with a framework from which they (and other sites) surely profited (directly or indirectly). The picture can be complemented with evidence from sites westward of Constantinople, along the northern coast of the Aegean (Figure 69). Recently, new important data was published for five sites in this area: Thasos, Amphipolis, Kepia, Philippi and Abdera. The pottery, however, mainly comprised the early and mid-Roman periods, mostly ESC: though this distorts the quantitative and chronological results (Figure 70), it does reflect the prominence of this ware in the Aegean-North.844 LRC formed the majority of the imported tablewares on late Roman Thasos.845 The handful of material published from Samothrake mostly concerns ESC.846 In an area where comparatively little is known with regard to imported tablewares, the late Roman pottery from Demetrias — where LRC seems to be dominant, though ARSW is noted — is significant as it probably belonged to that part of the Aegean in which LRC dominated (note that not much further south ARSW is generally more common, cf. supra, 5.3.3).847 It Japp 2007, 58. Hayes 1992. 844  Malamidou 2005. 845  Abadie-Reynal and Sodini 1992, 87-88. 846  Love 1969. 847  Eiwanger’s (1981) Ware IIA mostly matches ARSW (cat. nos II.1-19, Tafeln 2-3 (note that cat. nos II.86-87, II.89-90, II.92 are rather to be classified as ARSW — perhaps also cat. nos II.132-135, Tafel 9, and certainly cat. nos IIIa.41-43, Tafel 41), his Ware IIB matching LRC (cat. nos II.21-85, II.88, II.91, II.93-94, Tafeln 3-7). Further recognised is ESC (see e.g. Tafeln 31, 34-36; confusion may arise concerning cat. nos III.69-91, IIIa.13-17, as this shape occurs in both ESC and LRC), possibly also ESB (cat. nos III.120-136, IIIa.31, Tafel 37, perhaps including later examples of Hayes form 60? — note that similar vessels were also made in Athens; perhaps also cat. no. IIIa.38, Tafel 41). Cat. no. III.177 is reminiscent of Athenian carinated bowls of mid-Roman date (Grigoropoulos 2009, 414-415, 436-437, 459-460, nos 56-63, 67; Hayes 1972, 407-408, 2008a, 442, fig. 7, 2010, 25, 31, fig. 7, of 3rd to 4th-century date; Reynolds 2004b, 225, 247, fig. 13.17 (possibly late 4th

is noteworthy to observe in Figure 70 that both ARSW and LRC increased between the mid- and late 4th century. LRD only appeared in small quantities by the later 6th and 7th century.848 Except for the finds from Constantinople which Hayes had himself studied, at the time LRP was published he did not have published evidence from the Aegean-North at his disposal.849 ESC was found during the excavations of the Great Palace,850 and Hayes also observed that LRC was already commonly found by 400, yet ARSW did not experience the decrease during the mid-5th century (though slight, cf. Figure 70), regaining in quantity again after the mid-6th century.851 LRC was very common in late5th and early-6th century deposits from the Great Palace and Saraçhane excavations, yet ‘with only rare scraps’ of ARSW.852 In Hayes’ deposit 30, dated between the mid6th and -7th century, LRC form 10 was ‘relatively rare’, with ‘abnormally large quantities of dishes of Form 109’ of ARSW, as well as a possibly local tableware gaining its position on the market.853 At Assos, located relatively close to the main production site of LRC at Phokaia, Zelle observed that ARSW was near-absent during the 5th and first half of the 6th century, possibly resulting from the Vandal occupation of Africa. Both the proximity of Assos to Phokaia (and Gryneion), as well as its intermediate position between Phokaia and Constantinople, strongly influenced the import pattern for Assos.854 The resurgence of ARSW during the second half of the 6th century was not recognised at Assos. For Stobi — where a (supposedly) regionally-made tableware was recognised — it was emphasised that ARSW follows an Aegean pattern, starting at around the mid-3rd century, with the 4th and 5th century being the major period of import. Late ARSW forms seemed uncommon, represented by single examples of Hayes forms 107 and 108.855 LRC was mainly attributed to forms 2 and 3, with the major period of import falling in the 5th century.856 The single fragment of LRD connects well with an Aegean pattern.857 Other evidence is meagre: ARSW was identified at Abdera (n=3) and Thasos (n=12); a single fragment of LRD was noted at the latter.858

842  843 

century); Robinson 1959, 61-63, pls 12, 37, nos K20-22, K24-25, K27: ‘common in 3rd century, pre-Herulian fills’). 848  Hayes 1972, 418, 1992, 7. 849  Hayes 1972, 418. 850  Hayes 1972, 418, n. 1. 851  Hayes 1992, 5, 7. 852  Hayes 1972, 418. 853  Hayes 1992, 7. An additional c. 742 ARSW, c. 1300 LRC and c. 160180 LRD fragments are listed. 854  Zelle 2004, 87-88. 855  Anderson-Stojanović 1992, 55-56. 856  Anderson-Stojanović 1992, 55. 857  Anderson-Stojanović 1992, 55, pl. 40, no. 347. 858  Malamidou 2005, 50-51, 185, fig. 71, no. 939.

102

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:39 UTC All use subject to https://about.jstor.org/terms

The Mid- and Late Roman Periods: Discussion and Chronological Phasing 5.3.5 Aegean-East This region is bounded by Pergamon in the north, and runs south to include the Datça peninsula. A total of 958 fragments has been collected (Figure 71). The proportions are considerably different compared to other parts of the Aegean. LRC accounts for over 83% of the imported tablewares, thus confirming Hayes.859 LRC is considered as imported, despite the proximity of a number of sites to its main area of production.860 That LRC did (or could?) not monopolise the tableware market here provides an interesting contrast with Africa Proconsularis, where virtually no other tableware than ARSW was found during this period. The dominance of LRC, here the main tableware of Late Antiquity, generally continued northward, past Assos to include Constantinople and Thasos (cf. supra). Material has been gathered from Didyma, Emporio, Ephesos, Labraunda, Methymna, Priene, Samos, Sardis and Smyrna; LRC regularly is the dominant red slip tableware (Figure 72).861 The Canadian excavations at Mytilene also suggest that ESC, and later LRC, were the most important tablewares at the site; a similar scenario applies to Hephaestia on Lemnos.862 LRC also dominates over ARSW at Halikarnassos, in a ratio of c. 3.2:1.863 Next to the more plentiful material from Smyrna, Emporio, Sardis, Ephesos and Samos, LRC was also dominant at Didyma (18 out of 24 recorded specimens) and only just at Labraunda (seven out of 13 specimens). Nevertheless, the figures do not fail to show that LRC was the most commonly (imported) tableware between the late 4th and the 7th century. Recently published evidence from Priene (where Phocaean is the most common class of LRC) and Smyrna (4th-5th century ARSW, whilst LRC captures all but the entire range of forms from LRP, and a substantial variety in the stamp repertoire) conforms to this picture.864 This is much less so at Ephesos and Samos (c. 58% and 48% respectively), and considering their proximity to one another this need not be coincidental. The fact that Ephesos remained a major economic pull in the region for overseas traffic during late Antiquity could explain the higher percentages of ARSW, of which sites further north, where LRC had a stronger influence, received smaller proportions. The rather high percentage of LRD for Samos should be noted. Most of these sites are coastal (except Sardis, which nevertheless was connected Hayes 1972, 368-369, 418. See Langlotz 1969; Hayes 1980; Vaag 2003, 2005. 861  Didyma: Wintermeyer 1980; Emporio: Boardman 1989; Ephesos: Beyll 1993; Gassner 1997; Ladstätter 1998; 2005; Meriç 2002; Mitsopoulos-Leon 1991; Outschar 1997 (also see Turnovsky 2005, and several recent contributions in, for example, Ladstätter 2009); Labraunda: Hellström 1965; Methymna: Lamb and Pryce 1940; Priene: Zahn 1904; Samos: Unterkircher 1983; Sardis: Rautman 1995a (also see Stephens Crawford 1990 for finds from the Byzantine shops); Smyrna: Doğer 2007. Also see Berndt 2003 for Miletos. 862  Mytilene: Williams and Williams 1989, 171, 178, 1990, 189-190, in addition to a local terra sigillata identified at the site; Hephaestia: see, for example, Camardo 2002; Greco et al. 2003, 1041. 863  Hansen 2003, 198. 864  Priene: Yilmaz 2007; Smyrna: Doğer 2007. 859  860 

Figure 71. Aegean-East: pie chart showing the proportions of the different wares (n=958) (© Philip Bes/ICRATES Project).

to the coast by a road), where foreign tablewares may have been more widely available. Which wares were more popular inland remains obscure; Sardis in any case received all major tablewares. Some sites are thought to produce (part of) their own need for tablewares. At Sardis, shapes inspired by LRC were made in a presumably local fabric that accounts for half or more of the tablewares from two well-fills.865 Figure 73 shows that quantities of both ESC and ARSW were scanty during the 3rd and first half of the 4th century. The scarcity of ARSW should not be surprising, but the near-absence of ESC is noteworthy. ARSW increased very modestly and gradually after the mid-4th century, perhaps a marginal spin-off of the general increase of ARSW in the east. LRC increased gradually around the same time, and gained ground towards the mid-5th century. It attained a first peak in the decades around 500, followed by a sharp decrease reaching a low around the mid-6th century. This was soon followed by a second period of steep growth to reach a second, even higher peak around 600. This very sharp fall-and-rise is not observed to such an extent in any of the regions already discussed. After the mid-6th century, when the second period of growth of LRC set in, ARSW increased substantially, again best seen as a spin-off of the increasing quantities of ARSW flowing into the Aegean, notably the Aegean-Southwest. LRD now made its appearance, and for this period was also noted in the Aegean-North, whereas on Crete and the Aegean-Southwest LRD is also represented by earlier types. The increase of ARSW, LRD and, above all, LRC from the mid-6th century onwards coincided with one another. All three wares disappeared during the 7th century, although at least some sites were still supplied

865  Rautman 1995a, 42, 79-80. For the manufacture of LRC at Ephesos: Ladstätter and Sauer 2002, 2005.

103

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:39 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East

Figure 72. Aegean-East: map showing the ware proportions of the different sites (© Philip Bes/ICRATES Project).

Figure 73. Aegean-East: line chart showing the development, in absolute numbers, of the different wares (n=730) (© Philip Bes/ICRATES Project).

104

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:39 UTC All use subject to https://about.jstor.org/terms

The Mid- and Late Roman Periods: Discussion and Chronological Phasing

Figure 74. Asia Minor-West Central, -South Coast and -East Central: map showing the ware proportions of the different sites (© Philip Bes/ICRATES Project).

until a very late date.866 Considering that at Sardis LRC form 10 is generally ‘uncommon’,867 the large number of form 10C from Emporio is noteworthy (217 out of a total of 225 specimens for this region). In fact, Phocaean LRC is completely absent from the early 7th-century well deposit at Sardis: Rautman’s ‘Imitation LRC’ now accounts for c. 50% of the tablewares, which somehow recalls the Saraçhane finds. There, deposit 30, which ‘spans the mid-6th to mid7th centuries’, contained less LRC, form 10 being ‘relatively rare’. A larger segment was now claimed by ARSW and a presumed local tableware.868

that, in particular, the Aegean-Southwest saw increasing quantities of ARSW flowing in (cf. supra). 5.3.6 Asia Minor-West Central

No overall quantifications appear to be available for any of the sites except for excavation-specific ones, for example Sardis.869 As for relative developments, LRC was the dominant tableware for the entire late antique period in the Aegean-East, and nearly the sole tableware in the late 5th and early 6th century.870 Only during the second half of the 6th and early 7th century did ARSW hold a sizeable share of the market, again around the same time

This region roughly corresponds to the eastern part of the provincia Asia and western Galatia, and by treating inner Anatolia separately we can again see to what extent major red slip tablewares travelled beyond the ‘façade maritime’. This matter largely remains unanswered, since tablewares have rarely been published. This poses two related questions: whether this is the result of a lack of research that mostly focuses on coastal and near-coastal sites, and whether Asia Minor’s coastal zone consumed the largest share of imported tablewares. Two sites can be noted (Figure 74): a LRD form 9C/10 from Amorion (which can now be supplemented with a handful of ARSW and LRC fragments);871 which possibly echoes Hayes’ observation that LRD ‘during the first half of the seventh century seems to have enjoyed a greater vogue than ever before’.872

See e.g. Emporio: Boardman 1989; Hayes 1972, 418. Rautman 1995a, 79, and Table 1. 868  Hayes 1992, 6-7. 869  Rautman 1995a. 870  Already suspected by Hayes 1972, 418.

871  Tomber 1992, 216; Böhlendorf-Arslan 2007, 275-277; interestingly, several profiles seem somewhat unusual. 872  Hayes 1972, 424.

866  867 

105

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:39 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East Secondly, three ARSW fragments from Sagalassos.873 A LRC fragment from Pessinus may be mentioned.874 5.3.7 Asia Minor-South Coast This region comprises coastal Lycia and Pamphylia, wellpopulated regions during Late Antiquity;875 thus far 108 fragments were collected (Figure 75). The data indicates that LRC (c. 9%) was not common. That LRD attains nearly 40% suggests that Xanthos and Perge were economically more related to Cyprus (Figure 74).876 ARSW is the most common (50%), which sets it quite apart from, for instance, Asia Minor-Cilicia, yet the general low quantity greatly hampers a fair judgement of the evidence. This can be supplemented with evidence from Limyra and Lyrbe. A total of 2720 diagnostic fragments were excavated in Limyra, c. 3% of which were late Roman tablewares: c. 49% of that 3% was classified as (Cypriot) LRD, a further 42% as CRS-imitation; ARSW was represented only by ‘a few pieces’, and LRC was ‘equally sparse’. A remaining 7-8% concerned tablewares of other periods.877 The socalled CRS-imitation was also recognised at Perge, and an ‘Imitation Cypriot Red Slip’ was classed at Anemorion.878 Finds from the Ptolemaion in Limyra include LRD which is ‘found in the same amount as ARS’, whereas LRC ‘is not as numerous as the previous ones [i.e. ARSW and LRD]’.879 This does not tally completely with the collected evidence, yet the prevalence of ARSW and LRD is important to note. Pottery found in excavations in Limyra’s western city also showed a predominance of LRD, both Cypriot and other (local/regional?) classes; ARSW was scarcely represented, whereas LRC was absent.880 At Lyrbe, LRD appears to have been more common than ARSW (no quantification is available).881 That LRC is found at Limyra and Xanthos (a city where ‘all three major wares were in active competition for the market’),882 both in Lycia, yet is not noted (or published) at Perge and Lyrbe, is worth mentioning.883 Pamphylia was clearly not Bes forthcoming a; a LRD Hayes form 7 can be added, as well as several further fragments of Cypriot LRD. Thoen 2002. 875  Foss 1996. 876  Xanthos: Armstrong 2005; Pellegrino 2003; Perge: Atık 1995; Fırat 2003. 877  Vroom 2005, 291-294. 878  Atık 1995, 161, quoted in Vroom 2004, 294. See e.g. 164-165, nos 355-357; at Anemorion, Williams 1989, 50, which need not belong to the same category; also see Poblome et al. 2001a; Poblome and Fırat 2011; Kenkel 2007; Vroom 2005. As it appears, multiple sites in southern Asia Minor were manufacturing CRSW-inspired shapes, as Atık suggests, yet another example of a regional production of morphologically alike tablewares. The ware that Atık mentions concerns a category whose shapes find similarities in both CRSW and SRSW. The result of this is of course that, presumably, not all published SRSW was in fact manufactured at Sagalassos. The city of Ariassos is another likely member of this koinè, where overfired tableware was found during a visit on 19 June 2009. One diagnostic fragment strongly resembled SRSW form 1B230. 879  Eisenmenger 2003, 194. 880  Yener-Marksteiner 2009, 227-229. 881  Atık 1998, 102-103. 882  Hayes 1972, 418. 883  As imported pottery at Sagalassos is generally thought to have come through Perge, it is interesting to note that no LRC has been noted thus far. 873 

874 

Figure 75. Asia Minor-South Coast: pie chart showing the proportions of the different wares (n=108) (© Philip Bes/ICRATES Project).

beyond the distributional range of LRC, so other factors were at play; the distribution of LRD surely was one of these.884 Finally, ARSW, LRD and LRC are now reported from Patara, though the catalogued LRD and LRC hardly represent typical forms.885 The quantitative role of LRD is more complicated to probe, for example in comparison with western Asia Minor-Cilicia (cf. infra), though it must have been distributed in quantity. SRSW also enjoyed a certain popularity at Perge and Lyrbe during the early and mid-Roman periods. To what extent this persisted into Late Antiquity, here and elsewhere, cannot presently be answered. 5.3.8 Asia Minor-East Central Imported tablewares from the interior of eastern Anatolia are equally scarce here (cf. supra, 3.2.8). Single fragments of ARSW and LRD have been recorded from Porsuk (Figure 74),886 to which can now be added 29 LRC fragments from Tavium, located some 140km east of Ankara. Tavium in particular was located along major roads that connected major urban centres such as Constantinople and Ankyra to southern and eastern Turkey. A scarcity or absence of major red slip tablewares in non-coastal areas definitely need not imply an economic backwater. The most telling example is Sagalassos, whose lively Potters’ Quarter catered for the tableware needs of the city and the wider region. SRSW was only fairly recently identified,887 and it is most likely that other workshops were active within Turkey, and elsewhere, that catered for more limited markets.888 Also, local tableware industries can have a profound effect on Now also Kenkel 2007 for finds from Pednelissos. Korkut 2007. 886  Porsuk: Abadie-Reynal 1989b; Tavium: Weber-Hiden 2003. 887  SRSW was originally identified in the late 1980s, the typo-chronology was published in Poblome 1999. The workshops’ output also included oil lamps and figurines (Murphy and Poblome 2011, 32-33, 2012, 208). 888  Poblome 2004a. 884  885 

106

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:39 UTC All use subject to https://about.jstor.org/terms

The Mid- and Late Roman Periods: Discussion and Chronological Phasing the proportion of imported tablewares, as Sagalassos but for instance also Carthage show. 5.3.9 Asia Minor-Cilicia Cilicia is fairly well represented in the published record. The sites nearly cover the entire (predominantly coastal) region, and include large(r) urban centres such as Diokaesareia, Tarsos and Anemorion, but also rural Domuztepe and the monastery at Alahan. Jones’ treatment of pottery from Tarsos remains of great importance for Cilicia Pedias,889 and the recently published pottery from Elaiussa Sebaste offers an important landmark between Anemorion and Tarsos. Material (n=283, Figure 76) was collected from Alahan, Anemorion, Diokaesareia, Domuztepe, Küçük Burnaz and Tarsos.890 ARSW accounts for the highest percentage (c. 39.6%), LRC represents nearly 33%, and LRD comprises 26.5%; small quantities of ESC and ESD were recognised. Geographically, Cilicia Tracheia presents an altogether different picture compared to Cilicia Pedias (Figure 78).891 Moreover, these link up well with neighbouring regions: Cilicia Tracheia with Asia MinorSouth Coast, and Cilicia Pedias to the Levant-Interior North. Zoroğlu’s quantifications of LRD and LRC in Cilicia also bear this out: LRD attains 87% at Anemorion, yet only 4% at Tarsos, whilst LRC accounts for 83% at Tarsos, and 14% at Anemorion.892 Although Alahan does not fit this pattern, with a dominance of ARSW (Figure 77), recently published evidence from Elaiussa Sebaste furnishes a different picture, which shows that LRD and LRC take turns between the mid-5th and 7th century, whilst ARSW in fact has ‘[a] definitely minor role’.893 The recently published survey finds from Diokaesareia are, strictly taken, not easy to correlate. ARSW (n=130) is only marginally more common than LRC (n=120), whilst LRD (n=25) is infrequent. It can be noted that the ratio LRD:LRC better matches Cilicia Pedias (Figures 77-78), where LRC dominates.894 Figure 79 captures the quantitative developments for Cilicia Tracheia and Cilicia Pedias.895 ESC is scarcely represented, with two fragments at Anemorion, as is ESD with a single fragment at Tarsos. ARSW remained rather modest during the 3rd and early 4th century, grew steeply towards the late 4th century, and after a steep fall quantities remained stable during the second half of the 5th and early 6th century. A rise, a fall and a last Jones 1950. Alahan: Williams 1985; Anemorion: Williams 1977, 1989 (Meyza 2007, 101 for a critical note); Diokaesareia: Kramer 2005; Domuztepe: Rossiter and Freed 1991; Küçük Burnaz: Tobin 2004; Tarsos: Jones 1950. 891  Cf. Zoroğlu 2005, 248 (for lesser sites), referring to Hayes 1972, 419 (for Tarsos and Antioch). 892  Zoroğlu 2005, 247, figs 6a, b. 893  Ricci 2007, 175. LRC forms 3 and 10, and LRD Hayes forms 2 and 9 are ‘far more frequent than others’. 894  Kramer 2012, 17-20. 895  Data from the Tarsos Republican Square and Kelenderis excavations (Zoroğlu 2005) provides only form indications. For example, 467 specimens from Tarsos were only generally attributed to LRC form 3. 889  890 

Figure 76. Asia Minor-Cilicia: pie chart showing the proportions of the different wares (n=283) (© Philip Bes/ICRATES Project).

increase characterise the 6th century. Thereafter, quantities dwindled as elsewhere, although some of the latest forms are comparably common. LRC grew fairly evenly from the late 4th century on, reaching a peak around 500. Following a substantial drop, quantities revive somewhat around 600, after which quantities gradually dwindle during the 7th century. LRD form 1 is only moderately represented. A first peak in LRD is noted around 500. Following a serious drop, quantities of LRD rose again and reached an even higher peak in de years around 600. LRD gradually disappeared during the 7th century. The latest LRD forms seem uncommon; at least in the Tarsos Republican Square excavations ARSW and LRC appear to have supplied the bulk of imported tablewares in the second half of the 6th and 7th century; this is, for example, not the case for Eliaussa Sebaste.896 Even though more evidence is available nowadays, these developments nevertheless echo Hayes’ original observations that (1) ARSW may have been virtually the only imported tableware between the mid-3rd and early 5th century;897 (2) during the second half of the 5th and early 6th century LRC was dominant, whilst LRD held a second position;898 and (3) there was a resurgence of ARSW from the mid-6th century onwards [though note Figure 79]. Whereas Hayes refers to the latest ARSW and LRC forms being imported to Antioch into the 7th century, he does not speak about Tarsos, as later LRD and LRC forms were absent; later ARSW forms were not plentiful.899 This, as the Tarsos Republican Square excavations show, could well relate to the chronological occupation of Gözlü Kule in Tarsos.900 The predominance of LRD at Anemorion clearly shows that this part of Cilicia continued to rely on Cyprus for Ricci 2007, 175. Hayes 1972, 419, with references to Tarsos and Antioch. 898  Hayes 1972, 419; Zoroğlu 2005, 248. 899  Hayes 1972, 419. 900  Zoroğlu 2005, 246-248, figs 3-5. 896  897 

107

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:39 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East

Figure 77. Asia Minor-Cilicia: map showing the ware proportions of the different sites and surveys (© Philip Bes/ICRATES Project).

part of its tablewares, probably mostly through direct contacts.901 Whether or not Cilician harbours acted as final destinations, as ports of call or both remains to be seen; still, the considerable percentages of ARSW and LRC demonstrate that the region was well-connected to interregional trade. LRC was the prevailing imported tableware in Cilicia Pedias, not only at Gözlü Kule,902 but also at Domuztepe, located approximately 100km inland. This supposed pattern in part links up with the LevantInterior North (cf. infra). 5.3.10 Levant-Coastal North

Figure 78. Column chart, capturing the differences (in percentages) of ARSW, LRD and LRC between Cilicia Tracheia and Cilicia Pedias (© Philip Bes/ICRATES Project).

Relevant material, 513 fragments in total, has been collected from ‘Ain Dara, Antioch, Epiphaneia, Gindaros and Déhès (Figure 80);903 an important gap is now filled by material from Apamea, which indicates a dominance Similar patterns can be observed for Kelenderis (Zoroğlu 2005) and the Rough Cilicia survey (Blanton 2000). 902  Jones 1950, 205. 903  ‘Ain Dara; McClellan 1999; Antioch: Waagé 1948; Kondoleon 2000; Epiphaneia: Christensen 1986; Lund 1995; Gindaros: Kramer 2004; Déhès: Orssaud 1980. During a visit to Seleukeia Pieria on 5 October 2005, ARSW (e.g. forms 104 and 105) and LRC (e.g. form 10, and form 3 in different variants) were often noted. 901 

108

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:39 UTC All use subject to https://about.jstor.org/terms

The Mid- and Late Roman Periods: Discussion and Chronological Phasing

Figure 79. Asia Minor-Cilicia: line chart showing the development, in absolute numbers, of the different wares (n=219) (© Philip Bes/ICRATES Project).

sites and modest quantity of the evidence, a geographical distinction can be proposed (Figure 81).

Figure 80. Levant-Coastal North: pie chart showing the proportions of the different wares (n=513) (© Philip Bes/ICRATES Project).

of LRC. This corresponds surprisingly well with Cilicia Pedias and the Levant-Interior North (cf. infra, 5.3.11), less so with some sites in the northern coastal zone.904 LRD (9.3%) is more common compared to Cilicia Pedias, where it was only identified at Tarsos. However, ARSW (c. 55.8%) and LRC (c. 35%) dominate the picture, which is in sharp contrast with Cilicia Pedias, where LRC was most common. Notwithstanding the small number of 904 

Vokaer 2013.

ARSW occurs in roughly equal percentages at Antioch and Epiphaneia, yet it is LRC that dominates at Gindaros,905 and Apamea. LRD was identified at all sites except ‘Ain Dara, yet is relatively most common at the more southerly located Epiphaneia (c. 13%), and particularly at Beirut (cf. infra);906 this mirrors the distribution of LRD in the southern Levant (cf. infra). That the proportions between Antioch and Gindaros, a tell located approximately 50km northeast of Antioch, should differ so considerably is noteworthy. Antioch continued to be one of the main (urban) centres in the east during Late Antiquity; Gindaros presumably was located within its sphere of influence. Alternatively, the data available is insufficient, site-specific circumstances at Gindaros may have been chronologically restricted or dispersed, or all three wares did not always travel beyond Antioch, itself a major consumer. Figure 82 shows the quantitative developments: ARSW was the main tableware from the early 3rd until the mid-5th century — Hayes indeed considered ARSW to continue into the early 5th century907 — and from around 400 the ware continued into the 7th century with a fickle, downward course. LRC superseded ARSW by the mid-5th century, and continued into the 7th century, interrupted 905  Kramer 2004. Quantifications are provided, yet overall quantities seem small. 906  Sodini 2000, 189 907  Hayes 1972, 419.

109

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:39 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East

Figure 81. Levant-Coastal North: map showing the ware proportions of the different sites (© Philip Bes/ICRATES Project).

Figure 82. Levant-Coastal North: line chart showing the development, in absolute numbers, of the different wares (n=337) (© Philip Bes/ICRATES Project).

110

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:39 UTC All use subject to https://about.jstor.org/terms

The Mid- and Late Roman Periods: Discussion and Chronological Phasing quantitatively during the second and third quarters of the 6th century. LRD was most common during the second half of the 5th and early 6th century, yet is represented by forms of its entire period of manufacture.908 This finds corroboration in LRP: LRD and LRC dominated during the second half of the 5th century, and continued to do so till ARSW reappeared in larger quantities by the late 6th century;909 Figure 82 does not bear this out. Recently published evidence from the Beirut Souks excavations allows us to trace developments, albeit not continuously, for Beirut. In the early 3rd century, ESA is still dominant, continuing existing trends, with a token presence of ARSW. Not much later, by 230-250, ARSW was already roughly twice as common as ESA, which fits more general trends for this period. Between the mid- to late 4th century, ARSW dominated the tableware spectrum, yet by the early 5th century this had fallen back dramatically, whilst LRC rose in prominence. By the late 6th century the role of ARSW and LRC certainly was not negligible, yet it was LRD that took the lead. The dominance of LRD, followed by LRC and ARSW, is also attested in a mid-6th century context excavated by the French team (BEY 002) (n=554; 54%, 20% and 15% respectively).910 That these trends, especially for the late 6th century, may look a bit at odds with Figure 82, suggests that Beirut participated in (partly) different exchange patterns. As a matter of fact, Beirut conforms much better to the southern coastal Levant (cf. infra), as also seen in the different phase maps (Figures 100-103). Rather than being part of the core eastern markets for ARSW and LRC, the central Levant is closest to Cyprus (as is Cilicia Tracheia: cf. supra, 5.3.9, Figure 78), which we thus see reflected in this Beirut evidence.911 5.3.11 Levant-Interior North Data was collected from sites mostly located along or near the Euphrates: Athis, Doliche, Dura Europos, Halabiyya/ Zenobia, Resafa, Tell Fakhariyah, Tell Rifa’at survey, Umm el-Tlel and Zeugma.912 A small group from Zeugma was recently published, datable to the late 4th and early 5th century;913 all five fragments are ARSW. The authors argue persuasively that the absence of LRD and LRC provides a terminus ante quem for the date of this group,914 which the collected evidence here confirms more generally. Rather than considering local/regional causes to explain the scarcity of (slipped) tablewares (though note the Also mentioned by Hayes 1972, 419, referring to Waagé 1948, 56-57. Hayes 1972, 419. 910  For the Beirut Souks, see Reynolds 2011. For the French excavations, see Bonifay 1996, summarised in Sodini 2000, 189. 911  Reynolds 2010, esp. 111-114, tables 1-3, 6. Also see Reynolds 2011 for an overview of imported tableware trends. 912  Athis: Harper 1980; Doliche: Höpken et al. 2008; Dura Europos: Cox 1949; Halabiyya/Zenobia: Orssaud 1991; Resafa: Konrad 1992; Mackensen 1984; Tell Fakhariyah: Kantor 1958; Tell Rifa’at: Kenrick 1978, 1981; Umm el-Tlel: Majcherek and Taha 2004; Zeugma: AbadieReynal 2005b. 913  Abadie-Reynal et al. 2007. 914  Abadie-Reynal et al. 2007, 183-184. Early forms of LRD and LRC are attested in a contemporary group from Caesarea Maritima (Tomber 1999a). 908  909 

Figure 83. Levant-Interior North: pie chart showing the proportions of the different wares (n=807) (© Philip Bes/ICRATES Project).

relatively small total quantity, n=101), however, this could equally reflect the more general drop off in the quantitative distribution of ARSW that began around this time (cf. infra, chapter 6). Figure 83 captures the proportions (n=807). ARSW (c. 28%) is fairly common, LRD is only marginally present (1.5%). LRC, however, is the most common tableware with over 70%. With the exception of Athis, Dura Europos and Halabiyya/Zenobia (where ARSW is more common), the prevalence of LRC is found at every site here (Figure 84). Admittedly, numbers can be small, and as such less liable to withstand the comparison with a site like Zeugma, for which 862 fragments were reported from the French excavations.915 Both Zeugma (n=862) and Resafa (n=497) can therefore serve as benchmarks, illustrative of the common presence of LRC. Of further note is the rather large number of the late LRC form 10C (c. 600675) at Resafa, and its identification at Doliche and in the Tell Rifa’at survey,916 although it is not encountered everywhere.917 Different variants of LRC form 10 are known from Antioch; at Tarsos (Gözlü Kule) form 10 appears to be absent,918 yet large quantities were found in the Republican Square excavations.919 Figure 85 illustrates the pattern in detail. The majority of the finds fall in the interval between the late 5th and the first half of the 7th century. Third- and 4th-century ARSW forms were found in modest quantities, and the upsurge during the second half of the 4th century is also recognised here, which could be its eastern-most distribution point. Kenrick, however, noted ‘a puzzling absence of pottery attributable to the second, third and fourth centuries A.D.’ Abadie-Reynal 2005b, 523-524. Höpken et al. 2008, nos 35-38; Kenrick 1981, 439, 448. 917  Mackensen 1984, passim. 918  Zoroğlu 2005, 244, on the problems of this publication. 919  Zoroğlu 2005, 246, fig. 4; no variants are given. 915  916 

111

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:39 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East

Figure 84. Levant-Interior North: map showing the ware proportions of the different sites (© Philip Bes/ICRATES Project).

Figure 85. Levant-Interior North: line chart showing the development, in absolute numbers, of the different wares (n=725) (© Philip Bes/ICRATES Project).

112

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:39 UTC All use subject to https://about.jstor.org/terms

Framing Late Hellenistic and Early Roman Tablewares in the Tell Rifa’at survey.920 ARSW dwindled to only a token presence between the early 5th and the mid-6th century, yet between the mid-6th and -7th century ARSW witnessed a second period of increased distribution. LRC is only marginally present at first, yet witnessed a very strong peak in de decades around 500,921 followed by the familiar very sharp decline, then rises again to reach a second peak in the early 7th century. LRD remains scarce through the late Roman period. The fact that LRC arrived at large urban centres such as Antioch and Tarsos in large quantities, being the dominant imported tableware from the second half of the 5th century onwards, possibly into the 7th century,922 reflects the continuation of pan-(eastern) Mediterranean exchange patterns. Thereby, sites located up to c. 400km inland received substantial quantities of LRC (e.g. Resafa), and to a lesser extent ARSW (e.g. Halabiyya/Zenobia). At least some of these sites had a military or supervising role; several are thought to have formed part of the Strata Diocletiana,923 and the dominance of LRC can conceivably be placed in this context (cf. infra, chapter 6); on the other hand, the sites concerned were active or occupied during the period when LRC was most intensively distributed, i.e. the second half of the 5th and first half of the 6th century. This, however, does not explain why late LRC forms continued to arrive in large quantities at Resafa (79 out of the 497 fragments belong to form 10C). Qusair as-Saila exemplifies another such case. Apparently, this military settlement on the Strata Diocletiana was abandoned about 580, yet a number of LRC form 10A (n=7) and 10C (n=6) were identified.924 Non-military occupation possibly continued, the site maintaining its links with existing exchange patterns, plausibly through nearby Resafa and Soura.925 5.3.12 Levant-Coastal South926 The sites studied lie in relatively close proximity to one another: Caesarea Maritima, Jalame, Magen, SamariaSebaste, Tel Anafa, Tell Keisan and Tel Mevorakh; Porphyreon is the only more distant site;927 597 fragments were collected (Figure 86). LRC is the most common imported tableware (41.5%), LRD comes second (c. 40%), followed by ARSW (by and large 18%). Late ESC (single fragments at Caesarea Maritima and Samaria-Sebaste)928 Kenrick 1981, 439. Hayes’ earliest LRC forms are scarcely recognised, known more plentifully from elsewhere in the Levant. 922  Hayes 1972, 419, 2001, 279. 923  Butcher 2003, 417, fig. 191. 924  Mackensen 1984. 925  Mackensen 1984; Konrad 1992. 926  The recently published volume on Sepphoris (Meyers and Meyers 2013), unfortunately, could not be consulted. 927  Caesarea Maritima: Adan-Bayewitz 1986; Bar-Nathan and Adato 1986; Magness 1992; Riley 1975; Tomber 1999a; Jalame: Johnson 1988; Slane and Magness 2005; Magen: Feig 1985; Porphyreon: Domżalski 2002; Samaria-Sebaste: Kenyon 1957; Crowfoot 1957; Tel Anafa: Slane 1997; Tell Keisan: Landgraf 1980; Tel Mevorakh: Rosenthal 1978. 928  Caesarea Maritima: Riley 1975, 41-43, fig. on 42; now also Johnson 2008b, 31-32; Samaria-Sebaste: Crowfoot 1957, 343-344, fig. 83.11. 920  921 

Figure 86. Levant-Coastal South: pie chart showing the proportions of the different wares (n=597) (© Philip Bes/ICRATES Project).

and ESD are scarce, and combined represent about 0.5% of the collected data. All four wares are catalogued among the recently published evidence from Ashkelon: LRC ‘appears to be more abundant’ than LRD, ARSW perhaps comes third, and a single fragment of ESC is attested. Recently published finds from Caesarea Maritima present roughly the same scenario: ARSW comes only third after LRD and LRC; the latter two are ‘well represented’. Both Ashkelon and Caesarea Maritima thus conform to the general trends for this region. Among the late Roman pottery from Tel Hashash, located within the urban confines of modern Tel Aviv, LRD and LRC (and ERSW) were noted, but no ARSW.929 The site-based proportions show that LRD and LRC were generally most common (Figure 87). LRD, closely followed by LRC, was dominant at Jalame — Shiqmona and Sumaqa can be added.930 The proportions are more diverse for Caesarea Maritima, where ARSW attained c. 24%, possibly reflecting its significance as an important economic pulling factor of the (southern) Levant. A similar suggestion was made for Ephesos, where the proportion of ARSW was also higher and somewhat in contrast with surrounding sites, where LRC clearly dominated. Compared to ARSW and LRD, LRC appeared to be more common at sites located in the north or northeast part.931 At Shiqmona, for instance, LRD and LRC comprised c. 44% and 45% respectively, ARSW c. 10%. LRC was also most common at Tell Keisan and Shelomi932 (78% and 53% respectively), LRD accounting for 17% and 33%, Tal and Taxel 2010, 117-118, fig. 24. Fig. 24 does not really well match LRC Hayes form 10. 930  Shiqmona: Amir 2006; Sumaqa: Kingsley 1999. 931  Although LRD (ARSW is unclear) was noted, LRC was ‘the most common of the Late Roman Red Ware (LRRW) types’ encountered in a survey of the Upper Galilee: Frankel et al. 2001, 66-68, fig. 3.11; see also table 3.5 and pls. 36-37 for possible intra-regional variation. 932  Dauphin and Kingsley 2003. 929 

113

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:39 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East

Figure 87. Levant-Coastal South: map showing the ware proportions of the different sites (© Philip Bes/ICRATES Project).

Figure 88. Levant-Coastal South: line chart showing the development, in absolute numbers, of the different wares (n=473) (© Philip Bes/ICRATES Project).

114

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:39 UTC All use subject to https://about.jstor.org/terms

Framing Late Hellenistic and Early Roman Tablewares whilst ARSW was uncommon (5% and 14% respectively). Further north, LRC was most common at Porphyreon.933 In general, it appears that the proportions at sites located more towards the central Levant differ from Caesarea Maritima. On the other hand, proportions from Beirut (cf. supra) showed a different pattern, with LRC comprising 20%, whilst LRD was most common (54%). Further south, at Samaria-Sebaste (n=23), ARSW (n=11) was commonest, followed by LRC (8) and LRD (n=2). The small quantity, however, hampers a clear interpretation.934 The line chart of the collected data (Figure 88) shows that ESC and ESD were rare. ARSW gradually increased from the first half of the 3rd century onwards, and reached a peak in the second half of the 4th century.935 Despite very low numbers, it is interesting to see that ARSW marginally increased during the second half of the 4th century, which probably mirrors the general increase of ARSW during this period. Afterwards quantities decreased and forms postdating the late 4th century are much less represented. LRD and LRC, as elsewhere, joined ARSW from the late 4th century onwards. LRC, marginally attested around the mid-4th century, increased from the late 4th century onward, reaching a peak during the first half of the 6th century, and — temporarily interrupted — continued to be fairly common into the early 7th century: specimens of the latest form 10C were identified at Caesarea Maritima, Porphyreon, Samaria-Sebaste and Tell Keisan;936 LRC is also not uncommon at Sumaqa.937 LRD also appeared by the late 4th century, represented by form 1 (the horizontal segment in Figure 88). After a first peak in the early 6th century, a sharp drop toward the mid-6th century followed, after which a strong increase is noted, reaching a second peak in the later 6th century. LRD dominated the tableware import market after the mid-6th century. If this dominant position of LRD, from the early/mid-6th century onwards (but note the drop), is replicated by new evidence, this could have a wider bearing on Hayes’ observation that LRD ‘seems to have enjoyed a greater vogue than ever before’ in the first half of the 7th century.938 ARSW and LRC witnessed a period of marginal modest growth after the mid-6th century. Interestingly, this does not stand alone, as this was already noted for AfricaCyrenaica, Crete, the Aegean-North and -East, Asia Minor-Cilicia and Cyprus — and to a lesser extent for the Levant-Interior North and -Interior South — and which was already recognised by Hayes: ‘[t]owards the middle of the sixth century African products became very common again everywhere, though Late Roman C still retained the market in the East’.939 933  No full quantification is yet available, although Domżalski (2002) states that LRC dominates. 934  Crowfoot 1957. 935  Hayes 1972, 420. 936  Caesarea Maritima: Magness 1992, 151, fig. 67.16, (Johnson 2008b, 33, 131); Porphyreon: Domżalski 2002, 80-82, fig. 16; Samaria-Sebaste: Crowfoot 1957, 359, 361, fig. 84.23; Tell Keisan: Landgraf 1980, nos 9-15. 937  Kingsley 1999, 275-276. 938  Hayes 1972, 424. 939  Hayes 1972, 423.

5.3.13 Levant-Interior South This region includes the eastern and northeastern parts of the Trajanic province of Arabia, and extends south- and westwards to Africa-Egypt. Relevant material (n=1527) was collected from 21 sites and areas (Figure 89): Ammata, El-Haditha, ‘En-Boqeq, Gadara, Hammat Gader, Hammath Tiberias, Hippos, Jerusalem, Kallirhoe, Khirbet al-Karak, Khirbet ed-Deir, Nabratein, Nessana, Pella, Petra, Philadelphia, Ramat Rahel, Rehovot-in-the-Negev, Sa’neh/Sountha, Upper Zohar and southern Sinai.940 The proportions differ somewhat from those for the LevantCoastal South, as ARSW, for example, is more common (c. 21.4%). Differences are also substantial for LRD (33% here, c. 44% in the Levant-Coastal South), but less so for LRC (c. 45% here, c. 41.5% in the Levant-Coastal South). Recent reports from Hippos confirm the presence of ARSW, LRD and LRC in a wide variety of forms, including several late forms (e.g. LRC form 10C).941 All three wares are now also reported from around Paneas, including LRC form 10C.942 LRC was common in the northern part of this region, around and south of the Sea of Galilee, for instance around Ammata (c. 64%) (Figure 90). LRC was roughly

Figure 89. Levant-Interior South: pie chart showing the proportions of the different wares (n=1527) (© Philip Bes/ICRATES Project). Ammata: Kaptijn 2009, 215-218, table 4.59, figs 4.157-158; ElHaditha: Parker 1994; ‘En Boqeq: Gichon 1993; Gadara: Kenrick 2000; Hammat Gader: Ben-Arieh 1997; Hammath Tiberias: Dothan 1983; Hippos: Mlynarczyk 2002, 2003, 2004, 2005a; Jerusalem: Hayes 1985b (now also see Hayes 2008b); Magness 2005; Rosenthal-Heginbottom 2005; Kallirhoe: Clamer and Magness 1997; Khirbet al-Karak: Delougaz 1960; Khirbet ed-Deir: Calderon 1999; Nabratein: Meyers et al. 1982; Nessana: Baly 1962; Pella: Watson 1990; Petra: Horsfield and Horsfield 1942; Schneider (C.) 1996; Sieler 2004; Philadelphia: Koutsoukou and Najjar 1997; Ramet Rahel: Aharoni 1962; Rehovot-in-the-Negev: Rosenthal-Heginbottom 1988; Sa’neh/Sountha: Lenoir 2003; Upper Zohar: Harper 1995; southern Sinai: Calderon 2000. 941  Mlynarczyk 2006, 2007, 2008. 942  Paneas: Hartal 2009, 161, 163, fig. 10.1. Note that not all profiles match their typological identification: e.g., nos 10-11 on fig. 10.1 are identified as ARSW form 99 which, particularly for no. 11, is perhaps too different. 940 

115

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:39 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East

Figure 90. Levant-Interior South: map showing the ware proportions of the different sites and surveys (© Philip Bes/ICRATES Project).

Figure 91. Levant-Interior South: line chart showing the development, in absolute numbers, of the different wares (n=1078) (© Philip Bes/ICRATES Project).

116

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:39 UTC All use subject to https://about.jstor.org/terms

Framing Late Hellenistic and Early Roman Tablewares as common at Hippos and Jerusalem, yet proportions for ARSW and LRD here are considerably different. A possible intraregional pattern was suggested for the previous region, and some sites here might link up with this: LRC was more common at northern sites such as Ammata, Khirbet al-Karak and Tell Keisan; Kapernaum can be added here,943 as well as nearby Horvat Kur.944 Further south, at ‘En Boqeq on the southwest shore of the Dead Sea, LRC still accounted for c. 48%, followed by LRD (c. 44%), whereas ARSW represented only 7.7%. Then again, this is not dissimilar from the proportional range at Hippos up north. No sites to the north (i.e. Jerusalem and SamariaSebaste; note their low totals!) and south of ‘En Boqeq have comparable percentages of ARSW, except for a few sites in the Levant-Coastal South where ARSW comprised no more than c. 10-15%, though higher percentages were noted for instance for Caesarea Maritima (Figure 87). ‘En Boqeq potentially continues the line of sites where LRC was the major imported tableware. The fort or control post at Upper Zohar (n=129), not far to the northwest of ‘En Boqeq, presents a rather different picture:945 ARSW accounts for 45%, followed by LRC (c. 38%) and LRD (17%). ARSW (86%) is by far the commonest imported tableware at Petra (n=73); LRD and LRC are represented by no more than a handful of specimens. Figure 91 shows that ARSW was virtually the sole imported tableware during the 3rd and 4th centuries, continuing as the dominant tableware during the first half of the 5th century.946 Interestingly, the slight drop toward the mid-5th century cannot unequivocally be traced here; ARSW, however, is rather uneven during the 5th and 6th centuries, and quantities are low. Still, after quantities were fairly low during the second half of the 6th century, ARSW reaches a peak around 600, which might mirror the pattern during the second half of the 6th century observed elsewhere. LRC, appearing in the late 4th century, only increased towards the late 5th century, reaching a peak in the early 6th. Also here, the sharp decrease is followed by an increase, with a second yet much lower peak in the early 7th century. LRD is the least common, and although most forms of its entire production span are attested, later forms are better represented.947 The southern Negev and Sinai rarely figure in the published record, yet a small quantity of evidence is available from several monastic communities in mountainous southern Sinai.948 Amongst the admittedly very small quantity, ARSW is represented by seven fragments, LRD and LRC with one and two respectively. In comparison with Kellia, the author notes the absence of LRD as the dominant imported tableware during the second half of the 6th century (see Figure 91).949 Small quantities of all three Loffreda 1974, 2008a, 2008b, 2008c. Finds (as yet unpublished, full RBHS quantification) studied by the author. 945  Harper 1995, 115, doubting whether its purpose was in fact military. 946  Hayes 1972, 417-424. 947  Hayes 1972, 424. 948  Calderon 2000. 949  Calderon 2000, 199, 201; Hayes 1972, 424. 943  944 

wares nevertheless reached sites in an arid, mountainous environment some 250km from the (Mediterranean) coast.950 5.3.14 Africa-Egypt The proportions (n=270) are shown in Figure 92.951 ARSW is, with c. 52%, the best represented of the three main imported tablewares; LRD (34%) and LRC (13%) take up second and third position.952 The main focus rests on the major port Alexandria,953 as well as data from Kellia and Pelusium (Figure 93)954 — two fragments of ARSW are now known from Adulis further south.955 Unfortunately, the evidence published by Rodziewicz cannot be quantified since only hints are provided as to the importance of different wares and forms. Although ‘[l]ocally made wares continued to satisfy most of the demand for tableware throughout Late Roman times, though the African products were valued, being apparently something of a luxury’,956 the presence of the three main wares of late Antiquity indicate that connections between (notably) Alexandria and other parts of the Roman World were still active.

Figure 92. Africa-Egypt: pie chart showing the proportions of the different wares (n=270) (© Philip Bes/ICRATES Project).

Dahari 2000, 3-6. ERSWs dominated the Egyptian tableware market, at least during part of the late Roman period: Hayes considered that their introduction did not take place prior to the late 4th century The manufacture of the different ERSWs continued into the 7th, possibly into the early 8th century: Hayes 1972, 387, though see Sieler 2008, 273, even if ‘it seems most likely that the large-scale pottery-production at Nag el-Hagar did not begin before the mid-fifth century [...]’ (276). Also Hayes 1972, 387-401, 420-421, 1980, 530. 952  Recent evidence from Schedia, located 25km southeast from Alexandria, points to the presence of ERSWs (most common?), ARSW, LRD and LRC, with scraps of ESA, ESB and ITS in earlier deposits: Martin 2008. 953  Empereur 1998. 954  Alexandria: Rodziewicz 1976; Bonifay and Leffy 2002; Kellia: Egloff 1977; Pelusium: Rodziewicz 1994. 955  Adulis: Peacock 2007, 96. 956  Hayes 1972, 420. 950  951 

117

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:39 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East

Figure 93. Africa-Egypt: map showing the ware proportions of the different sites (© Philip Bes/ICRATES Project).

Figure 94. Africa-Egypt: line chart showing the development, in absolute numbers, of the different wares (n=215) (© Philip Bes/ICRATES Project).

118

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:39 UTC All use subject to https://about.jstor.org/terms

Framing Late Hellenistic and Early Roman Tablewares The percentage of ARSW may be noted in comparison with the Levant-Coastal South, where LRD dominated at some of the coastal sites, yet where ARSW comes third (c. 18%). Alexandria, presumably, conforms more to (a) Mediterranean pattern(s), precisely the reason for Hayes to discuss Alexandria and its hinterland and the rest of Egypt separately.957 It is unfortunate that Gaza is nearly absent in the published record;958 its location in a transitional zone between eastern North Africa (with Alexandria as its main urban centre) and the (southern) coastal Levant is expected to clarify some of the observations made here. That ARSW is common at Alexandria (and Kellia) appears no isolated case. At Ephesos, in an area that was dominated by LRC, ARSW was more common (c. 20%) than the regional percentage (12.4%). At Caesarea Maritima, located in an area where LRD played a significant role, ARSW was also relatively more common (c. 24%) than the regional percentage (c. 18%). That ARSW was somewhat more common at Alexandria, Ephesos and Caesarea Maritima, as well as Antioch (59.2% vs 55.8%), could hint at the significance of these cities in tying together economic exchange across the (eastern) Mediterranean. Figure 94 shows that ARSW experienced the upsurge during the second half of the 4th century also here, which here generally lasts longer. After the mid-5th century quantities decreased unevenly. With three peaks attested around 500, 600 and the mid-7th century, LRD follows an odd course, though generally coinciding with the dwindling of ARSW. LRC hardly plays a role of importance, even if it is most common in the first half of the early 6th century, as elsewhere. The rather common presence of LRD at Alexandria presumably reflects direct contacts with Cyprus. Hayes had observed that LRD was very common at Abu Mena in the late 5th and early 6th century, pondering that the workshops may consciously have destined part of their output for Alexandria959 — a direct supply of LRD probably also headed for Cilicia Tracheia and (parts of) the Levant-Coastal South. This line of exchange is not reflected in comparable percentages of ERSW at Cyprus, which strongly suggests that ERSW was only rarely taken out from Alexandria/Egypt to Cyprus.960 As a matter of fact, ERSW and LRD are both rare in the Aegean (as well as further west): except for Africa-Cyrenaica, both seldom travelled westwards. Finally, LRC increased during the 6th century, and may have been one of the major tablewares at Alexandria by the late 6th century.961 Elsewhere, the proportions at Hermopolis Magna do not differ that much from those at Alexandria, although in light

of Rodziewicz’s absence of quantities such statements remain tentative. ERSW (Hayes’ ERSW-A) accounts for c. 78%, ARSW for c. 21.5%, while LRD and LRC combined do not exceed 0.5%.962 ERSW (c. 86%) is even more common at Mons Porphyrites and its surroundings, the remainder being ARSW.963 These sites, and others (not) listed here, are in general agreement with Hayes’ observation that ‘[l]ocal Egyptian products predominate, and Late Roman C and Cypriot R.S. Ware are more or less absent’.964 Hayes further observed that ‘[u]nlike elsewhere in the East’ Egypt (besides Alexandria) showed ‘no visible decline in imports from North Africa in the fifth century’.965 That one of the more common ARSW forms of the Vandal period, Hayes form 84 (c. 440-500), was ‘[r]are à Kôm el-Dikka’, yet shows up plentifully at Hermopolis Magna and Nag el-Hagar, strengthens this observation.966 At Kom el-Nana, not too distant from Hermopolis Magna, only ERSW and ARSW were noted, the latter of minor quantitative importance.967 Koptos also seems to conform to these trends, since a handful of ARSW and LRD fragments was noted amongst a dominance of ERSW,968 and ERSW(-A) was also most common at Nag el-Hagar (Praesentia?), which was in fact produced at the site; neither LRC nor LRD were noted, yet the quantity of ARSW ‘seem[s] impressively large’. That Hayes form 84, common also at Hermopolis Magna further north, is ‘by far the most common form’ at Nag el-Hagar echoes Hayes’ observation on Egypt’s import pattern of ARSW during the 5th century, outside Alexandria.969 5.3.15 Cyprus The last region under discussion is Cyprus. Earlier, the island could be divided within the context of the manufacture and distribution of ESA and ESD.970 Lund suggested that eastern Cyprus was primarily tied to exchange with the northern Levant, resulting in a predominance of ESA (cf. chapters 3-4), whilst ESD appears to have been more common on the island’s western part, considered as its area of manufacture. Lund’s suggestion is interesting: the threshold of 50% for a certain category of tableware is a working methodology in the reconstruction of regional and supraregional exchange patterns. Published tablewares of the mid- to late Roman periods predominantly reflect western and southern Cyprus. Some evidence is available for central Cyprus, whilst northern and northeast Cyprus largely remain uncharted. Data was collected from Ayios Philon, Kopetra, Kourion, Maroni Petrera, Palaipaphos, Panayia Ematousa, Paphos, and Bailey 1998, 1-2. Tomber 2007a. Hayes 1972, 421. 965  Hayes 1972, 421. 966  Rodziewicz 1976, 37; Bailey 1998, 2. 967  Faiers 2005, 64-99, 173, table 2.4. A few ARSW fragments were found in the survey, yet ERSWs dominate also here (180-211). 968  Herbert and Berlin 2003, 116-118, fig. 93, 123-124, 126, fig. 99. 969  Sieler 2008. 970  Lund 1999, 8-9, 19, fig. 9, 2006b, 36-37, fig. 5. 962  963  964 

Hayes 1972, 420-421: ‘[o]nly in and around Alexandria does the normal Mediterranean pattern of imports prevail’. 958  Gaza was probably an important outlet for LRA 4, manufactured in Gaza and its hinterland. 959  Hayes 1972, 420. 960  Hayes 1972, 420. 961  Hayes 1972, 420-421. 957 

119

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:39 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East means that ARSW is only modestly present, not exceeding 10% (combined with ERSW). Still further east is Panayia Ematousa,974 and a summary of the ceramic finds included a total quantification of the late Hellenistic to late Roman tablewares: LRC was the chief tableware, accounting for nearly 57% of the finds. LRD represented not more than 21%, closely followed by ARSW with 17% (ERSW comprised c. 4%). Recently published evidence from Kourion, which mostly pertains to the (later) 6th and 7th century, conforms well to the pattern for (western) Cyprus, pointing to a ‘predominance of the local [LRD] (possibly made in northwest Cyprus), with moderate quantities of [LRC] (almost all of 6th-century types), and [ARSW], which comes to the fore in the seventh century’. Additionally, ERSWs are noted infrequently, appearing in mid-7th century deposits.975 Figure 95. Cyprus: pie chart showing the proportions of the different wares (n=594) (© Philip Bes/ICRATES Project).

Salamis.971 As very little had been published on tablewares from late Roman Cyprus when LRP appeared, Cyprus consequently appears only briefly.972 A substantial body of evidence has been published since. The proportions, based on 594 fragments, are shown in Figure 95. As might be expected, considering its presumed origin, LRD dominates, yet by only little more than half of the finds (c. 56.4%). LRC comes second, comprising c. 28% of the finds, and with about 14.3% ARSW is fairly uncommon; ESC and ESD combined make up about 1%. The site-specific proportions do not reflect the general proportions (Figure 96). LRD’s dominant position is reflected at Palaipaphos (57%) and Paphos (c. 65%), both on western Cyprus. With c. 20% at Paphos and about twice as much at Palaipaphos, LRC presents a rather different picture. Percentages of ARSW range between c. 5-20%. From Kourion, on the southwest coast, the evidence is rather meagre: LRD and LRC are each represented by seven fragments, ARSW by one.973 Percentages instead of sherd counts are available for Maroni Petrera, located on the south coast, which show that LRD was most common (43%), whilst ARSW and LRC are represented by small percentages; note the differences with appendix 4b. Further east along the coast, full quantifications are available for Kopetra (n=1246) and the hinterland survey (n=842): LRD dominated with c. 51% and 58% respectively, which conforms well to the general picture. LRC comes second with 40% and 35% respectively, which Kopetra: Rautman 2003; Kourion: Neuru and Soren 1987; Maroni Petrera: Tomber 2002; Palaipaphos: Lund 1993; Panayia Ematousa: Lund 1996a, 1998; Paphos: Hayes 1991, 2003; Guidice et al. 1996, 1999; Guidice and Narbone 2000, 2001; Malfitana 1992, 1993, 1994; Malfitana et al. 2004; Salamis: Diederichs 1980. 972  Hayes 1972, 419-420, 1980, 528-529. 973  Neuru and Soren 1987. 971 

The available percentages for Palaipaphos (in the west), Kopetra (on the south coast) and Panayia Ematousa (in the southeast), echo Lund’s division of late Hellenistic and early Roman Cyprus into a western and an eastern zone. Some additional evidence is available from the east coast: at Salamis, ARSW (n=6) and LRD (n=4) were catalogued, yet no LRC.976 At Ayios Philon (n=39), each of the three wares accounts for about a third of the finds, which is not particularly helpful here. If indeed the higher proportion of LRC at Panayia Ematousa was the result of the site belonging to (a) different exchange zone(s), this could also explain the higher percentages of LRC at Kopetra. A survey carried out by the Sydney Cyprus Survey Project (SCSP) focused on a number of sites in the northern foothills of the Troodos Mountains in central Cyprus. A site-based count per tableware category resulted from the project’s aim to identify industrial and agricultural settlements. The total count indicated that LRD was prevalent (c. 57%), followed by ARSW (21%), LRC (15%), ESC and ERSW (each about 3%). Despite the greater proximity to the south coast, the composition of the surface pottery collected by the SCSP more closely resembles Paphos and Palaipaphos, and to some extent Kopetra. In a pottery workshop for coarse wares at Dhiorios, 29 out of 31 fragments were attributed to LRD, whilst ARSW and LRC were represented by a single fragment each.977 This low number nevertheless roughly conforms to the proportions along the west coast and in central Cyprus: a predominance of LRD, with secondary roles for ARSW and LRC. The dominance of LRD recurs in Figure 97, and remained the major tableware between the later 5th through the 6th and 7th centuries. ESC is scarcely represented, which accords to the eastern Mediterranean evidence (apart from the Aegean). Some late ESD is still typologically present Lund 2006a, b; preliminary reports, Lund 1996a, 1998. Hayes 2007, 436. 976  Diederichs 1980. 977  Catling 1972. 974  975 

120

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:39 UTC All use subject to https://about.jstor.org/terms

Framing Late Hellenistic and Early Roman Tablewares

Figure 96. Cyprus: map showing the ware proportions of the different sites and surveys (© Philip Bes/ICRATES Project).

Figure 97. Cyprus: line chart showing the development, in absolute numbers, of the different wares (n=502) (© Philip Bes/ICRATES Project).

121

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:39 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East in the 3rd century. ARSW remained uncommon from the early 3rd right through the mid-6th century, even if a minor increase occurred during the second half of the 4th century, as elsewhere. Hayes noted a more common appearance, and at Soli in the northwest ‘a fair amount of fine North African ware was being imported in the latter part of the third century, just as elsewhere in the East’.978 ARSW did enjoy an increase during the late 6th and first half of the 7th century; in other regions, growth took place earlier, in the second half of the 6th century. It is not until the later 5th century that an increase of LRD and LRC is noted. LRC gradually increased from the early 5th century onwards, yet a marked rise started just after the mid-5th century, reaching a peak in the first half of the 6th century (as noted elsewhere). After a short phase of shrinkage, a second increase followed that reached a second, slightly lower, peak in around 600. LRD only became the major tableware after the late 5th century, reaching a peak in the early 6th century, then proceeding to a sharp decrease and increase, and arriving at an even higher peak in the years around 600. Hayes had indeed observed that ‘Cypriot R.S. Ware suffered a marked change in fortune for half a century or so, though it revived towards A.D. 600’.979 In the course of the 7th century all three wares disappeared, yet LRD maintained its dominant position.

978  979 

Hayes 1972, 419. Hayes 1972, 423-424.

5.4 Tableware Distribution in the Mid- to Late Roman East The quantitative, chronological and geographical dimensions of the regional evidence provide the building blocks for further interpretation. This section synthesises this regional evidence per chronological phase. Absolute quantities for all classes, phase by phase, are presented in appendix 4b. 5.4.1 Phase 5 (c. 200/225-325) ARSW, accompanied by only ESC till the late 3rd century and whose distribution pattern mainly comprised the Aegean — though recent evidence shows that ESA and ESB continued well into the 3rd century — gradually expanded quantitatively and geographically in the eastern Mediterranean after the early 3rd century. Basically only ESC and ARSW circulated within the eastern Mediterranean during this phase, though recent evidence has shown ESA and ESB to continue well into the 3rd century, albeit on a reduced, regional scale. Whereas ESC is mostly confined to the Aegean, ARSW attained a wide yet concentrated distribution throughout the east. To define

Figure 98. Map showing the distribution, in absolute numbers, of ESC and ARSW during phase 5b (c. 225/250-325) (© Philip Bes/ICRATES Project).

122

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:39 UTC All use subject to https://about.jstor.org/terms

Framing Late Hellenistic and Early Roman Tablewares

Figure 99. Map showing the distribution, in absolute numbers, of ARSW during phase 5a (c. 175/200-225/250) (© Philip Bes/ICRATES Project).

when both wares, especially ARSW, began to become more commonly distributed within the east is problematic, since certain forms of both wares have rather long date ranges (e.g. ARSW form 50A, c. 230-325); at Ephesos, in any case, ESC and ARSW ousted ESB by the mid-3rd century.980 The mere presence of a form at a particular site is illustrative in gauging the geographical action radius, or extent, of both wares during this phase, yet fails to probe the more precise dimensions (chronological, quantitative, intensity) of distribution. Hayes provides some clues as to the morphological development of a number of his forms, such as ESC form 1 (c. mid-2nd to the third quarter of the 3rd century), which tends to become more open and shallow, and generally smaller in size as it developed.981 This is generally tricky with fragmentary material, but can be overcome partly through the use of (published) archaeological contexts, which may provide a more precise indication of when a form first arrived at a site. Moreover, quantified archaeological contexts, preferably of high quality, offer insights into the proportions of different wares.

Figure 98 illustrates the distribution of both wares during phase 5(b) (appendix 4b). ESC’s main area of distribution clearly is the Aegean, with occasional examples found further afield: e.g. in Cyrenaica (Berenice), Asia MinorSouth Coast (Xanthos), Cyprus (Paphos) and the LevantCoastal South (Samaria-Sebaste), and which can be supplemented with findspots in the Black Sea, Cyrenaica and Italy.982 The distribution was nevertheless mostly concentrated throughout the Aegean; new data that has become available since LRP and A Supplement to LRP leaves the outlines of its distribution pattern intact, yet helps to condense it.983 Hayes observed that ‘[i]t is probable that it was equally common in Cyrenaica, though the evidence at present is rather scanty’ ― Kenrick indeed identified ESC at Berenice, and quantified data from Ptolemais is now available984 ― and that it could have played a secondary role next to Pontic pottery in the Black Sea.985

Hayes 1972, map 13, and the discussion on 317-318. E.g. Malamidou 2005. 984  Kenrick 1985, 257-265, esp. 257, 259; Domżalski 2012, passim. 985  Hayes 1972, 317-318. 982  983 

980  981 

Ladstätter 2008, 97. Hayes 1972, 318.

123

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:39 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East The classification of ARSW into fabrics A, C, D etc.986 has, thus far, found only rare following in the eastern Mediterranean, yet a more common use potentially throws a more detailed light on the relative importance of the different places/regions of production.987 For phase 5, the distribution of ARSW seems to have been thinner yet much more widespread compared to ESC, as small quantities are found throughout the eastern Mediterranean. ARSW is most common at Berenice, and shows a mere scatter in the Aegean which presumably is related to the dominant position of ESC, even if the mid-3rd century was pivotal in Athens and Argos.988 Further east and on Crete, however, finds become marginally more numerous at places, for example, at Antioch and Petra. The overall chart shows that ARSW gradually increased after the early 3rd century, partly due to the start date of some popular forms of this phase, e.g. forms 45A-B and 50A, but principally shows that the distribution increased quantitatively. Hayes had already placed the start of this growth around 240-250, a change that ‘correspond aussi à un déplacement des centres de production des sigillées claires africaines’ that now ‘proviennent [...] de Tunisie centrale’.989 AbadieReynal also draws our attention to the fact that Argos, and for that matter the east more generally, received a very restricted form repertoire compared to western sites, which mainly concerns forms 45A-B (c. 230-320) and 50A (c. 230-325): ‘donc d’importations visiblement organisées à partir d’un lieu de production et selon des critères très précis, ne laissant pas la place aux hasards des initiatives de negotiatores’.990 In consideration of the long date ranges of these forms, several archaeological deposits shed a bit more light on this issue. The first archaeological clue is related to the Persian incursions during the years 253-256. Amongst others, Zeugma and Dura Europos are known (historically, archaeologically) to have been destroyed in the course of these events and specimens of form 50A have come to light at both sites. In fact, Dura Europos offers a more secure date since it was not really reoccupied afterwards991 and thus ‘gives a firm terminus ante quem for the date of its arrival en masse, which may be placed around A.D. 24050’.992 Not many deposits have been dated to the mid-3rd century: examples are known from Corinth, Athens, Stobi, Troia, Knossos and Tenos; two recently published deposits (dated to 225-250) from Ephesos, which contain an early ARSW Hayes form 50A and a form 14/15 respectively, should be mentioned here, as well as a Severan deposit (c. 190-225/235?) with a (very?) early ARSW Hayes form 50A and late variants of ESB.993 Two more recently published deposits from Ephesos further document the thin Bonifay et al. 2012, 42, fig. 1. Carandini et al. 1981; for a recent example, see e.g. Sieler 2004. 988  Athens: Hayes 2008a, 71-72; Argos: Abadie-Reynal 2007, 171-173. 989  Hayes 1972, 423, and maps 4-5, 23-24, 2008a, 71-72; Abadie-Reynal 2007, 171. 990  Abadie-Reynal 2007, 171. 991  Cox 1949, 14-15; Heath 2011, 72, fig. 3-20. 992  Hayes 1972, 421. 993  See n. 126. 986  987 

yet persistent presence of 3rd-century ARSW.994 A late3rd to early-4th century drain fill from Corinth, besides a small portion of residual tablewares, contained local/ regional (Attic, Corinthian) wares as well as ESC and ARSW.995 Groups K and M, layer V, at Athens seemingly contained mostly Attic tablewares with only a few ARSW, and group M layers VI-VII contained only three locallymade vessels.996 At Olympia, ARSW was common by the second half of the 3rd century; unfortunately, no figures are available.997 At Stobi, lot 1295 (c. 225-325) contained less than a handful of ARSW and ESC fragments.998 In a small well deposit at Troia (c. 260-275), red slip tablewares are represented only by ESC.999 A group of small deposits from Knossos contained specimens of both ARSW and ESC, yet the latter gained a near-monopoly in the Severan deposits in the Unexplored Mansion.1000 From Tenos, one fragment of ARSW and ESC each were published from a deposit dated to the mid-3rd century.1001 This rather limited evidence makes it difficult to gauge the significance of ESC and ARSW, and clearly more quantified data is needed. All mid-3rd century deposits come from sites in the Aegean, except two, from Paphos and the recently published Beirut evidence.1002 This does not entail a regional, Aegean pattern, as the distribution of ARSW in the eastern Mediterranean implies. This scarcity of available deposits from outside the Aegean simply prompts a greater reliance on the typo-chronological evidence. The evidence, inevitably incomplete, nevertheless supports the view that the distribution of ARSW changes during phase 5. ARSW forms that do not postdate the mid-3rd century were clearly less widely distributed, suggesting that the distribution of ARSW was more irregular or modest during the late 2nd and early 3rd century, possibly into the second quarter of that century (phase 5a) (Figure 99) than during phase 5b (Figure 98).1003 In fact, the change in the intensity of the distribution pattern of ARSW could well have coincided with a change in morphology. This twofold pattern (phases 5a-b) strongly resembles Hayes’ mapping of ARSW between the 2nd and mid-3rd century,1004 which sharply contrasts with his map 61005 that illustrates the distribution of forms 48-50 (particularly form 50A) of the mid-3rd and throughout the early 4th century. Ladstätter 2010a, 193, Fundkomplexe A-Z/2-SR 5a (250-275) and A-Z/2-SR 11 (3rd century), containing specimens of Hayes forms 45A and 50A. 995  Slane 1994, 127-168, esp. 129-135. Also see n. 814 for a small deposit from Titanè tentatively dated to the first half/second quarter of the 3rd century. 996  Robinson 1959, groups K and M, layers V-VII; note that this well may have been either operative or derelict. 997  Schauer 2003. 998  Anderson-Stojanović 1992, lot 1295, 54, 56, 158, nos 343-344, 350a, 514. 999  Tekkök et al. 2001, 365. 1000  Hayes 1983; Sackett 1992, 158-159, 163 (fig. 6), 242-249. 1001  Etienne and Braun 1986, group H. 1002  Hayes 2003, deposit 8; Reynolds 2010, 111-112, table 1 (deposit of 230-250). 1003  Hayes 1972, 417-424, noting certain regional differences for ARSW. 1004  Hayes 1972, maps 3-5, 454-455. 1005  Hayes 1972, map 6, 455. 994 

124

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:39 UTC All use subject to https://about.jstor.org/terms

Framing Late Hellenistic and Early Roman Tablewares

Figure 100. Map showing the distribution, in absolute numbers, of ARSW, LRD and LRC during phase 6 (c. 325-400/425) (© Philip Bes/ICRATES Project).

5.4.2 Phase 6 (c. 325-400/425) ARSW increased substantially till the late 4th, possibly the early 5th century, after which quantities begin to decrease. This phase is further characterised by the set off in the distribution of LRD and LRC from the mid-/late 4th century onwards; both experienced a gradual increase into the 5th century. Early LRD is chiefly found in the southeast Mediterranean, whilst LRC primarily circulated in the Aegean. By the end of phase 5, ESC had more or less disappeared from the (mostly Aegean) market, though given the resemblance of ESC and LRC forms production presumably continued (on a reduced scale) (cf. supra, 2.2.3, 2.2.8).1006 It is generally accepted that ARSW attained a very wide distribution throughout the eastern Mediterranean during phase 6, which roughly spans the 4th and early 5th century, and possibly enjoyed its most extensive geographical reach.1007 The overall chart (Figure 58) shows a marked increase around the mid-4th century that lasted until the end of that century, during 1006  1007 

Hayes 1972, 317, 417; Vaag 2005. Hayes 1972, 423, maps 6-9, 25-26.

which ARSW more than doubled in quantity. Moreover, ARSW was virtually the only tableware in the eastern Mediterranean with a supraregional distribution during a large part of the 4th century. ARSW was joined by LRD and LRC only from the mid-/late 4th century onwards, though clearly these initially kept a moderate position into the early 5th century. A number of ARSW forms fit the chronological boundaries of this phase rather well, whereas the date range of other forms runs into phase 7, for instance the common form 67. As the larger part of their date range falls into phase 6 these will be included here; that said, the map for phase 6 can thus be used indicatively in the discussion of phase 7. The collected data does not fail to show a manifest increase, geographically as well as quantitatively, in particular in the (western) Aegean and the Levant (Figure 100). In comparison to phase 5, ARSW now also reached more inland sites.1008 Figure 100 presents a more condensed pattern of distribution compared to Hayes’ map 6, represented for instance by finds now also along the Nile. This insight could very well result from an 1008 

125

Also shown by Hayes 1972, maps 6-8, 455-456.

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:39 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East actual intensified distribution into non-coastal zones, yet we should not ignore the fact that much more evidence is available nowadays. ARSW now very commonly occurred in the Aegean-Southwest and on Crete, whereas it appears to be rather scarce in the Aegean-East, where LRC probably already dominated the market from the mid-4th century onwards.1009 Little evidence is known for most of the southern coast of Turkey, though some ARSW is published for a number of coastal and inland sites in Cilicia. ARSW also fairly commonly appears at Cyprus’ south coast, and is represented at Alexandria.1010 In the northern Levant, ARSW has been identified at Antioch and Epiphaneia, and a synthesis is now available for Zeugma, as well as midto late 4th century evidence from Beirut1011 that shows that ARSW was predominant — though by the early 5th century it had shrunk dramatically, matched by a strong presence of both LRD and LRC. ARSW also found its way to sites at great distances from the coast, such as Resafa, Athis and Tell Fakhariyah. In the southern Levant, ARSW appears in quantities at Jerusalem, Caesarea Maritima and Jalame, while smaller numbers have been recognised at Hippos, Magen, Kallirhoe and Samaria-Sebaste. Further inland, a notable quantity has been published from Petra, and contemporary forms have also been identified at Rehovot-in-the-Negev. The distribution of LRD1012 mainly encompassed Cyprus, Cilicia, the southern Levant and coastal Egypt. Import of LRD into the southern Levant was already quite considerable; Caesarea Maritima surely was one of its main inlets. In comparison to LRD, LRC appears to have had a somewhat wider impact during its initial phase, as substantial numbers of early forms showed up at sites in the Aegean-East (e.g. Assos, Ephesos); smaller quantities are known from Emporio and Smintheion.1013 Early forms are also attested around the Aegean, notably in central Greece and at Gortyna, where they were already quite common; additional specimens are known from Constantinople and Stobi. Outside the Aegean, early forms were noted at Butrint,1014 Berenice, Anemorion, Palaipaphos, Paphos, Kourion, Kopetra, Antioch, Gindaros and Epiphaneia. LRC relatively swiftly acquired a wide distribution by the early 5th century, which stretched from the Aegean to the Levant, the characteristics of which are also illustrated by Hayes.1015 The well-dated harbour deposit from Caesarea Maritima, published by Tomber, throws some light on the distribution of early LRD and LRC.1016 All three loci, datable to the very end of the 4th or the early 5th century, contain ARSW as well as early forms of LRC but above all LRD. LRD Hayes 1972, 317. Bonifay and Leffy 2002; Rodziewicz 1976. 1011  Reynolds 2010, 112-113, tables 2a-3. 1012  Hayes 1972, 372-373 for LRD form 1, and map 17. 1013  See Hayes (2008a, 85-86) for recent remarks on the chronology of his early forms of LRC. 1014  Butrint: Reynolds 2002, 2004b. 1015  Hayes 1972, map 14, 459. 1016  Tomber 1999a. 1009  1010 

is in fact the most common, and had already attained ascendancy over ARSW by this time, whilst LRC is sparsely represented. It reflects the Levant-Coastal South pattern (cf. supra, 5.3.12). Few archaeological deposits were encountered in the literature: examples dating to the second half of the 4th century or slightly later were found at Alexandria, Anemorion, Corinth and Stobi, in which ARSW was the sole late Roman Red Slip Ware catalogued; a late 4th-early 5th century group from Zeugma can now be added.1017 At the same time, a destruction deposit from Kenchreai contained an Attic fragment, illustrating the significance of local and regional wares, such as Macedonian Grey Ware at (e.g.) Stobi.1018 Despite the small quantities, in lots 169, 282, 306-307, 1629, 1783 and 2393 from Stobi, which were all dated to the late 4th century, ARSW is uncommon in comparison to Macedonian Grey Ware and Pannonian Lead-Glazed.1019 Lots 1370 and 2663 dated to the late 4th and early 5th century each contained a fragment of Macedonian Grey Ware,1020 and a deposit of similar date from Corinth contained two Attic and one ARSW fragment.1021 Further deposits of similar date are known from Athens, Constantinople and Knossos. At Athens, an Attic fragment was found in group M, layer IX; deposit 3 at Constantinople contained a LRC fragment;1022 and several deposits at Knossos contained ESC, with not more than a handful of ARSW and LRC.1023 Williams’ group A7 from Anemorion, dated to the first quarter of the 5th century, contained ARSW (n=2) and LRC (n=1) besides much residual material.1024 At Athens, Robinson’s group M, layer X contained only two Attic fragments, thus similar to layer IX (cf. supra).1025 Locus 3042 at Butrint, admittedly beyond the scope here, contained three ARSW and one LRC fragment.1026 Deposit 7 from the Saraçhane excavations contained a single LRC fragment.1027 Finally, lots 274, 278, 293-295, 297-300, 302-303, 514 and 1016 at Stobi contained ARSW (n=8), LRC (n=1), Macedonian Grey Ware (n=5) and Pannonian Lead-Glazed (n=3).1028 In summary, this phase is particularly characterised by the expansive growth of ARSW, possibly into the early 5th century; LRC in particular attained a wide distribution relatively quickly, whilst LRD remained primarily restricted to Cyprus and the southeast Mediterranean. 1017  Alexandria: Bonifay and Leffy 2002, 40, 69; Anemorion: Williams 1989, 6, 11, 15: group 6 that only contains residual ESA and ESD; Corinth: Slane 1990, 55-56: locus 1999 that contained three ARSW fragments; Stobi: Anderson-Stojanović 1992, 53, 56, 70: lots 523, 1314, 1404, of which only the latter contained two fragments of ARSW; Zeugma: Abadie-Reynal et al. 2007. 1018  Adamsheck 1979, 90-91. 1019  Anderson-Stojanović 1992, 56-57, 67-70, 76-77. 1020  Anderson-Stojanović 1992, 68-69. 1021  Williams II and Zervos 1985, 78-79. 1022  Hayes 1992, 91-92. 1023  Hayes 1983: deposits AE (N) Topsoil (121-122, 132), AE4 (121-122), and Baulk 5 (118, 120-122). 1024  Williams 1989, 39, 44, 47. 1025  Robinson 1959, 111. 1026  Reynolds 2002, 222. 1027  Hayes 1992, 92, 152. 1028  Anderson-Stojanović 1992, 46, 48, 56-58, 60, 68-69, 71, 76.

126

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:39 UTC All use subject to https://about.jstor.org/terms

Framing Late Hellenistic and Early Roman Tablewares This is mainly based on the overall quantities, since the available archaeological contexts are insufficient in quantity and/or quality to investigate their distribution in more chronological detail. Only the fully quantified harbour deposit from Caesarea Maritima firmly proves that ARSW, LRD and LRC all circulated in the eastern Mediterranean by the late 4th, early 5th century, each with its quantitative, chronological and geographical characteristics. These distribution patterns were to change considerably during phase 7.

and Kelenderis (c. 59%),1033 LRD is the most common late Roman Red Slip Ware, whereas it accounts for not more than 28% at Elaiussa Sebaste;1034 further east, in Cilicia Pedias, it is more or less absent (cf. infra). Westwards, LRD is known from Tocra and Berenice, and it also sporadically appears at Gortyna, Knossos, Argos, Corinth, Athens and Halieis. Nearly no finds are known from the Aegean-East and -North, which reflects the predominance of LRC. Generally speaking, as in phase 6 LRD occurs only sporadically west of the line Alexandria-Lycia.

5.4.3 Phase 7 (c. 425/450-500/525)

Also here certain forms overlap with the previous and/ or subsequent phases, especially in relation to phase 6; in particular this concerns ARSW form 67, nevertheless attributed to phase 6, yet whose date range runs between c. 350 and 500.1029 A large number of ARSW forms nevertheless fall well within this phase’s range. The date range of LRD form 1 (c. 375-475; n=94) overlaps equally with phase 6, therefore it is included here again. Whereas this creates a distorted view concerning the distribution quantity of LRD, not including it causes this form to be under-represented. Figure 101 presents the geographical and quantitative distribution during phase 7. The distribution of LRD has clearly expanded in comparison with phase 6 (even if form 1 would be omitted here), yet the nature of this change appears to be rather quantitative than geographical, as LRD continued to predominate in the same regions (although specimens of forms 2 and 3 now appear at new sites). LRD is now more frequent on Cyprus, where it was only marginally represented during phase 6. LRD is also slightly more common now in the coastal Levant (e.g. Sumaqa), yet still seldom travelled beyond the coastal zone, indicated by the few specimens of form 3 at and around Resafa.1030 Finds from Alexandria provide clues that LRD grew in importance,1031 which was also the case at Perge and Anemorion, and LRD regularly appears in late Roman deposits at Limyra;1032 these regions could very well have been in direct trade with Cyprus. Further evidence is available for Cilicia, notably Cilicia Tracheia: in the Rough Cilicia survey (>89%), at Anemorion (c. 67%)

As during phase 6, LRC was already found in modest numbers in parts of the eastern Mediterranean, and higher quantities were noted in the Aegean-East and Crete. In phase 7 the existing distribution pattern greatly expanded; it remained dominant in the Aegean-East, where it is now also found at Emporio, Labraunda and Didyma, and in greater numbers at Constantinople. LRC was now also much more common on Crete but especially in the Aegean-Southwest, comprising 80-90% of the imported tablewares at late 5thcentury Athens;1035 finds from the Methana, Keos and Pylos surveys demonstrate that it now clearly made its way into rural areas. LRC is the most common tableware at Gortyna during this phase (Figure 101). How Knossos relates to this is more difficult to say, since late Roman pottery has not been published in great quantities. Further east, LRC also commonly appears on Cyprus, at Paphos and Palaipaphos, to a lesser extent at Maroni Petrera, Kourion and Kopetra (cf. supra). LRC not only increases quantitatively in the Levant, notably the coastal zone, but now also appears at more sites. There seems to be a difference, however, between the Levant-Coastal North and -Coastal South: LRC and LRD, to a lesser extent ARSW, appear to be more common in the Levant-Coastal South, yet it is noteworthy that material is relatively plentifully available, and includes Caesarea Maritima, Jerusalem, Sumaqa, Jalame and Hippos. A detailed distribution map, illustrating the distribution of LRC at about 135 sites in the southern Levant, unfortunately lacks a chronological and quantitative dimension.1036 ARSW was the most common imported tableware in the LevantCoastal North — at least during part of this phase (Figure 82) — from where only Antioch and Epiphaneia are known in some detail. At Zeugma, a considerable distance from the coast, LRC was dominant in the French excavations.1037 Unfortunately, only the ARSW is published in detail. Hayes judged that LRC and LRD ‘had virtually taken over the market’ in Syria by the second half of the 5th century;1038 at least for LRC we can see this reflected in Figure 82. The evidence from Epiphaneia seems to offer some clarity. Lund observed that ARSW, despite being the most common late Roman Red Slip Ware found, diminished after the late 4th century and did not increase again in the later 6th century.

Bonifay 2004, 171-173, fig. 92. Qusair as-Saila, Resafa: Mackensen 1984. 1031  Rodziewicz 1976, 46, formes D7 (‘[e]xiste assez couramment à Alexandrie, à Nea Paphos et à Antioche’) and D8 (‘[a]pparait très souvent à Alexandrie, à Nea Paphos et à Antioche’). 1032  Yener-Marksteiner 2009; Eisenmenger 2003; Vroom 1998, 2004, 2005.

Rough Cilicia: Blanton 2000; Anemorion: Williams 1985, 1989; Kelenderis: Zoroğlu 2005. 1034  Elaiussa Sebaste: Ferrazzoli 2003a, b. Ricci 2007, 175, indicates that LRD held a minor position compared to LRC during the second half of the 5th century. 1035  Hayes 2008a, 85. 1036  Kingsley 2003. 1037  Abadie-Reynal 2005b. 1038  Hayes 1972, 419.

ARSW was still found in the eastern Mediterranean, yet more restricted in quantities and geographical range, a decrease that continued into the early 6th century. LRD and LRC both experienced their first period of growth, LRD reaching a first peak around 500 but decreasing soon after: its distribution still mainly comprised Cyprus, parts of southern Turkey and the southeast Mediterranean. LRC went through a marked increase during the second half of the 5th century, and reached a peak in the early 6th century.

1033 

1029  1030 

127

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:39 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East

Figure 101. Map showing the distribution, in absolute numbers, of ARSW, LRD and LRC during phase 7 (c. 425/450-500/525) (© Philip Bes/ICRATES Project).

In accord with the observations by Hayes, LRC began to appear in the 5th century, and virtually all LRD belonged to the period 450-550.1039 This could very well mean that the period around 400 was a turning point in the distribution and use of imported tablewares. The geographical and quantitative distribution of ARSW contracted significantly during phase 7, although specimens still travelled deeply inland, for example to Halabiyya/ Zenobia and Resafa.1040 The quantitative decrease is also seen in Figure 58: ARSW, after dwindling early in the 5th century, further decreased into the early 6th century. The combined evidence shows that the geographical reduction was mainly quantitative; ARSW is not attested at fewer sites (68 here, 69 in phase 6). This decrease also occurred when LRD and LRC considerably expanded. ARSW was relatively quickly ousted by LRC in the Aegean-North and -East, where during phase 7 ARSW was clearly reduced to only the occasional arrival. ARSW also gave some way to LRC on Crete and in the Aegean-Southwest. Lund 1995. The rather wide date ranges of certain forms do not allow pinpointing a more precise date of arrival.

1039  1040 

In summary, Figure 58 bears out the geographical and quantitative increase of LRD and LRC. The increase of LRD was roughly restricted to the coastal zones closest to Cyprus. LRC, on the other hand, increased strongly and now encompassed a large part of the eastern Mediterranean. ARSW, however, experienced a period of diminishing quantitative distribution, a change that had set in around 400, thus at least about a generation or so prior to the Vandal invasion of Africa Proconsularis. 5.4.4 Phase 8 (c. 500/525-575/600) ARSW went through a second phase of growth that encompassed the second and third quarters of the 6th century. LRD went through a similar development: a second (?) phase of growth that started toward the mid-6th century, reaching a peak toward the end of that century. LRC markedly decreased after its peak in the early 6th century, although it increased again in quantity after the mid-6th century to reach a second, albeit lower peak in the years around 600. The dispersion of LRD, which expanded essentially in quantitative terms, did not change too much compared to 128

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:39 UTC All use subject to https://about.jstor.org/terms

Framing Late Hellenistic and Early Roman Tablewares

Figure 102. Map showing the distribution, in absolute numbers, of ARSW, LRD and LRC during phase 8 (c. 500/525-575/600) (© Philip Bes/ICRATES Project).

phase 7 (Figure 102). It was dominant on Cyprus and at Anemorion; this was presumably more generally the case for Cilicia Tracheia and the coastal zone in the direction of Perge. LRD was also still found in the coastal Levant (e.g. Sumaqa), yet in the Levant-Coastal South it appears to have given way (slightly) to LRC; Hayes considers LRD to have been ‘temporarily in relapse’ during the first half of the 6th century,1041 which Figure 58 more or less bears out. At Beirut, however, LRD was the dominant imported tableware in a late 6th century context.1042 Further west, LRD continued to appear sporadically at Aegean sites (at Methana, for instance), and in Cyrenaica (Tocra, Apollonia). LRC experienced a period of quantitative and geographical growth; Figure 58 shows that LRC continued its prosperity in the first half of the 6th century (cf. supra). This growth is mainly on account of the common forms 3E and F. Yet, LRC had reached a low by the third quarter of the 6th century, after which an immediate and strong increase is noted, attaining a second though much less prominent 1041  1042 

Hayes 2008a, 86; possibly echoed in Meyza (2007, 103). Reynolds 2010, 114, table 6.

peak in the years around 600. LRC was also still common on Crete and in the Aegean-Southwest. Especially for the latter no drastic change in the distribution pattern is observed, although the regional evidence indicates that LRC was decreasing up to the third quarter of the 6th century; not long before, ARSW had again become the most common late Roman Red Slip Ware. At Athens, however, LRC ‘remains the commonest fine ware in Agora groups of the 6th and early 7th centuries’.1043 Outside the Aegean-East, LRC is found at, for instance, Tocra, Apollonia, Alexandria, Xanthos, Anemorion, Alahan, Tarsos, Küçük Burnaz and Domuztepe. A third important zone of distribution comprises Cyprus and the Levant. On Cyprus, while still second to LRD, LRC is found in large quantities at Paphos, Kopetra and Palaipaphos, and fragments have been identified in central Cyprus and prolifically at Panayia Ematousa.1044 LRC is common at Jalame and dominant at Jerusalem and Porphyreon in the Levant-Coastal South, and a single specimen was noted at, for instance, Magen, whilst in the Levant-Interior South it is attested at Philadelphia, Nessana, Rehovot-in-the1043  1044 

129

Hayes 2008a, 85. Central Cyprus: Given and Knapp 2003.

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:39 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East Negev, Petra, Upper Zohar and as far as the south Sinai. The evidence obtained from the northern Levant creates no clear picture. LRC is identified at Epiphaneia, Antioch and Gindaros, and travelled great distances to reach Resafa (where it is particularly common) and Athis. ARSW witnessed a peak (Figure 58) during the second and third quarters of the 6th century. Following a low in the third quarter, strong growth began so as to reach a peak in the early 7th century. ARSW is found throughout the eastern Mediterranean, yet certain parts relied much more on tablewares such as LRD and LRC. ARSW is found in Cyrenaica and on Crete, where ARSW more than doubled by the later 6th century. ARSW also became more common again in the Aegean-Southwest after the mid-6th century; at Athens, for instance, ARSW became ‘relatively plentiful again’, yet had to put up with a second position after LRC.1045 At Constantinople, ARSW was about as common as LRC during this period, though quantities from St. Polyeuktos are small.1046 ARSW now seems to cover a greater part of the Aegean, though as during phases 6 and 7, LRC accounted for the majority of the tablewares in the Aegean-East. The evidence from Asia Minor-South Coast and Cilicia provides a mixed picture, yet the relative scarcity of evidence, especially for the former, does not allow us to sketch a detailed picture. ARSW occurs at Perge, Anemorion, Kelenderis, Alahan and Tarsos.1047 In the northern Levant ARSW has been found at Antioch, Gindaros, Epiphaneia, Zeugma, Resafa, Halabiyya/ Zenobia and Athis. ARSW is not very common at all in Cyprus and the southern Levant. ARSW thus arrived in the latter regions, thereby occasionally travelling inland, but generally ARSW only came third after LRD and LRC. 5.4.5 Phase 9 (c. 575/600-700) ARSW gradually disappeared from the eastern Mediterranean, and LRD dwindled after its late 6th-, early 7th-century peak. LRC also gradually disappeared, even if it remained exceptionally strong in the eastern Aegean. Despite the fact that the 7th century marks the end of the widespread distribution of the late Roman Red Slip Wares, late forms of all three are attested across the eastern Mediterranean. In addition, production and distribution may have continued on a regional scale into the early 8th century.1048 During the late 6th and 7th century (Figure 103), the workshops manufacturing ARSW, LRD1049 and LRC were still active, yet with an even more limited morphological spectrum in comparison to phase 8. Figure 58 shows that all three wares witnessed a period of growth towards the Hayes 2008a, 73, 85. Hayes 1992. 1047  Kelenderis: Zoroğlu 2005. 1048  Bonifay 2003, 128, 2005, 570. For a period of transition in Syria and Jordan, see Villeneuve and Watson 2001. 1049  The Well Form, of 7th-century date, was introduced by Williams (1989). Hayes now regards the appearance of his form 10 to have occurred around the mid-6th century (2008a, 86). 1045  1046 

later 6th and/or early 7th century, all reaching a peak in the decades around 600. Never before did all three wares grow simultaneously. The distribution of LRD continued more or less along previous outlines, although its presence in the LevantCoastal South withdrew, even if it still occurred in considerable quantity at, for example, Sumaqa and Shiqmona. It appears that LRD is now found at more sites beyond the traditional distribution pattern, even if this usually pertains to one or a few specimens. Although LRD generally remained the main tableware on Cyprus, especially western Cyprus, LRC and ARSW are not uncommon on its eastern half: both are common at Salamis (where quantities are in fact low) and Panayia Ematousa. LRD continues to be found in the southern Levant, even reaching Upper Zohar and the southern Sinai. Further north it is noted at Antioch, and a few specimens were noted at and around Resafa. In the LevantCoastal North, however, not a lot changed during phase 9, as ARSW and LRC remain the major imported late Roman Red Slip Wares. LRD continued to be the main imported ware at Anemorion, is common at Kelenderis, and a single specimen reached inland Porsuk; less than a handful of specimens are identified at Perge and Xanthos. At Elaiussa Sebaste, LRD and LRC were apparently not too different between the mid-6th and -7th century, yet LRD acquired a ‘distinct dominance’ in the second half of the 7th century.1050 Small quantities of LRD were noted at Apollonia, Berenice and Tocra, yet it is more common at Alexandria. Finally, LRD appears to be considerably more common now in pockets in the Aegean; notable quantities have been found at Emporio and Constantinople. Following the steep drop toward the mid-6th century, LRC also enjoyed a period of increased distribution, even if the overall quantity now circulating was considerable lower. More so than for LRD, the distribution of LRC changed considerably: it is essentially quantitative, and can be observed in different parts of the eastern Mediterranean. Late LRC forms are still found in Cyrenaica, and a high number of form 10 was noted at Gortyna, where ARSW is nevertheless more common. In the Aegean, the distribution of LRC changed considerably. Whereas it continues to dominate the Aegean-East, the quantitative distribution is now reduced to a few specimens per site; possibly, the several hundreds of specimens of form 10 that were noted at Emporio echo a more realistic distribution scenario for parts of the (eastern) Aegean. LRC also generally decreased in favour of ARSW in the Aegean-Southwest, where an interesting pattern can be distinguished.1051 Whereas most sites in southeastern Greece still received late LRC forms, including Athens where ‘what evidence is available [...] suggests that the ware remained popular’,1052 these are absent at Kenchreai and virtually absent in

Ricci 2007, 175. Bes forthcoming a. 1052  Hayes 2008a, 85. 1050  1051 

130

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:39 UTC All use subject to https://about.jstor.org/terms

Framing Late Hellenistic and Early Roman Tablewares

Figure 103. Map showing the distribution, in absolute numbers, of ARSW, LRD and LRC during phase 9 (c. 575/600-700) (© Philip Bes/ICRATES Project).

Boeotia.1053 At Corinth, late LRC forms are found but LRC is generally superseded by ARSW. At Constantinople, LRC is about as equal as ARSW, and a single specimen has been identified at Stobi. Eastwards, it appears at Xanthos, and Hayes form 10 makes sporadic appearances at Anemorion, Alahan, Diokaesareia and Domuztepe; recently published evidence from Elaiussa Sebaste shows that form 10 also occurs there.1054 On Cyprus, late LRC forms are more common although LRD clearly dominates. A substantial number was found at Kopetra, and specimens of form 10 were identified at Paphos, Palaipaphos, central Cyprus and Panayia Ematousa. LRC form 10 seems scarce at Alexandria, where ARSW and LRD dominate. Finally, in the Levant, LRC (form 10) still arrived yet a dramatic withdrawal can be observed, particularly in the southern half (Figure 88). During phase 9, quantities dropped for Jerusalem and Upper Zohar, at the latter in favour of ARSW; it is further found in south Sinai, Petra, Samaria-Sebaste and Hippos. In the Levant-Coastal North, LRC is also still found: a new findspot is Déhès in the Syrian limestone massif. The northern coastal Levant is 1053  1054 

See n. 836. Ricci 2007, 175.

in marked and intriguing contrast with the Levant-Interior North. Also during phase 9, LRC clearly is the imported tableware par excellence, although at Resafa ARSW is now slightly more common. Because of the considerable differences between the coastal and interior northern Levant, the LRC finds from Zeugma could be helpful: one would expect the region around Resafa to be supplied by means of the Euphrates, for which Zeugma was a possible point of redistribution. ARSW also increased during the late 6th and early 7th century, and reached a second peak around 600. ARSW gradually disappeared during the 7th century, and its distribution possibly contracted to parts of central North Africa into the early 8th century.1055 ARSW is the major tableware in Cyrenaica, and at Gortyna. In the AegeanSouthwest, ARSW dominates at sites away from southeast Greece, though probably not Athens. Less than a handful of ARSW fragments were noted at Stobi, and it was about as common as LRC at Constantinople. In the Aegean-East, the largest quantity was found at Emporio, where with great distance it came second after LRC. No more than a 1055 

131

Bonifay 2003, 128, 2005, 570.

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:39 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East few specimens have been identified at Ephesos, Sardis and Didyma. It is further found at Perge, Anemorion, Alahan, Tarsos, Diokaesareia, Küçük Burnaz and Domuztepe; on Cyprus, ARSW is found at Paphos, Palaipaphos, Kopetra, Salamis and Panayia Ematousa, though LRD is more common now compared to phase 8, including eastern Cyprus. In Egypt it appears in roughly similar quantities as LRD; as the main evidence derives from Alexandria, these proportions can only be seen as indicative.1056 In the southern Levant, ARSW is, amongst others, noted at Gadara, Hippos, Caesarea Maritima, Jerusalem, Upper Zohar (where it in fact exceeds LRD and LRC), Petra and the southern Sinai. In general, except for the small

settlement at Upper Zohar, published quantities are small, not surpassing ten specimens. In the northern Levant also not much changed; at Antioch ARSW remained dominant. ARSW also reached Halabiyya/Zenobia, Athis and Resafa and surroundings (where a marked increase can be noted) on or near the Euphrates. All in all this shows that ARSW still reached both larger and smaller, as well as coastal and inland, settlements. Although the published record inevitably contains (considerable) gaps, the data at hand captures certain (substantial) changes during the 3rd through 7th century Chapter 6 seeks to explain these changes.

1056  Rodziewicz (1976) provides only indications for the relative presence of forms, rather than quantified data.

132

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:39 UTC All use subject to https://about.jstor.org/terms

Archaeopress

Chapter Title: Interpretation of the Quantitative, Chronological, and Geographical Evidence for the Mid-to Late Roman Period Book Title: Once upon a Time in the East Book Subtitle: The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East Book Author(s): Philip Bes Published by: Archaeopress. (2015) Stable URL: https://www.jstor.org/stable/j.ctvr43kch.11 JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected]. Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at https://about.jstor.org/terms

Archaeopress is collaborating with JSTOR to digitize, preserve and extend access to Once upon a Time in the East

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:41 UTC All use subject to https://about.jstor.org/terms

Chapter 6

Interpretation of the Quantitative, Chronological, and Geographical Evidence for the Mid- to Late Roman Period Introduction1057 As was discussed in chapter 5, ARSW was probably the most widely distributed late Roman Red Slip Ware, and scholarly research continues to provide a progressively better insight on its key aspects: the typo-chronological framework, distribution patterns, and the concept of manufacture. Ceramic manufacture in Africa Proconsularis involved tablewares, amphorae, and cooking wares as well as categories such as lamps and architectural ceramics.1058 During the Roman imperial period pottery production was not tied to one place, but shifted within Africa Proconsularis (Figure 7). That large quantities of, amongst others, ARSW were shipped over great distances and are found throughout the (eastern) Mediterranean is remarkable in itself. A notion can be obtained concerning the proportions and general development of each of the three main wares by collecting available evidence en masse. It was pointed out that [1] considerable differences existed in the development of each ware; that [2] site-based proportions within a region presented differences and similarities; and [3] that the proportions for each region show to what extent a region agrees with its neighbouring regions, or for that matter the eastern Mediterranean at large. These differences, in principal, seem to reflect the site, regional and/or supraregional level. The nature of influencing factors was probably wildly diverse, and their effect(s) on production and distribution could have been both direct and indirect, and also the degree and length of the effects could have ranged from profound to superficial. Furthermore, if Fulford and Parker1059 are basically right in supposing that tablewares constituted only a minor part of a ship’s cargo, if at all always present, and rode piggy-back, the question presents itself to what extent tablewares were a structural part of (certain) exchange patterns. Recently, Bonifay asked the important question whether ‘[l]a céramique africaine’ can be used as ‘un indice du développement économique?’;1060 in other words, whether ARSW can be used as a proxy for general economic development(s). Evidently, as elsewhere, Bonifay does not limit himself to ARSW, but incorporates other categories of African pottery into his discussion, notably the amphorae that were far more important economically. By rephrasing Bonifay’s question, we can explore whether or not, through mapping the regional and supraregional tableware evidence, it is possible to see that Condensed versions of this chapter have been published: Bes and Poblome 2008, 2009. 1058  See Bonifay 2004, e.g. 23-25, 54, figs 9-11, 24; Bonifay et al. 2012. 1059  Fulford 1987; Parker 1992. 1060  Bonifay 2003.

evidence as a general index of exchange patterns in the eastern Mediterranean during the mid- and late Roman period. If so, what mechanisms and factors underlie these distribution patterns? 6.1 Interpretation of the Collected Evidence ‘Crisis’ is the word traditionally used when the 3rd century is discussed. During part of the 3rd century, emperors generally did not occupy the imperial seat for more than a few years. The Empire had to cope with invasions by Persians in the east, Goths from central Europe and similar upheavals in parts of North Africa. Furthermore, both in Gaul and in the east there were militant movements which attempted to throw off the yoke of Roman occupation or control. The devaluation of money and the increasing popularity of Christianity also further characterised the period. This historical tradition of crisis, however, has been challenged from diverse angles in more recent years. Possibly one of the more important considerations is that not all parts of the Empire were (equally) affected, and those that were may have experienced any troubles of the time differently.1061 The 3rd century also seems to have been a period of change on more modest levels. The pottery manufactories of southern Gaul declined towards the later 2nd century, perhaps the early 3rd century, while in the east, ESA, ESB and ESD disappeared from supraregional exchange patterns, even if production and distribution (likely) continued on a reduced scale. The remaining major manufactories in the east were Sagalassos and the Pergamon region. These, however, for one or more reasons did not or could not fill this gap. Thus, during the 3rd century the eastern Mediterranean appears somewhat bleak in terms of the distribution of supraregional tablewares. The extent in which settlements fell back on locally or regionally manufactured tablewares is thereby poorly understood. Whether the gradual increase of ARSW in the course of the 2nd and 3rd centuries reflects such wider developments remains to be seen; clearly ARSW did not become common in the east prior to the mid-3rd century. One has to bear in mind, however, that ARSW was more common at Berenice (and Sabratha) during the later 1st to (earlier?) 3rd century. This tentatively indicates that its geographical distribution gradually expanded, with Cyrenaica being the eastern limit of this (early) distribution phase.

1057 

1061  Pieri 2005, 146; Poblome 2006a; also Southern 2001, 1-13 who rather sees the 3rd century as a period of change.

133

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:41 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East 6.1.1 The Distribution of ARSW in the 3rd Century ARSW, sporadically attested in the east prior to the 3rd century, may in fact have met part of the demand for quality tablewares. The available evidence indicates that ARSW became gradually more common towards the mid-3rd century, a development that had already come to fruition by the years 240-50.1062 At Athens, in any case, ESC had to give ground around this time to locally-manufactured tablewares as well as ARSW.1063 Early ARSW forms are only sporadically attested at eastern sites, a situation which changed with 2nd and early 3rd century forms. These are attested at major urban centres such as Corinth, Ephesos, Paphos and Antioch, though not exclusively. Moreover, many sites where ARSW forms of this period are attested are coastal sites, which could reveal the major shipping lanes into and out of the eastern Mediterranean; the distribution of the earliest (and the latest) ITS in the east is thought of in similar terms. Evidence from eastern sites remains scarce but if more could be found to substantiate Hayes’ hypothesis, this would imply that at least part of the exchange networks continued to operate in the 3rd century. Whether ARSW was deliberately used to replace part of the eastern demand for tablewares, following the withdrawal of ESA, ESB and ITS, remains to be seen. After his death, the emperor Septimius Severus ‘left a surplus of grain to the amount of seven years’ tribute, or enough to distribute seventy-five thousand pecks a day, and so much oil, indeed, that for five years there was plenty for the uses, not only of the city, but also for as much of Italy as was in need of it’ [sic].1064 From an archaeological and economic point of view, Mattingly argued that the number of presses in Tripolitania and Africa increased in the course of the 2nd century.1065 This development may have started already by the late 1st century, and is thought to demarcate growth in the surplus production of olive oil, a development that continued into the 4th century.1066 In addition, Rickman considered that of the two main suppliers of grain, Egypt and Africa, the latter had become the more important quantitatively by the late 1st century, as such contributing to the distribution of ARSW that started around that time.1067 Archaeological evidence to support this hypothesis comes from Rome’s Monte Testaccio, where Spanish and North African oil amphorae dated to the 2nd century are found, and Ostia, where large quantities of African olive oil amphorae were found, spanning the late 1st to the 4th centuries.1068 The supply of Spanish olive oil to Rome, however, began to dwindle Hayes 1972, 419; Reynolds 2010, 111-112, tables 1-2a, for Beirut: ARSW comprised c. 51.5% around 230-250, c. 91.4% by 325-350, and 76% by the late 4th century. This fell back to c. 17% by the early 5th century. 1063  Hayes 2008a, 52, 71-72. 1064  Scriptores Historiae Augustae, Septimius Severus XXIII (427). 1065  Mattingly 1995, 581-582: ‘[t]he African share of the Mediterranean market of olive oil built up to massive proportions in the period from the second to fourth century AD’; also see Peacock et al. 1990, 83. 1066  Hitchner 1993, 502-506, esp. 502. 1067  Rickman 1980, 231. 1068  Hitchner 1993, 500.

by the later 2nd century; after the early 3rd century, Africa became a/the major supplier of olive oil within the annona framework.1069 The chief consumer markets of these products were western, amongst others those involved in the state-driven annona civica (the city of Rome) and the annona militaris (the army). These products do not seem to have had a major quantitative importance in the east, even if Baetican Dressel 20 oil amphorae from time to time do turn up at eastern sites.1070 The background of these eastern finds remains unclear: these may have ‘escaped’ the annona framework, or had, perhaps, always been commercialised goods. It is interesting to see that the quantities of ARSW supplied to Berenice (and Sabratha) began to diminish rapidly after the early 3rd century. Possibly, the supply of ARSW to these sites was diverted in favour of the supply of olive oil (and other goods) to Rome, which indeed, as Hayes mentioned, continued to receive vast quantities of ARSW during the 2nd and 3rd centuries, together with Ostia and sites in the South Etruria survey.1071 On the other hand, Sabratha and Berenice may have found an acceptable surrogate for African tablewares. A further point of interest is that ARSW became rather common in the east during the course of the 3rd century, starting c. 240-250. Whether the actual production output increased so that more could simply be distributed is a matter beyond the current scope. It is interesting nevertheless to note that this occurred around roughly the same time that African agricultural products became increasingly important to Rome, and when smaller quantities of ARSW arrived at Sabratha and Berenice. Some (indirect) evidence is available to support the argument that exchange to and within the east became partly a state affair by the mid-3rd century, perhaps even earlier, or at least that contacts with specific relevance to the east increased. First, ARSW arrived at Dura Europos by the mid-3rd century, and possibly contemporarily also at Zeugma; these sites were located in frontier zones, as such were important to the Empire’s (military) control, and thus possibly the focus of increased official supply. This is not to say that ARSW was only or specifically shipped to military sites. Secondly, Pisidia experienced a ‘second wave’ of Romanisation during the 3rd century: Pisidian cities became involved in the supply of soldiers and goods to the military via Pamphylian harbours such as Perge and Side, that had their significance in shipping lanes between west and east.1072 This seems to comply with the socioeconomic development of landowning in the 3rd century (cf. infra). On a more modest level, yet not necessarily less significant, is the probable non-coincidental resemblance

1062 

Pieri 2005, 146; Mattingly 1995. This process had started already earlier: Harris 2011, 164. 1070  Bonifay 2005, 572, on the scarcity of African amphorae in the east, perhaps even prior to the second half of the 5th century (574): ‘[t]out au long des IIe et IIIe s., la présence des amphores africaines classiques reste assez anecdotique’. It is the author’s impression, however, that in some regions (Boeotia, for instance, for which see Bes accepted b) the presence of African (Tunisian and Tripolitanian) amphorae can be described as not uncommon. Much more quantified evidence is nevertheless required. 1071  Hayes 1972, 416. 1072  Mitchell 1999; Cavalier 2007. 1069 

134

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:41 UTC All use subject to https://about.jstor.org/terms

Interpretation of the Quantitative, Chronological, and Geographical Evidence for the Mid- to Late Roman Period between ARSW form 50(A) and SRSW form 1B101.1073 Moreover, African products remain archaeologically rare or no longer traceable in the east for this period. Another explanation is the role played by Alexandria, from where Egyptian grain was shipped to Rome-Portus within the context of the annona. Portus, where goods arrived from across and beyond the Mediterranean, acted as a centre of redistribution, which created opportunities for (a range of) goods for the return voyage eastwards. In Egypt, Alexandria acted as another centre of redistribution, and ships probably touched at intermediate harbours, for instance to take on fresh water: whereas this is no proof whatsoever, such systems that encompassed the (entire) Mediterranean provided sufficient opportunities for goods to attain a wide — whether thin or not — dispersion (in the east). Recently, Abadie-Reynal highlighted a scenario to explain the demise of ESC and the consequent growth of ARSW during the mid-3rd century, perceiving that ARSW had certain intrinsic advantages over ESC. On a broader scale, Greece had been importing grain (and other agricultural produce, attested through amphorae) from Asia Minor, a supply that became disrupted during the course of the 3rd century. If Africa did indeed take over the supply of grain (hypothesised to have occurred only in, or mostly after 330, cf. infra), and only grain in light of the supposed scarcity of African amphorae in the east during this period, in order to explain the growing distribution of ARSW, it nonetheless ignores the role that Egypt could have played.1074 Does the supposed disparity between ARSW and African amphorae in the east, otherwise plentiful in Rome as a result of the annona, imply that African grain and ARSW (and amphorae?) were shipped eastwards (mostly) as commercial cargoes? The distribution pattern of ARSW of this period, notably represented by Hayes forms 45A-B and 50A, is extensive yet still rather thin. It reached sites across the eastern Mediterranean, even as far inland as Petra, Dura Europos and Zeugma. Unfortunately, the context of many of these finds is not or at best poorly known, which consequently hampers a more precise determination when the inflow of ARSW actually started; we should think that this was probably dissimilar for different parts of the east. These developments fit the perspective that is considered more and more appropriate to approach the ‘crisis’ of the 3rd century, and partly contrasts the opinion expressed by Southern: ‘[t]he decline of trade in the third century adversely affected the urban centres. Long-distance trade diminished sharply, one reason being endemic insecurity of the land and sea routes’ and ‘[t]he old system of converting surplus produce into money via the urban markets had been radically altered when tax payments were converted from monetary mode to payment in kind; the urban markets depended on an interchange of cash payments for various goods. As self-sufficiency became the order of the day the urban markets were perhaps replaced by rural markets, some of them centred on villages, others on Poblome 1999, 292-293, 296, 305, 355-356 (figs 19-20). 1074  Abadie-Reynal 2007, 172-173. 1073 

large estates. The polarisation of wealth that concentrated most of it in the hands of the owners of these large estates ensured that economic and political power accrued to the landowners’.1075 This suggests that supraregional trade had become a system stripped to its bare essentials, focussing on the (strictly) necessary exchange and supply; this likely included the military. By the late 3rd century the situation in the Roman Empire had returned to relative calm. Imperial administration had regained a sense of continuity, and actually went through a major reorganisation. Military unrest and wars both within and against the empire had been settled, and the financial difficulties had been handled by Diocletian; his well-known Price Edict served to control the inflation that had troubled the Empire.1076 The 4th century began more peacefully. 6.1.2 After the early 4th Century ARSW became more common during the course of the 3rd century, even if it remained a rather modest presence. After the early 4th century, a twofold change can be observed in its distribution: 1) A quantitative growth, roughly spanning the second half of the 4th century; 2) Intensification of the geographical distribution of ARSW into non-coastal areas (though note that specimens already arrived at inland sites previously, e.g. Dura Europos). Compared to the 3rd century, this implies a fundamental shift in the supply of ARSW, since Sabratha, Berenice, Rome and Ostia now received ARSW only in modest quantities. This major shift in the distribution of ARSW reflects what should probably be seen as the most decisive element: the founding of Constantinople in 330. Although the actual founding of Constantinople probably acted only as a catalyst (since ARSW had already begun to become more common in the east in the mid-3rd century) and did not affect the distribution of ARSW in direct terms, it nevertheless designated a major shift in the orientation of economic (and other) patterns that had developed in previous centuries. The new capital implies the emergence of a new imperial pivotal point in administrative, political, military, cultural, diplomatic, and economic terms, and it is in this broad context that the reorientation in the distribution of ARSW during the second half of the 4th century needs to be seen. Constantinople thus attracted, as Rome had done during the late Republic and early Empire and to a certain extent continued to do so, a wide range of goods from across its territories to fulfil the needs and desires of its population, which included not only the rich(er) échelons of society, 1075  Southern 2001, 267. For a partly contrasting ceramic view, see Reynolds 2010, 98. 1076  Pieri 2005, 146-147.

135

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:41 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East but also a sizeable share of the population which had to be fed. Constantinople’s hinterland may have been unable to answer this demand: it had to obtain primary foodstuffs (grain, olive oil, wine, amongst others) from elsewhere. Rome had for example procured Egyptian grain, its most important source, and African olive oil, in previous centuries. Pieri visualises a major shift from 330 onwards, when Egyptian grain now supplied Constantinople, whilst African grain was shipped to Rome; in years of famine or crop failure it could call on Egypt, as well as other regions.1077 Under normal circumstances, this division of the grain supply would thus be institutional.1078 And what are we to make of grain from Cyrene, or the Black Sea?1079 It is hard to imagine that the capital, literally on the threshold of the Pontic basin, ignored this potentially important source. Actually, this notion might help to explain the Pontic distribution of Phocaean LRC — and evidently that of many other goods. This has recently been challenged by Bonifay, who does not deny that ‘dès avant 330, le blé d’Égypte a été détourné de Rome pour alimenter la future ville de Constantinople’.1080 Explaining the decrease of ARSW in the east after the Vandal conquest, which in fact starts prior to the Vandal conquest, he conjectures whether ARSW was not shipped along with African grain. This premise is based on the scarcity of African ceramic products other than ARSW during Late Antiquity, except for spatheia that were not uncommon in the east between the late 5th and 7th century. The important consequence of this hypothesis is that Africa, next to Egypt, could also have acted as a supplier of grain at least up to the Vandal conquest. The precise role of the African grain (requisitioned by the state as part of the annona?; resorting to Africa in periods of insufficient crop yields?; buying grain commercially?) is not clear. It is clear, however, that ARSW and presumably African (agricultural) products increased in importance during the second half of the 4th century. Whether that was grain, olive oil, fish products or any other product, and whether it was shipped within official, private and/ or commercial modes of exchange, the bottom-line is that the founding of Constantinople affected the exchange mechanisms with/of Africa. Interestingly, the Theodosian Code includes specific regulations that forbade other goods to be brought along with officially-requisitioned cargoes.1081 To what extent this explains the paucity of ERSW and LRD (if Egyptian grain ships indeed passed Cyprus on their journey westwards) in the Aegean cannot be made out.1082 If these regulations were indeed observed, and if African cargoes were (partly?) also state-requisitioned, this does not Pieri 2005, 148. Rickman 1980, 198-199. 1079  Rhodes and Osborne 2003, 486-493, no. 96; Braund 2007. Both pertain to pre-Roman periods. 1080  Bonifay 2003, 120. 1081  Pharr 1969, 399, book XIII.8. 1082  Rautman 2003, 242: ‘[i]t is understandable that few examples of Egyptian Red Slip and LR7 amphoras reached the valley, even though [Cypriot] fine wares and amphorae are abundant in Egypt’. 1077  1078 

make it easier to interpret the widespread abundance of ARSW in the east. Did this result from more commercial shipping?; were these regulations not observed?; was ARSW a product in demand on its own terms? The founding of Constantinople contributed indirectly to the sizeable increase of ARSW from the mid-4th century onwards. Yet, this peak was only a relatively short-lived period, as by the very end of the 4th century quantities began to diminish. This decrease was not insignificant, which suggests that smaller quantities of ARSW flowed into the eastern Mediterranean. If indeed tablewares serve as a measure for the size and volume of the economy, this implies that the period of intensified economic contacts between Africa and the east were relatively short-lived; surely matters were more complicated than this. First, the dwindling quantities of ARSW in the east need not per se signify that lesser goods were drawn from Africa. It surely is not coincidental that by the late 4th century, if not somewhat earlier, the workshops manufacturing LRD and LRC were (fully?) operational. It has been suggested that ARSW form 84 in particular provided decorative and/ or morphological inspiration to LRD form 2 and LRC form 3.1083 Such inspiration, be it directly or indirectly, will have taken place more often, yet requires a much better understanding of the production, infrastructure and typo-chronology (of LRD and LRC). Especially LRC may have been reaping the benefits: it is not unlikely that the proximity of its area of manufacture to the new capital even prompted or stimulated its (initial) production and distribution. Secondly, LRC attained a slim yet wide distribution even during its first period of production. LRD quickly reached several overseas markets, yet its distribution was geographically more limited. Assuming that this represents competition between these three wares is overlooking the fact that these wares supposedly were of secondary (or tertiary) importance economically, and thus were rather not influential factors. On the other hand, in an economy of substitution, increasing the production of tablewares in the wake of (intensified?) agricultural production (or other raw materials or goods) results in the demise of another tableware/product.1084 Thirdly, agricultural products from eastern regions became more common: from the mid-4th century onwards these were carried in Riley’s characteristic LR Amphorae. This could represent standardisation in packaging agricultural produce instigated by the state; agricultural production in the Levant is argued to have intensified.1085 Fourthly, it is essential to characterise the products carried by these eastern amphorae, so as to determine to what Bonifay 2005, 577-578, fig. 3. Poblome 2004a. 1085  Riley 1979; Rautman 2003, 256, concerning the 5th-6th centuries, with references; Reynolds 2005b, 575-579. 1083  1084 

136

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:41 UTC All use subject to https://about.jstor.org/terms

Interpretation of the Quantitative, Chronological, and Geographical Evidence for the Mid- to Late Roman Period extent alternatives were sought for products from Africa and elsewhere. Figure 100 (phase 6) holds certain aspects that could reflect these changes. First, LRC was common in the AegeanEast, in Gortyna and in Berenice. Since the latter two sites are comparatively well-published, this could indicate that these (or their areas more generally, for that matter) were located along the sea route(s) that connected Africa and the Aegean, with ships carrying LRC and other goods on the return journey (LRC plausibly not reaching Carthage and Sabratha,1086 with noted exceptions); the presence of LRC in the east is to be seen similarly (cf. infra). Several different routes may have existed for the voyage out of Africa: one linked up with the Aegean-Southwest, either via the Corinthian Gulf or along the southern Peloponnese, which may well explain the near-dominance of ARSW during this period in this region. The Aegean-Southwest and -East present a distinct pattern during this phase, with ARSW being uncommon in the latter and LRC being scarce in the former. Further east, the evidence presents a more mixed picture: whereas ARSW continued to dominate, LRC only came third although its distribution spans a larger area. That early LRC is relatively common in Cyprus and the Levant-Coastal South could well echo an easterly route out of the Aegean; therein, Palestinian amphorae may well be one element of return cargoes.

out by recent evidence from Athens,1089 and more generally implied by the collected evidence in this study, which designates ‘a situation of gradual decline, into which the invaders marched, rather than one of sudden change’.1090 Does this relate — concerning central Tunisia — to ‘a dislocation in the [amphora] industry at some time in the early 5th century, when kilns clustering around urban centres gave way to production on rural estates’?1091 Did this shift (temporarily) disrupt tableware manufacture in its wake, not to recover until some time in the early 6th century? If indeed ARSW began to subside quantitatively several decades before 439, if not already in the later 4th century, this suggests less of a rupture than previously assumed. In this case, the causes for this decrease need to be sought elsewhere. The Vandal invasion did not leave Africa (including its economy) untouched. Figure 58 shows that ARSW basically diminished in quantity from the early 5th century on, even if one needs to reckon with possibly regionally-distinct patterns.1092 All this makes this period fascinating, but not necessarily easier to define. If the Vandal invasion was neither the initial nor the main cause of a decline (in the distribution of ARSW), its effects may have been less defining, lasting relatively briefly. The contracting distribution of ARSW coincides with phases of growth for LRD and LRC. With the question whether these could be related, the following two scenarios are plausible: 1) The distribution of ARSW diminished for certain reasons, with the resulting gap being filled by LRD and LRC; 2) The shift in exchange patterns eastwards following the founding of Constantinople created new momentum, which cranked up the economic and agricultural motor of eastern regions. As a result, the manufactories of LRD and LRC blossomed, consequently affecting the need or demand for tablewares from further afield. To centralise the debate around tablewares, however, is to ignore the wider context.

6.1.3 The Vandal Period It has been pointed out that the distribution patterns of the late Roman Red Slip Wares underwent considerable changes after the late 4th-early 5th century. First, ARSW was now distributed on a quantitative reduced scale, a process that had already begun around 400, yet geographical contraction is not noted (Figure 101).1087 LRD and LRC, on the other hand, were experiencing quantitative expansion; LRC was now very widespread, whereas LRD continued to circulate mainly in the southeast Mediterranean, only occasionally being attested west of the axis Alexandria-Perge. The distribution pattern of ARSW during part of the 5th century has often been coupled to the Vandal invasion in 439 — and the subsequent occupation of North Africa — in a rather straightforward manner, and on more than one occasion this operated as an interpretative framework, determining the extent that these events affected Africa’s economic and artisanal framework. Originally, Hayes, Waagé, Abadie-Reynal and others believed that the Vandal invasion was the/an immediate cause of the decrease of ARSW in the eastern Mediterranean.1088 On the other hand, Fulford pointed out that ARSW was diminishing already from the early 5th century onwards, that is, a generation or so prior to the Vandal invasion. This is now also borne

Fulford and Tomber 1994. Reflected in recent figures from Beirut: Reynolds 2010, 113, table 3; cf. supra, n. 1062. Also see Bes and Poblome 2009, 84, fig. 3, for a nearly identical pattern (note the difference c. 460-480). 1088  Hayes 1972, 423; Waagé 1948, 56-57; Abadie-Reynal 1989a, 150. 1086  1087 

Fulford hypothesised that the supply of agricultural produce required by the Empire continued during the Vandal occupation of Africa. Since Africa was no longer Roman territory, agricultural produce could not be requisitioned as taxes, yet it could be bought commercially. If indeed grain, olive oil and so on were sold commercially to the imperial authorities, the economic framework and the resulting inflow of wealth into Africa could be coupled to the increase of eastern amphorae at Carthage. As a matter of fact, the Empire may have increasingly resorted to eastern sources for its demands in subsistence goods, which indirectly fits the evidence for an increase in the distribution of LRD and LRC from the mid-5th century onwards. Evidence from 1089  Hayes 2008a, 72: ‘it can now be argued that the political and administrative changes in the Empire ca. A.D. 400 affected the supply of fine wares much more than did the arrival of the Vandals (as was once thought)’. 1090  Fulford 1984, 113. 1091  Peacock et al. 1990, 82. 1092  Hayes 2008a, 73.

137

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:41 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East the Aegean supports a shift in exchange patterns. At Argos, by the mid-5th century c. 20% of the imported amphorae were of eastern provenance, predominantly LRA 1, LRA 4 and LRA 5, whereas relative quantities of the Aegean LRA 3 diminished during the second half of the 5th and early 6th century.1093 At Argos, the main period of import of eastern amphorae spans the second half of the 5th and the first half of the 6th century, roughly contemporary with the major export phase of LRC.1094 This pattern seems to recur at Corinth, where an increase in the variety of Palestinian amphorae was noted by the late 5th century.1095 Concerning the tableware evidence, first, ARSW remained rather common in Africa-Cyrenaica, Crete and central and southern Greece, but in comparison to phase 6 clearly had to put up with a second place. By this time, however, LRC had clearly gained a considerable share of the market, now almost the sole (known) tableware in the Aegean-East. Although ARSW and especially LRC are common at a number of sites in southern Asia Minor, Cilicia Tracheia, Cyprus and the Levant-Coastal South, LRD dominated, most likely reflecting direct, regionalised exchange between Cyprus and its closest coastal zones.1096 At sites along Cyprus’ west coast, however, LRC seems to be marginally more common during this phase even if LRD prevails. Cyprus and the southern Levant (Figure 101) show a considerable degree of similarity in the proportions of ARSW, LRD and LRC. The dominance of LRD reflects the axis between both regions, possibly partly connected through Paphos and Caesarea Maritima.1097 The LRC (rather common) and ARSW (small quantities) in both regions were carried by ships from further west that, given the scarcity of LRC at Perge and Anemorion, possibly departed from the Lycian coast, although the role of one or more other redistributive centres (e.g. Alexandria) cannot be ruled out. In the southern Levant-Coastal South, although published evidence is scarce, a different situation appears. LRD is near-absent, and ARSW is the most common late Roman Red Slip Ware, which provides an interesting connection to the peculiar pattern concerning Egypt, which was observed by Hayes, Fulford, Bonifay and others.1098 Egypt continued to receive ARSW of the second half of the 5th century in comparably larger numbers than elsewhere. Although located inland, at Hermopolis Magna about half of the catalogued ARSW comprises Hayes form 84 (440-500), a form otherwise not very common in the east; at Kôm el-Dikka in Alexandria, form 84 was a rare appearance.1099 The southern Levant-Coastal South possibly is the easternmost stretch of this Egyptian pattern. Egypt surely was principally not in want of grain, yet olive oil had to be imported: Bonifay states that those Abadie-Reynal 1989a, 151, 2001, 285-286. Abadie-Reynal 2001, 284-287. 1095  Slane and Sanders 2005, 285-286. 1096  Rautman 2003, 242-243, 253, 258. 1097  Rautman 2003, 242-243, 253, 258, who sees Salamis and Kition as primary beneficiaries, whilst Paphos was reached by coastal trade given the building materials at Kopetra that were manufactured at Paphos. 1098  Hayes 1972, 423; Fulford 1984, 259; Bonifay 2005, 568-569. 1099  Bailey 1998, 2; Rodziewicz 1976, 37: ‘[r]are à Kôm el-Dikka’. 1093  1094 

amphora types for which olive oil has been determined as their primary content ‘se rencontrent principalement à des époques hautes (Tripolitaine I) et/ou dans les zones géographiques ne produisant pas ou peu d’huile d’olive (Égypte).’.1100 Egypt is considered as the chief supplier of grain to meet the imperial needs in feeding Constantinople as well as provisioning the military forces. Routes taken by these ships ran either via Cyprus, or along the African coast westwards, reaching the Aegean via Crete. This was probably the continuation of a system that had been in operation at least since the 2nd century BC, to which century dates the oldest (written) evidence regarding grain supplies to Rome.1101 Ships leaving the Aegean on their return journey, which may well explain part of the distribution of LRC and ARSW, likely also carried goods that were specifically required in Egypt: in this way (African) olive oil may have found its way to Egypt. 6.1.4 The Later 5th to Mid-6th Century Whereas Bonifay, with Fulford and Reynolds, considers the period of c. 440-475 ‘une période de relative stagnation dans l’économie de cette région’, the very early 6th century seems to have marked a change for the better. According to Bonifay, the later 5th and early 6th century ‘pourraient marquer, d’une certaine façon, un retour à la situation antérieure’.1102 This change can also be partly concluded from the collected evidence (Figure 102). From around 500 ARSW visibly increases, yet is followed by a dip (during this period both LRD and LRC experience a peak); from about the third quarter of the 6th century — at the very end of the reign of Justinian (cf. infra) — quantities increase once more. This has several implications. First, this suggests that African tablewares again became more widely available. Secondly, and this is more important and shown not to stand alone, the fortunes of ARSW recovered somewhat several decades prior to the Justinianic reconquest of 533. ARSW began to dwindle about a generation or so prior to the Vandal conquest, and as the intervals are of roughly equal length, one could argue that this is largely a typo-chronological issue, and that we need to shift this forward in time on either end by about two to three decades, in order to make the evidence run parallel to the Vandal occupation of Africa. But that would be writing history the wrong way round. Moreover, whereas indeed several of Hayes’ date ranges have been adjusted, sometimes marginally, sometimes by several decades, Bonifay’s1103 typo-chronology is based to a large extent on Tunisian evidence, and should only tighten the typo-chronological framework. A very detailed interpretation based on ARSW alone seems nonetheless hazardous. All in all it thus seems that the situation of Africa was improving, at least economically: ‘[t]he coming of more peaceful conditions at the end of the fifth century may have permitted the African wares to circulate Bonifay 2005, 575, 2003, 122. Rickman 1980, chapter II, esp. 36-37; Rautman 2003, 242. 1102  Bonifay 2003, 123. 1103  Bonifay 2004. 1100  1101 

138

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:41 UTC All use subject to https://about.jstor.org/terms

Interpretation of the Quantitative, Chronological, and Geographical Evidence for the Mid- to Late Roman Period more freely again, but the main recovery occurred under Justinian, presumable as a result of his reconquest of Africa’.1104 Some time after the death of Geiserik (477), and Vandal military turmoil in the west, contacts may have bettered also diplomatically; the hostile atmosphere between the two parties seems to have settled down in the final decades of the 5th century.1105 These improving relations between Constantinople and Africa toward the later 5th century possibly resulted in increasing economic activity between the two, and the Aegean-Southwest, located centrally on the axis AfricaConstantinople, may have been the first to profit: ARSW could again gain in importance, even if Figure 67 shows this to occur already in the early 6th century.1106 The distribution of LRD and LRC continued within the existing zones of supply, even if both experienced a decrease in the mid-6th century. That the major phase of the distribution of LRC during the later 5th and first half of the 6th century coincided with the major export phase of Proconnessian marble to the east is surely no coincidence.1107 The east, however, did not remain peaceful. Although improving contacts between Africa and Constantinople probably resulted in increasing (economic) contact, the east had to cope with a series of disruptive events. There were the Persian invasions of 540-1, followed by Roman counteractions. Furthermore, in 541 and the years that followed, the plague ravaged throughout the Mediterranean, and in 542 reached Constantinople.1108 Also, some of the eastern regions and its major cities were struck by a number of earthquakes in the course of the 6th century. Although the effects of these events on the populace and the urban and rural fabric must have been considerable, for instance in the case of Antioch, it is far more problematic to infer from these events any consequences for regional and supraregional networks of manufacture and exchange. Despite the resurgence of ARSW on the eastern markets, it is generally clear that its distribution pattern of the second half of the 4th century was not attained anew. LRD had firmly established itself as a class of tablewares, admittedly in a comparatively limited area. Also LRC continued to be an important product in the east, with a distribution pattern that at its peak spanned the entire eastern Mediterranean, even spreading into the western Mediterranean in substantial quantities. It becomes more and more apparent that the second half of the 4th century, in Mediterranean history, was a period during which circumstances were exceptional, making it a unique phase with regard to Roman tablewares: ‘[m]ais c’est aussi le seul moment de l’histoire où la Méditérranée orientale est pleinement intégrée au reste du bassin méditérranéen du point de vue de ses approvisionnements en céramique fine’.1109 Hayes 1972, 423. Hornblower and Spawforth 2003, 1581. See Abadie-Reynal 2007, 174, on the 4th century. 1107  Papacostas 2001, 115. 1108  Evans 1996, 151-165. 1109  Bonifay 2005, 568; also Sodini 2000, 181; Hayes 2001, 279; 1104  1105  1106 

6.1.5 The Later 6th Century to the ‘End’ of Antiquity By the late 6th century the collected tableware evidence indicates no signs of decrease. In fact, it points out that all three wares witnessed a final phase of growth (Figure 58). ARSW shows an uneven course throughout the 6th century, as does LRD, which seems to have reached its greatest extent outside its core zone of distribution, even if the attested quantities usually do not exceed one or a few specimens (Figure 103). LRC, however, although its distribution still spanned a large part of the eastern Mediterranean, appears to have declined more rapidly after the early 7th century. In comparison to its heyday, the later 5th and first half of the 6th century, quantities reduced markedly during the mid-6th century, yet went through a last phase of growth during the second half of the 6th and early 7th century. All this generally seems to reflect Bonifay’s ‘relatif dynamisme économique’ of the late 6th and early 7th century.1110 Also here Figure 58 marks certain interesting features that could clarify aspects of the exchange of goods. It has already been noted that the Aegean-Southwest may have been the first to profit from the intensified or renewed contacts between Africa and Constantinople; this can be confirmed for this period as well, even if not all sites probably gained equally, if at all. The characteristic LRC form of this period, 10A-C, shows a particular distribution in the Aegean-Southwest. Whereas form 10A-B (c. 575625) occur regularly at, for instance, Argos and Corinth (and at Gortyna), the later variant 10C (c. 600-675) is much rarer, and appears to be absent at Argos, Corinth and Athens. No specimens of LRC form 10 or variants were recognised in the Boeotia survey (though see n. 836, cf. supra). Where attested, including the late 10C, form 10 is attested only in small numbers, and mainly restricted to the southeast (e.g. Methana, Halieis). At Corinth, and in (eastern) Boeotia, however, contemporary ARSW forms are more common, most recognisably Hayes form 105.1111 This is in marked contrast to other parts of the Aegean, where LRC continued to dominate. In the 7th century, however, things changed considerably. Whilst the AegeanSouthwest and Crete continued to receive substantial quantities of ARSW and LRC, the Aegean-East and -North on the other hand faced considerable changes (Figures 70 and 73). Whereas the supply of all three wares persisted at Emporio, possibly stimulated by its military character and location, sites on the mainland were almost bereft of their supply of the three major tablewares. That form 10 is ‘uncommon’ at Sardis1112 and ‘relatively rare’ at Constantinople is worth noting; perhaps less recognisable McCormick 2002, 55; Fulford 1984, 112. 1110  Bonifay 2003, 123-124, 2005, 571-572; Sodini 2000, 181-182; to a lesser extent see Fulford 1984, 114. 1111  Bes forthcoming a. Worthwhile is a thorough evaluation of the distribution of ARSW Hayes form 105, specifically using Bonifay’s (2004, 183-185, fig. 98) three subvariants (ST57A-C), which would refine our understanding of the latest phase of the distribution of ARSW in the (eastern) Mediterranean. 1112  Rautman 1995a, 79, table 1.

139

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:41 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East products came to meet local demands, such as Rautman’s ‘Imitation Late Roman C’ at Sardis. At Constantinople, however, both ARSW and a local tableware became more common.1113 The fact that LRC form 10C continued to arrive at inland sites as far east as Resafa (to which Hippos, for instance, can now be added) strongly contrasts with this Aegean pattern, even if the question, particularly pressing for the 7th century, when these specimens arrived exactly cannot be answered. LRD continued to dominate on Cyprus, be common in parts of Cilicia, and to arrive in the southern Levant; in Beirut it was the most common imported tableware by the late 6th century.1114 In fact, the latter region not only shows a more diversified picture compared to phase 8, in addition quantities also dwindled. The military or administrative (?) outposts at Upper Zohar and ‘En Boqeq continued to receive substantial quantities of all three wares into the 7th century, possibly reflecting a final phase of military or administrative control of the eastern limes. If that was the case, similar sites elsewhere could help clarify such a supply pattern; Emporio may have been another example in closer proximity of the capital. The Arab invasions to some degree disrupted the exchange networks that crossed the (eastern) Mediterranean. With regard to Cyprus, Rautman writes that ‘[u]rban and rural, coastal and inland, almost all of these places were touched by events that took place during the 7th century, of which the most consequential were the Persian invasions and Arab conquest of Syria, Palestine, and Egypt, followed by attacks on Cyprus itself beginning in 649. Over the span of two or three generations, the island’s political status shifted from a prosperous Late Roman province to an unstable frontier territory, balanced uneasily between Constantinople and Damascus. Military campaigns, shifting economic patterns, and other circumstances seriously affected settlement-patterns and quality of life’.1115 In the wake of the great geopolitical and military changes of the time, the exchange networks largely disappeared (quickly or gradually?) in the course of the second half of the 7th century, if not partly already earlier, and the distribution of red slip wares with them.1116

Constantinople in 330 surely was a prime cause for an important shift in exchange patterns. This predominance of ARSW perhaps marked a transitional period, but since ARSW continued to arrive in the following centuries, ties with Africa were maintained, albeit not always under the same circumstances. The shift of capital and the supposed increasing reliance of the Empire on eastern sources for the provisioning of goods and agricultural produce, for the annona civica and annona militaris, possibly partly explains the appearance or further development — major eastern amphorae of this period have predecessors1117 — of some of Riley’s LR Amphorae, as well as the grain supply from (only?) Egypt now (entirely?) directed to the new capital, but also the appearance of LRD and LRC by the late 4th century. After an initial phase, LRC especially attained a wide distribution in the east, whilst the distribution of LRD was and remained more restricted to Cyprus and its closest coastal zones. These wares, and more broadly the growth in production and distribution in the east, may have contributed to a downward trend for ARSW from around 400 onwards, so we should probably not regard the Vandal invasion of 439 as the sole cause for the (temporary) deterioration of exchange patterns and contacts between Africa and Constantinople.

As for the east, the 3rd century was a period comparatively empty of supraregional tableware distribution. ESC was mostly distributed in the Aegean, whilst after the early 3rd century ARSW gradually increased in the east. This situation changed rapidly, or at least markedly, from the mid-4th century onwards, when ARSW became widely available, and reached a geographical and quantitative peak during the second half of that century. Although not of immediate and direct influence, the founding of

Whereas part of the exchange was undoubtedly statedirected, the presence of LRD on Cyprus and the coastal zones of Lycia, Pamphylia and Cilicia Tracheia as well as the southern Levant could well indicate the continuation of embedded coast-to-coast patterns of a more commercial (?) character. State-governed exchange could also explain the Palestinian wine amphorae in the Aegean that travelled by way of Cyprus and southern Asia Minor. The reason why LRD was only rarely taken westwards may have to do with different circumstances, of which the imperial prohibition to carry goods other than the official cargoes is perhaps one. LRC, however, could have travelled along with ARSW and other goods eastwards. LRC is rather common in the Levant-Coastal South, which fits the dominance of LRC at Panayia Ematousa on eastern Cyprus. Interestingly, it is LRC and not LRD that is dominant in Cilicia Pedias and the Levant-Interior North, which from a market perspective would make more sense. This supposes a more artificial exchange network, and that LRA 1 was mostly manufactured in Cilicia (and on Cyprus from the early 6th century onwards) appears to be an important piece of the puzzle. The role of Alexandria may be seen in a similar context. Ships that provisioned the Empire with grain, carried goods on the return journey that, once arrived at Alexandria, were subsequently partly redistributed. Perhaps this function of Alexandria contributed to the continued high importation of ARSW, at a time when most if not all other parts of the east received this ware more sporadically. Other regions of importance for agricultural and/or other purposes can be seen in

Hayes 1992, 6-7. Reynolds 2010, 114, table 6. 1115  Rautman 2003, 259, emphasising, almost paradoxically, the geographically fortunate location of Cyprus. 1116  Arthur 2007, for a more socio-cultural and technological perspective.

1117  Reynolds 2005b. According to Reynolds (577) ‘[t]he needs of the annona militaris are unlikely to have been the original stimulus for the production and marketing of the type. It should be noted, furthermore, that LRA 4 and LRA 5 are relatively rare imports on the Lower Danube [...]’: the Aegean LRA 2 partly performed this role (Karagiorgou 2001).

6.2 Summary

1113  1114 

140

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:41 UTC All use subject to https://about.jstor.org/terms

Interpretation of the Quantitative, Chronological, and Geographical Evidence for the Mid- to Late Roman Period the same light. The fact that from the early 6th century onwards increasing quantities of ARSW again travelled eastwards, initially mainly to the Aegean-Southwest (with a trickling down to Constantinople?), could signify that the Empire relied on a division of sources. Although LRD continued to be common in its core zone of distribution, and ARSW in increasing quantity, in the decades around 600, LRC appears to have lost part of its distribution zone, and late forms are no longer attested everywhere, yet do occur at distant places. Finally, in the course of the 7th

century all three wares dwindled and disappeared from Mediterranean exchange patterns. Even if manufacture and distribution continued on a more reduced scale, the Arab invasions, perhaps coupled with the general state of the political, military and economic framework, eventually brought about the collapse of Mediterranean exchange patterns as they had existed for almost a millennium, and with that a connecting cultural element so characteristic of the Roman period: red slip tablewares.

141

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:41 UTC All use subject to https://about.jstor.org/terms

Archaeopress

Chapter Title: Thoughts on the Socio-Economic Significance of Red Slip Tablewares Book Title: Once upon a Time in the East Book Subtitle: The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East Book Author(s): Philip Bes Published by: Archaeopress. (2015) Stable URL: https://www.jstor.org/stable/j.ctvr43kch.12 JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected]. Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at https://about.jstor.org/terms

Archaeopress is collaborating with JSTOR to digitize, preserve and extend access to Once upon a Time in the East

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:43 UTC All use subject to https://about.jstor.org/terms

Chapter 7

Thoughts on the Socio-Economic Significance of Red Slip Tablewares 7.1 Socio-Economic Observations Pottery production is thought to have taken place on three general levels. First, that of ‘household or estate production’, archaeologically ‘the most elusive’.1118 Second, ‘professional workshops’ that mainly catered for local and (close-)regional markets. Several such wares can be tentatively listed for the period under study: Pink Ware,1119 Pink Powdery Ware,1120 Jerusalem Sigillata,1121 Kibyra Ware,1122 Athenian White-Painted Ware,1123 Macedonian Grey Ware,1124

Poblome 2006b, 353. Slane 1997, 371-375. 1120  Mlynarczyk 2005b, 2010, 362, now suggesting that ‘there seem to be little doubt that we are dealing here with a predecessor (or one of the predecessors?) of CS ware’, though Pseudo-Sigillata/Pink Powdery Ware could very well have run chronologically (partly) parallel with ESA and ESD. 1121  Magness 2005. 1122  Japp 2005, 2009b. 1123  Hayes 1972, 407-408, 2008c, 441-443, figs 6-7. 1124  Hayes 1972, 405-407, 1980, 534 (lxviii); Anderson-Stojanović 1992, 61-72.

Gryneion LRC,1125 Ephesian Red Slip Ware1126 and Asia Minor/Late Roman Light-Coloured Ware.1127 Circa 40% of the late Hellenistic to late Roman terra sigillata and red slip ware fragments from the urban survey at Tanagra remains unprovenanced. Though first-level production largely escapes us, it is clear that knowledge of regional production is equally still in its infancy. Both levels of production must have formed a plethora of workshops across the Mediterranean; only a handful or so are identified, archaeologically and/or archaeometrically. Their output must have been quite

Figure 104. The collected data concerning the late Hellenistic and early Roman eastern Mediterranean, showing core zones of distribution (© Philip Bes/ICRATES Project).

1118  1119 

Rautman 1995; Ladstätter and Sauer 2002, 2005; Vaag 2005. Iro et al. 2009, 55. 1127  Hayes 1972, 408-410, 1980, 534 (lxviii), 1992, 6-8; Domżalski 2012, 338, 340. 1125  1126 

142

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:43 UTC All use subject to https://about.jstor.org/terms

Thoughts on the Socio-Economic Significance of Red Slip Tablewares

Figure 105. The collected data concerning the mid- and late Roman eastern Mediterranean, showing core zones of distribution (© Philip Bes/ICRATES Project).

varied, dependent for instance on settlement size, its location and the time and effort (that could be) invested. The third level is believed to include the terra sigillata manufactories,1128 particularly those whose products supplied the basic evidence for this study. The outlines of two broad groups emerge based on this evidence, which strengthens the notion that reality beyond the ‘three basic levels of artisanal activity’ was indeed complex.1129 The first group includes ESB, ESC and ESD for the late Hellenistic and early Roman periods, and LRD for the late Roman period, and designates wares whose distribution was certainly substantial, yet, except for a thin scattering, clearly did not span the entire eastern Mediterranean. Still considerable, these were of more regional importance, perhaps somewhere intermediate between production levels 2 and 3. SRSW, which largely catered for the needs of a landlocked area of southwest Asia Minor should also be mentioned here; outside this area a thin yet remarkably wide scatter has been attested.1130 The second Poblome and Brulet 2005; Fülle 1997; Peacock 1982b, 114-128. Poblome 2006b, 353. 1130  Poblome 1999, 288, 339, fig. 1; Poblome and Waelkens 2003. See e.g.

group consists of wares whose distribution did encompass most or all areas of the eastern Mediterranean at some point during their period of production and distribution, if not beyond, and were clearly in a league of their own: ESA, ITS, ARSW and LRC. It is not necessarily (only) their quantitative impact which was most important, but particularly their supraregional distribution. Only a few of the major terrae sigillatae and red slip wares of the Roman World attained a supraregional distribution. Therefore, when considering the reconstruction of socioeconomic patterns, both on the level of production and exchange, the products of this second group could be regarded as exceptional. In this context, the discipline has long concentrated on the more representational end of the evidence. Even if scientific attention has made important progress, in fact, it has become more and more necessary for research to take the local, regional and supraregional levels of production and distribution into account.

1128  1129 

Gindaros, Diokaesareia, Ashkelon, Caesarea Maritima, Laodikeia ad Lycum; profiles sometimes hardly fit the SRSW typology strictu sensu.

143

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:43 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East Figures 104-105 present the collected evidence from the eastern Mediterranean for both periods studied here, showing the core zones of distribution for ESA, ESB, ESC and ESD (Figure 104, and LRD and LRC (Figure 105). Since ITS and ARSW were not eastern in origin, only the eastern dimension of their distribution can be considered here. To what extent the major terrae sigillatae and red slip wares relate to the substratum of household and workshop production in quantitative terms is a question unanswerable at present, nor anywhere in the near future, even if evidence is patchily available. The underlying factors of the exceptional context of tableware manufacture and circulation of ESA, ITS, ARSW and LRC merit a little more focus. Four mutually dependent factors could underpin a supraregional distribution. First, the ‘symbiosis between an active urban hub and a productive countryside, which we proposed as a regulatory factor in the creation of a commercial line of craft products, functioned on the scale of the socio-economic network of the production centre’.1131 In principle, this ‘urban hub’ had to support itself through a dynamic countryside, which could sustain the production of an agricultural surplus or for that matter any product or commodity, required or in demand (also) beyond the region.1132 A happy few of the urban inhabitants had the land and financial means to invest in agricultural production, and clearly did so when it (also) came to artisanal production. If ceramic tablewares were or could not be traded for their intrinsic value,1133 the surplus provided a means by which tablewares could travel to the outside world. The achievement of both, of course, required sufficient investments. Lund pointed out that the production of all major (eastern) wares of the 2nd century BC to the 7th century took place in the general vicinity of major towns, and with easy access to the Mediterranean.1134 Secondly, the existence or creation of (a) pulling force(s), towns, harbours, provincial capitals and so on, with a political-administrative, economic, religious, military, or combined function. Furthermore, their role could be intermediary, or that of producer and/or consumer. Examples include Corinth and Constantinople, each with their own blend of significant political-administrative, economic, military, and other features. A pulling factor need not necessarily have been a major settlement, as Upper Zohar illustrates. Thirdly, both pushing and pulling forces had to be directly or indirectly connected through lines of communication, composed of a complicated, interconnected network of shorter and longer routes, combining cabotage, open sea routes as well as land routes.

Poblome et al. 2007, 227. Fulford 1987; Poblome et al. 2007. 1133  Parker 1992. 1134  Lund 2003, 131.

Figure 106. Table showing the number of different ESA forms in its core zone (Levant and Cyprus), and elsewhere in the eastern Mediterranean (© Philip Bes/ICRATES Project).

Fourth, a political or other system that encapsulates these factors. The Roman Empire, and to some extent even the late Republic, served as such a framework for the Mediterranean. Under specific conditions both public and private interests created and/or stimulated regional and supraregional exchange, channelling commercial trade, the demand for an agricultural surplus and other commodities in order to sustain part of the population and the army, but, for instance, also elite demand. Socio-economic development as regards urbanisation, architecture, interior decoration, and the arts and crafts were stimulated following the Pax Romana.1135 In relation to the above, ESA is the obvious chronological starting point. Figure 104 captures its large core zone of distribution: Asia Minor-East Central (c. 95.6%), Asia Minor-Cilicia (notably Cilicia Pedias) (c. 82%), the Levant-Coastal North (c. 92.8%) and -Interior North (c. 97.7%), the Levant-Coastal South (c. 87.8%) and -Interior South (c. 78.2%). Given the presumed area of production, Antioch and its hinterland could have provided the urban and rural context in which ESA could develop and flourish. In fact, the largest variety of forms was attested at Antioch (Figure 106), and even if the pottery still derives from one publication, Hayes’ typo-chronology is — not for nothing — largely based on that publication.1136 The ESA from Kinet Höyük, located in the Issos plain and abandoned c. 60-50 BC, includes the entire late Hellenistic repertoire of ESA, as well as a considerable variety of BSP and ESA shapes, some of which have no parallel elsewhere.1137

1131  1132 

Berns et al. 2002; Meyer 2007. Waagé 1948. 1137  As yet unpublished finds, studied by the author and Jeroen Poblome. 1135  1136 

144

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:43 UTC All use subject to https://about.jstor.org/terms

Thoughts on the Socio-Economic Significance of Red Slip Tablewares Even if this is only one end of the chronological spectrum of ESA, this strongly suggests, speaking in terms of typology and distribution, that Kinet Höyük was located comparatively close(r) to where ESA was manufactured. What provided the wider economic basis, however, escapes us archaeologically, although recent work in the Amuq plain north of Antioch may well hold some clues.1138 On the other hand, if ESA is to be identified as Cicero’s rhosica vasa, it could have been more than just a product that rode piggy back along a(n) (agricultural) surplus; despite Cicero’s somewhat derogatory attitude, rhosica vasa thus could have been marketed on its own terms. ESA reached a number of central Mediterranean sites by the late 2nd, early 1st century BC, mostly a few standard (preferred?) forms (Figure 57), which suggests that ESA caught on, not only entering existing and/or developing exchange networks but more importantly, it could well reflect an intensified investment of time, people and money into expansion of the production output. This is perhaps a spin-off effect of the increasing meddling by the late Republic in the east, which increasingly pulled it into its sphere of political and military influence. Moreover, if Antioch was the prerequisite urban hub for ESA, the Seleucid kingdom in which it was located was leaving its glory behind it. ESA was being moved into new markets. The emporion of Delos is usually thought to have played an important role,1139 and its epigraphic and archaeological record sufficiently indicates that it attracted both Italian and eastern residents,1140 acting as a pivotal point in the (central) Mediterranean where west and east literally met. Unfortunately, the published record of ESA from Delos remains fairly meagre, and does not allow us to substantiate that claim archaeologically. Another important lacuna is the mid-1st century BC,1141 a period for which comparatively few deposits are published, making it impossible at this stage to determine whether or not ESA continued to arrive in the central Mediterranean. Even if the destruction of Delos in 87 BC meant a downturn in its role, the prolific ESA forms 3, 4A and 22A(-B) continued into the Augustan period, and clearly their distribution did not halt during the mid-1st century BC. The exceptional distribution of ESA during phase 1(a) stands out even more compared to ESC and ESD: their distribution was largely regional during the late Hellenistic period (phase 1). The early Empire saw considerable changes, with the arrival of ITS and ESB, western influences permeating eastern tableware morphology, and expanding distribution patterns for ESC and ESD. ESA was still found in the central Mediterranean. The distribution of ITS now included the east, yet its distribution is uneven. Corinth, Knossos, Gortyna and Berenice appear to have been notable consumers, yet their (relative) proximity to Italy should again be emphasised. Greece was long thought Yener 2006. Malfitana et al. 2005. 1140  Rauh 1993; Müller forthcoming. 1141  See e.g. Kögler 2010. 1138  1139 

to be devoid of any (major) production of terra sigillata, thus depending on external supply. Corinth, Patras and Koroneia, are now known to have had a local production, which puts the import of ITS into perspective.1142 At Berenice, however, all tablewares from or after the early 1st century onwards were imported.1143 That exchange between west and east passed through these regions could at least partly explain the prolific presence of ITS, and provides a different slant to the socio-cultural role ascribed to ITS, but since these regions are home to a fair number of coloniae this really faces us with a chickenor-the-egg scenario. The mixed picture of the collected evidence in the Aegean-Southwest (Figure 104) more generally reflects that intermediary role. Presumably the majority of the ITS that headed eastwards was consumed in central and southern Greece, Crete and Cyrenaica. The quantitative distribution of stamped vessels, as well as that of the earliest and latest forms roughly seems to confirm this. If Antioch played a vital role in pushing the distribution of tablewares in the late Hellenistic and early Roman periods, the late Roman situation was quite the opposite. With regard to the tableware evidence, the role of Antioch and its hinterland shifted from that of a producer to that of consumer, which mostly concerned ARSW and LRC. To explain the trigger of this pattern, as the Antiochene evidence alone is insufficient, the east at large needs to be considered, through which the outlines of (regional) patterns emerge. ARSW was most common in the AegeanSouthwest (c. 67.4%) and Africa-Cyrenaica (c. 70.7%), implying close connections with Tunisia. ARSW is also fairly common in the Levant-Coastal North (c. 55.7%). LRC dominated the Aegean-North (c. 60.2%, omitting ESC) and -East (c. 83.4%), Cilicia Pedias (Figure 78), the Levant-Interior North (c. 70.4%), the Levant-Coastal South (c. 41.5%) and -Interior South (c. 45.3%), implying lines of exchange between the Aegean(-East) and these eastern regions. Thus, two main axes emerge, both centered on the Aegean. First, that between Tunisia and the Aegean, which brought considerable quantities of ARSW to (especially) the Aegean-Southwest and Cyrenaica. The second is that between the Aegean and parts of the east. Crete presents a more balanced picture (Figure 62). The Aegean thus played a pivotal role in these axes, in the same way as late Hellenistic Delos may have done. In fact, these particular axes could only emerge with the founding of Constantinople, which will have been a major pull in the Mediterranean from 330 onwards, triggering the boom in ARSW in the second half of the 4th century, as well as the emergence of LRD and LRC by the late 4th century. In fact, the shift of the capital eastwards meant a reorientation of exchange patterns. The agro-economic importance of 1142  Corinth: Wright 1980; Slane 2003; Patras: Hübner 1996, 2003. Pottery manufacture at Koroneia is evidenced through numerous wasters, as well as kiln fragments, moulds and tools found during surface prospection between 2006 and 2011. See also Hagn 2003 for presumed local or regional products at Aigeira. 1143  Kenrick 1985, 494.

145

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:43 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East Tunisia included its potential of surplus production of olive oil, wine and/or grain. It was recently suggested that it may have been mainly grain with which ARSW could travel, for African amphorae and cooking wares are considered to be generally uncommon in the east.1144 The axis Aegean/Constantinople and parts of the east can be similarly specified. Wine and oil were drawn from these regions to supply Constantinople and the (Danubian) limes; certainly not all was destined to meet imperial and military needs. LRA 1 provided a stable economic basis in Cilicia Pedias, which possibly was of direct imperial concern.1145 Antioch’s hinterland is poorly known in pottery terms,1146 yet it is increasingly believed that the Dead Cities did not supply the large share of wine and/or olive oil they were once thought to be responsible for.1147 In fact, imperial authorities seemingly favoured Cilician traders for as yet unclear reasons.1148 Further south, the surplus scenario is ― in theory ― applicable to the Palestinian amphorae LRA 4 and LRA 5, also widely attested in the central and western Mediterranean.1149 Within the Aegean proper, the production and trade of LRA 2 was probably (partly) an answer to imperial requisition, including its supply to the Danubian limes.1150 Karagiorgou observes that LRA 1 and LRA 2 were purposely assembled at certain point(s) in the Aegean for further redistribution. Naturally, the imperial authorities were also concerned with the eastern limes. The single specimens of LRA 1 and LRA 2 at the small settlement of Upper Zohar (if at all military!) hardly reflect the material core of this concern.1151 Not only can these have arrived out of more commercialised exchange; supplies could also have come from less distant sources. It is within this framework that ARSW was shipped to the Aegean, and ARSW and LRC travelled further eastwards, perhaps as combined cargoes. LRD, resembling the distribution pattern of ESD, could have fulfilled the intermediary role in Lycia, Pamphylia, Cilicia Tracheia, Egypt and the southern Levant through coast-to-coast trade. Egypt, an important supplier of grain, would for that reason also have been connected to the Aegean/ Constantinople: however, since Egyptian amphorae and Bonifay 2003, 120, 2004, 2005. Decker 2001. 1146  Déhès: Orssaud 1980. 1147  Favoured by Wickham (2005, 443-449), though Pieri 2005, 83-85; Reynolds 2005b, 578 (upland Cilicia for the agricultural basis). Wickham does draw attention to the manufacture of LRA 1, considering the Limestone Massif together with ‘Antioch and its hinterland, Cilicia, and parts of Cyprus [...] as the main region that produced LRA I amphorae’ as its region(s) of origin. Archaeological evidence for the manufacture of LRA 1, however, has been predominantly attested in Cilicia and Cyprus: e.g., Demesticha and Michaelides 2001 (Paphos); Autret et al. 2010 (Soli); Burragato et al. 2007 (Elaiussa Sebaste); Pieri 2005, 69-85, esp. 80-81; Empereur and Picon 1989. 1148  Durliat and Guillou 1984; Dagron 1985; Pieri 2005, 84-85; Iacomi 2010. 1149  Kingsley 2001; Pieri 2005, 110-114, 125-127; Reynolds 2005b, 576 with regard to Constantinople. Though LRA 4 and LRA 5 are among the most widely distributed types, amphora manufacture in the late Roman (southern) Levant was not restricted to these alone (e.g. LRA 9/Agora M334). 1150  Karagiorgou 2001. 1151  Harper 1995, 30-31, 132, fig. 16, pl. 25. Both carry dipinti, which unfortunately are not further discussed. 1144  1145 

ERSW are rare outside Egypt, this third axis largely escapes us archaeologically. It nonetheless provides an interesting contrast to ARSW, for if both Tunisia and Egypt supplied agricultural surpluses that were (partly) instigated by the imperial authorities, why then is Egyptian pottery so comparatively scarce outside Egypt? An act in the Theodosian Code states that officially requisitioned cargoes were to carry nothing but the public cargo;1152 if this is rigidly applied to the evidence and we consider that Africa did supply agricultural surpluses within the annona system, it implies that ARSW travelled on its own merits. It would then have been shipped (mostly) commercially, although this hardly explains the situation. Perhaps the Egyptian output played a role, or the quality of its products. Another possible factor is the dominance of ARSW, LRD and LRC, leaving little room for other wares to manoeuvre. Yet an important period is overlooked: the 3rd and early 4th centuries. During this period only small quantities of recognisable wares (ESC, ARSW) circulated in the east, a process that may have started already in the later 2nd century. ESC was basically restricted to the Aegean, and ARSW was only beginning to make an impact in the east (Figures 98-99). Whether or not these are Lund’s ‘Centuries of Darkness?’,1153 a ceramo-centric perspective is certainly only part of the puzzle to improve our understanding of urban and rural life in the Roman east: recent contributions plead for a regional assessment of the archaeological evidence.1154 The current conviction that most major wares (ITS, ESA, ESB, ESD) went into decline in the course of the (second half of the) 2nd century, argues against any direct impact of 3rd-century troubles.1155 The continued manufacture of SRSW during this period testifies that, probably, much more was happening on a regional level. SRSW retained a regional distribution throughout its period of production (c. 25 BC-650/700). Why production could be sustained during the 3rd century may have been determined by specific factors, yet the workshops clearly did not take advantage of the relatively empty tableware landscape in the east, which could reflect ‘the symbiosis between an active urban hub and a productive countryside, [...] a regulatory factor in the creation of a commercial line of craft products, functioned on the scale of the socioeconomic network of the production centre. In other words, Sagalassos was not Pergamum or Ephesos’.1156 So, supraregional distribution of an artisanal product (e.g. tablewares) could only go as far as the countryside could support surplus production? The example of SRSW also shows that much more knowledge is needed about regional products; people still needed (ceramic) tablewares during these ‘centuries of darkness’, which perhaps could Pharr 1969, 399, book XIII.8: ‘[n]othing shall be placed upon a public cargo’. 1153  Lund 1992, 2006c. 1154  Poblome 2006a; Lund 2006c. 1155  Note that ESA continued on a regional scale up to (at least) the mid3rd century: Reynolds 2010. 1156  Poblome et al. 2007, 227; Abadie-Reynal 2007, 171-173. On the association between large(r) urban centres and terra sigillata/red slip ware manufacture more generally, see Lund 2003; Poblome 2004b, 29. 1152 

146

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:43 UTC All use subject to https://about.jstor.org/terms

Thoughts on the Socio-Economic Significance of Red Slip Tablewares

Figure 107. Stacked column chart showing the number of different forms per ware, per fifteen-year interval (ESC is omitted for reasons of clarity) (© Philip Bes/ICRATES Project).

no longer be sufficiently supplied by the late Hellenistic and early Roman terra sigillata manufactories. But inland Sagalassos, and surely others, had never been dependent at all on major tablewares; also during the Hellenistic period had Sagalassos and the wider region catered for its own tablewares. 7.2 Socio-Cultural Observations

into their daily lives. It is most apparent during the late 1st century BC and the first half of the 1st century, even if not all forms were equally common.1159 If, during the early Empire, ITS forms (or their equivalents in the more costly metal, or even other materials) were indeed important stimuli in deciding upon morphological change, the collected evidence can provide some hints as to the reception and integration of western terra sigillata in the east.

As a matter of fact, the (early) 3rd century marks another important development. The number of manufactured wares and forms gradually decreases, a process that accelerated after the early 2nd century (Figure 107).1157 Also, the general morphological repertoire greatly reduced by the 3rd century, a trend that may also have begun earlier. Small shapes, not necessarily only for drinking, were increasingly absent, a development that becomes especially apparent from the later 4th century onwards, signifying a major break with late Hellenistic and early Roman traditions; glass presumably replaced these small(er) vessels to some extent.

As a starting point, during the Augustan period and the ensuing decades ESA shows considerable changes. Following the model of typological negotiation, rather than from ITS directly, these changes were inspired by metal prototypes which were owned by the elite.1160 Whereas the production of Hellenistic forms continued (e.g. form 4B, 22B), with producers and customers probably keeping to existing traditions for some time, the morphological repertoire of ESA now also included ‘western’ forms (e.g. forms 37, 47, Figure 108). This process was to continue throughout the 1st century; those shown in Figure 108 are the more common ones.

The early imperial period was in this respect perhaps exceptional, and perhaps the ‘Augustan tableware boom’ inaugurated a phase of experimentation, with the desire for broader morphological luxury.1158 It is unclear in what ways different segments of society incorporated this morphological ‘luxury’, or terra sigillata more generally,

The evidence of stamps, particularly on ESA, is a reliable material expression of the model of negotiation. Rather than actual potters’ stamps, which occur on ITS yet rarely on ESA, most stamps on ESA (which moreover only occur on italicised forms)1161 have a positive connotation: salutations, greetings or wishes; possibly we see the materialisation of an eastern (Hellenistic) dining habit. Other wares also betray western influences, and ESB is the

1157  ESC is omitted from Figure 107: the large number of forms would obscure other trends. 1158  Poblome and Zelle 2002. A comparative study into other decorative expressions (metalware, glass, architecture, sculpture) provides an interesting framework, e.g. Berns et al. 2002; Meyer 2007; Willet 2012.

Bes and Poblome 2006, 158-159, fig. 6. Poblome et al. 2007, 229. 1161  Hayes 1985a, 11-12. 1159  1160 

147

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:43 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East

Figure 108. Morphological correspondence between some of the more common ITS forms (top left and right, after Ettlinger et al. 1990, 87, 91, Tafeln 18, 20) and ESA (bottom left and right) (© Hayes 1985a, Tavola V.12, and Tavola VI.16, reproduced with kind permission).

most particular, even if such influences are uncommon, if not near-absent in the later products.1162 Conversely, ESC, but especially ESD and SRSW show only very few such influences. The ‘tableware boom’ signifies a ‘fairly unique phase of empire-wide cultural integration’,1163 reflecting the Mediterranean hegemony that came about under Augustus. The distribution of ITS can be seen as a demonstration of that integration. Did buying and using western/westernstyle terra sigillata necessarily make people feel that they were being or becoming Roman? Other elements (taxation, law)1164 were more explicit, visible, compulsory, whilst the circulation and use of terra sigillata provides a more mixed message. This is most perceptible in the extent to which the different manufactories responded to (the) new morphological inspiration, even if the origins of that inspiration remained largely invisible to a considerable part of the population. If indeed customers perceived the appearance of new shapes, were they necessarily aware of their Italic connotation? For instance, the distribution of italicised ESA forms beyond its core zone (including Cyprus) is extensive, yet thin. Since most have been attested in larger urban centres,1165 an understanding of demand or perceived connotations is thus obscured. Moreover, ESA experienced a contracting distribution during the 1st century.

Zahn 1904; Zabehlicky-Scheffenegger 1995, 1996; Abadie-Reynal 2007, 115. 1163  Poblome et al. 2007, 222. 1164  MacMullen 2000. 1165  E.g. two common 1st century forms: 47 (197 entries, 15 from outside the core zone (Levant, Cyprus, Cilicia), e.g. Assos, Corinth, Ephesos) and 37A-B (127 entries, ten from outside the core zone, e.g. Berenice, Athens). 1162 

For a considerable share of the population, tablewares will have been ‘simple’ tools in their daily lives, and choice (quantity, morphological variety) may have been fairly limited. The degree to which they could afford these also depended on taste, place in society, etc. In many areas and towns the major wares will have formed a mere supplement to local or regional products. Although insufficient evidence is available to firmly study this on an eastern Mediterranean level, it is of interest because it suggests that the way people composed their tableware sets, which reflects their daily, social behaviour, will have differed — by figure of speech — from street to street. A chicken-and-the-egg situation presents itself: will they have incorporated new shapes into existing (dining) habits, or did (dining) habits change because of the availability of new shapes, and to what extent were locally- or regionallymade products influenced by new shapes? Archaeology seeks to map such behaviour, and the distribution of some particular forms offers some tentative clues. In the same vein, the eastern evidence of ITS reveals a particular pattern (Figure 109). Nearly half of the identifiable ITS concerns two basic shapes: plates with a vertical rim, flat floor and high ring base, Consp18-21 (c. 33.2%), and cups with a vertical rim and sloping body, Consp22-23 (c. 14.6%);1166 examples of Consp20.4 and Consp22.1.4 are illustrated in Figure 108 (cf. supra). The evidence from southern Greece is remarkably similar:1167 Consp18-21 comprise c. 36%, and Consp22-23 about 15%. This suggests that these plates and cups were preferred among eastern Mediterranean customers, and may have formed elements in a set, or service.1168 This largely depends of Roughly similar observations were made for Argos (Abadie-Reynal 2007, 61-62), as well as for Cossyra/Pantelleria (Ferrandes 2013, 393396, Graf. 1, Tab. 5), except for Consp34. 1167  Bes and Poblome 2006, 158-159, fig. 6. 1168  Bes and Poblome 2006, 158-159. 1166 

148

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:43 UTC All use subject to https://about.jstor.org/terms

Thoughts on the Socio-Economic Significance of Red Slip Tablewares

Figure 109. Column chart showing counts of the identified ITS forms, in percentages, in the collected data (n=548) (© Philip Bes/ICRATES Project).

course on whether there was intention, which may have been lost on the way, or customers bought and used these forms under their own terms. The relative lack of welldefined, good quality deposits, from a variety of social and functional contexts, does not make answering such questions any easier. Morphologically similar ESA forms (Figure 108) were among the most popular. This parallels the late Hellenistic period, during which ESA forms 3, 4A and 22A were preferentially distributed westwards (Figure 57). If the stamps on ITS from the eastern Mediterranean listed in the OCK (c. 5%) mirror the degree to which ITS was aimed at eastern markets, clearly it did not make a major, quantitative impact; the collected evidence shows ITS to be most common in Africa-Cyrenaica (c. 65.7%), Crete (c. 37.1%) and the Aegean-Southwest (c. 43.7%). Considerable percentages are also attested for Egypt (c. 32.8%). A few standard forms were traded that customers made part of their daily life without being ‘necessarily aware or even interested in the background of the things they bought and used’.1169 Again, their choices probably depended on their social background, economic situation, prospects, ambitions, etc. 7.3 Concluding Thoughts1170 Paradoxically, archaeology is one of those odd disciplines that move forward by looking back, and the same of course holds true for the (sub)discipline of ceramic studies. This 1169  1170 

Poblome et al. 2007, 230. See e.g. Poblome et al. 2013.

final section offers some concluding thoughts as well as a few ideas on how future research could progress. 7.3.1 Typology and Chronology Crucial to any study undertaken that concerns a material category, are typology and chronology. Whereas one or more typo-chronologies fortunately exist for all the wares discussed here — which otherwise would simply not have been possible — continued scrutiny about the (archaeological) dating criteria is required. Even if we will never be able to pinpoint very precisely when any form began to be made, when it experienced its phase of growth, popularity and decline (both locally and beyond, if more widely marketed), nor its very last production phase — let alone life cycles of use and reuse1171 — any attempt in that direction is likely to prove helpful. To that end, deposits of high archaeological value, from a variety of contexts, are imperative. 7.3.2 Quantity and Quantification For this study, 23,272 database entries from 173 sites and surveys from around the eastern Mediterranean were collected. When we recall John Lund’s estimate of the output of the ESA workshops, for one, that is thought to have comprised some 24 million vessels over the course of some 300 to 350 years, it would be a generous

1171 

149

Peña 2007.

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:43 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East understatement to say that this is very little to go by.1172 Although the extremely small size of this sample is obvious enough, in particular Figures 104-105 show the potential of undertaking a study with a wide geographical and chronological focus. I find it difficult to consider that the core tableware regions that thus emerge do not result from this approach. That said, by working strictly with published catalogues, backed up by quantified evidence where available, this study also served to evaluate the catalogue as an accepted means of publishing material. The answer, it appears, must be generally affirmative. That said, there is nonetheless much need for quantified data from a variety of sites, from a variety of deposits (recent work on burial contexts for example),1173 and from a variety of chronological scales, from across the (eastern) Mediterranean and beyond. I do not wish to claim that a single fragment of ESA matches the archaeological, art-historical or cultural impact, let us say, a Nikè of Samothrake, a Monte Testaccio or a Theodosian Harbour could have, to name but a few. In a certain way, however, that sherd of ESA might be as significant within its, in the broad sense of the word, context. This is regrettable, especially since pottery is still not accepted everywhere as a dating tool, let alone considered a valuable means that can contribute to our understanding of the ebb and flow of Mediterranean economies, of choices made by small inland settlements, of matters that deal with technology and style, and so forth. Presently, the field of study remains too dispersed thematically and methodologically, and more intrinsic appreciation for pottery as that ubiquitous archaeological material category that we ‘always’ speak of is desperately needed. Whilst much data was collected, much data was also omitted. In the process of selection, attention was primarily paid to larger catalogues as well as a variety of sites (e.g. coastal, inland, larger town, rural settlements). Much new and valuable data continues to emerge — in addition to the discovery of existing publications — from sites that are already included, as well as new ones; it is simply too much to list here, yet where appropriate, mention was made in the text. In the future, all this information will help to refine patterns and trends on a variety of geographical and chronological levels. 7.3.3 Local, Close-Regional and Supraregional Whereas the approach taken here is easily called supraregional, one instructive way forward is a good understanding of confirmed local production. This is necessary, not only to learn what morphological and decorative schemes local workshops came up with, but also to put things in perspective. The near-unavoidable case of SRSW has clearly shown there was no need to import red slip tablewares over land, over considerable distances, either throughout the Roman imperial period or before; the few imports that have been recognised play no quantative 1172  1173 

Willet and Poblome 2011. Slane and Walbank 2006; Winther-Jacobsen 2006; Rife et al. 2007.

role, yet these rarities do offer thoughts about personal possessions, travellers or traders returning home, etc. The case of Koroneia, in Boeotia, is another case in point. Local manufacture has been archaeologically confirmed from the late Cassical, early Hellenistic period onwards, and continued into the late Roman period. Quantities of Roman-period locally-made tablewares (open and closed) comprise c. 95% of the total tableware count for that period, and so to speak make the imported tablewares pale into insignificance, even if this does not undermine the broader import trends in which Koroneia participated. Knowledge of local products also facilitates an indirect insight into choices about tableware manufacture on that local level. The workshop(s) at Koroneia, for instance, followed a logic that resulted in a morphological repertoire that in its entirety has not much in common with major wares such as those discussed in this study. At the same time, however, some of the more popular forms do recall more widely traded shapes, for example ARSW Hayes form 50. The messages we (try to) decode and read are not, and never will be, clear-cut. 7.3.4 Scope and Category By no means is this considered a final word on the subject. On the contrary, it might or could be seen as a starting point that merely touches that proverbial tip of the ceramic iceberg. Work by Willet already takes a number of topics further — methodologically, thematically — and has also moved beyond the strict boundaries of ceramic as a medium.1174 The general outlines as presented in this study will need to be tested against further data, not only to evaluate the conclusions put forward, but also to refine patterns and trends. And of course we need to look beyond pottery strictu sensu, even if other material categories have their inherent problems (the recycling of glass, to name but one). One way forward, perhaps the only way forward, is increasing specialisation. That specialisation will continue to be based on geographical, chronological, thematic, or material categories (or otherwise), but may well be increasingly focused. It will also make matters much more complex; continuous research on Roman-period pottery from Tunisia is a case in point.1175 The rapidly snowballing variety and quantity of available data is both a blessing and a curse, one which requires large-scale and long-term collaborative efforts to deal with and interpret that data in smaller segments:1176 ‘the question must be broken down into smaller parts—trade in what goods, between what regions, at what times? Each time we have looked at a large dataset, e.g. shipwrecks, we have seen the possibility of certain biases in the data, which need to be tested by isolating a subset of the data […]’.1177 Within this context, regional ceramic trends also provide interesting case Willet 2012. E.g. Bonifay 2004; Bonifay et al. 2012. 1176  The Levantine Ceramics Project, which is coordinated by Andrea Berlin, is one such approach. 1177  Wilson 2008, esp. 245. 1174  1175 

150

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:43 UTC All use subject to https://about.jstor.org/terms

Thoughts on the Socio-Economic Significance of Red Slip Tablewares studies that can improve our understanding of intra- as well as interregional ceramic manufacture, exchange and consumption.1178 This would also be the point where closer collaboration between archaeology and related disciplines becomes desirable, if not necessary. The upsurge in network analysis is a case in point, and has been applied to (part of) the dataset.1179 Admittedly, although this study hardly deals with the large body of literature concerning the ancient Mediterranean economies, nevertheless it is worrying to note that the words ‘pottery’ and ‘ceramic’ do not occur in the foreword of a recently published conference proceedings entitled ‘Quantifying the Greco-Roman Economy and Beyond’.1180

Proper dating still relies heavily on ceramic evidence (as well), yet for decades the study of pottery has advanced beyond simple dating, and is regarded as a valuable tool to the study of a wide range of topics. With the right research agenda, supported by a well-thought-through digital platform, projects will create the right conditions to discover patterns and trends among the myriad of — always valuable — data that is and becomes available. Adding and contrasting, for instance, amphora data to the outlines sketched here, or tableware and/or amphora data from the western Mediterranean and/or the Pontic basin, will surely create charts and distribution maps that will be puzzling at first, but if that data is also properly dealt with methodologically and conceptually, these are surely avenues of very exciting potential.1181

But let us end on a positive note. Ever since the study of Hellenistic and Roman pottery emerged by the late 19th century, it has grown to become a discipline in itself.

Fischer-Genz et al. 2014, passim; Bes accepted b. Brughmans 2010; Leidwanger et al. 2014. 1180  De Callataÿ 2014. 1178  1179 

E.g. Reynolds 1995, 2010. For the Pontic basin, see for example Sazanov 2000; Smokotina 2014.

1181 

151

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:43 UTC All use subject to https://about.jstor.org/terms

152

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:43 UTC All use subject to https://about.jstor.org/terms

Archaeopress

Chapter Title: Bibliography Book Title: Once upon a Time in the East Book Subtitle: The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East Book Author(s): Philip Bes Published by: Archaeopress. (2015) Stable URL: https://www.jstor.org/stable/j.ctvr43kch.13 JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected]. Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at https://about.jstor.org/terms

Archaeopress is collaborating with JSTOR to digitize, preserve and extend access to Once upon a Time in the East

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:44 UTC All use subject to https://about.jstor.org/terms

Bibliography

Classical Authors Pliny the Elder. Naturalis Historiae. 1938-1963, translated by H. Rackham. Volumes I-X. Cambridge, MA, Loeb Classical Library. Seneca. Letters from a Stoic. 1979, translated by R. A. Campbell. Harmondsworth, Penguin Classics. Scriptores Historiae Augustae. 1960, translated by D. Magie. Volumes I-III. Cambridge, MA, Loeb Classical Library. Modern Authors Abadie, C. 1984. Estampilles de potiers italiques trouvées à Argos. Bulletin de correspondance hellénique 108, 425-446. Abadie-Reynal, C. 1987. Céramique romaine. In V. Karageorghis, O. Picard and K. Tytgat (eds.), La nécropole d’Amathonte, tombes 113-367, II. Céramiques non chypriotes. Études Chypriotes 8, 4558. Nicosia, Fondation A. G. Leventis. Abadie-Reynal, C. 1989a. Céramique et commerce dans le bassin Égéen du IVe au VIIe siècle. In V. Kravari, J. Lefort and C. Morrisson (eds.), Hommes et richesses dans l’Empire byzantin I. IVe-VIIe siècle, 143-159. Paris, P. Lethielleux. Abadie-Reynal, C. 1989b. La céramique romaine de Porsuk. Kazı Sonuçları Toplantısı 11, 221-228. Abadie-Reynal, C. 1995. Étude de la ceramique. In P. Marchetti and K. Kolokotsas, Le nymphée de l’agora d’Argos. Fouille, étude architecturale et historique. Études Péloponnésiennes 11, 295-320. Paris, De Boccard. Abadie-Reynal, C. 2004. Les modalités d’importation de la sigillée italique à Argos. In Poblome et al. 2004, 5966. Abadie-Reynal, C. 2005a. Trade relations in the Aegean Sea: the ceramic evidence from Argos between the 1st century BC and the 2nd century AD. In M. Berg Briese and L. E. Vaag (eds.), Trade relations in the Eastern Mediterranean from the Late Hellenistic period to Late Antiquity: the ceramic evidence. Acts from a Ph.D.seminar for young scholars, Sandbjerg Manorhouse, 12-15 February 1998. Halicarnassian Studies 3, 37-49. Odense, University Press of Southern Denmark. Abadie-Reynal, C. 2005b. Les sigillées africaines à Zeugma. In F. Baratte, V. Déroche, C. Jolivet-Lévy and B. Pitarakis (eds.), Mélanges Jean-Pierre Sodini. Travaux et Mémoires 15, 523-546. Paris, Association des amis du Centre d’Histoire et Civilisation de Byzance.

Abadie-Reynal, C. 2007. La céramique romaine d’Argos (fin de IIe siècle avant J.-C.-fin du IVe siècle après J.C.). Études Péloponnésiennes 13, Paris, De Boccard. Abadie-Reynal, C. and Sodini, J.-P. 1992. La céramique paléochrétienne de Thasos (Aliki, Delkos, Fouilles anciennes). Études Thasiennes 13. Paris, De Boccard. Abadie-Reynal, C., Martz, A.-S. and Cador, A. 2007. Late Roman and Byzantine pottery at Zeugma: groups of the beginning of the 5th century. In B. BöhlendorfArslan, A. O. Uysal and J. Witte-Orr (eds.), Çanak. Late Antique and Medieval pottery and tiles in Mediterranean archaeological contexts. Proceedings of the First International Symposium on Late Antique, Byzantine, Seljuk and Ottoman pottery and tiles in archaeological context (Çanakkale, 1-3 June 2005). BYZAS 7, 181-194. Istanbul, Ege Yayınları. Adamsheck, B. 1979. The Pottery. Kenchreai, Eastern Port of Corinth 4. Leiden, Brill. Adan-Bayewitz, D. 1986. The pottery from the Late Byzantine building (stratum 4) and its implications. In L. I. Levine and E. Netzer, 1986: Excavations at Caesarea Maritima 1975, 1976, 1979 – Final report. Qedem 21, 90-131. Jerusalem, Hebrew University of Jerusalem - Institute of Archaeology. Aharoni, Y. 1962. Excavations at Ramat Rahel. Seasons 1959 and 1960. Serie Archeologica 2. Rome, Centro di studi semitici. Akyürek, N. E. 1992. Terra sigillata aus dem Heiligtum des Apollon Smintheios. In Studien zum antiken Kleinasien 2. Asia Minor Studien 8, 125-165. Bonn, Rudolf Habelt. Amir, R. 2006. Pottery, stone and small finds from Shiqmona. ‘Atiqot 51, 145-161. Anderson-Stojanović, V. R. 1992. Stobi 1. The Hellenistic and Roman pottery. Princeton, Princeton University Press. Angeli, A. 2007. Κεραμική από το ρομαϊκό τειχος της Νικόπολης. In K. L. Zachos (ed.), Nicopolis B. Proceedings of the Second International Nicopolis Symposium (11-15 September 2002) Volumes I and II, 527-532. Preveza, Actia Nicopolis Foundation. Armstrong, P. 2005. The construction date of the baptistery at Xanthos. In F. Baratte, V. Déroche, C. Jolivet-Lévy and B. Pitarakis (eds.), Mélanges Jean-Pierre Sodini. Travaux et Mémoires 15, 547-553. Paris, Association des amis du Centre d’Histoire et Civilisation de Byzance. Armstrong, P. 2007. Rural settlement in Lycia in the eighth century: new evidence. In Symposium Proceedings Volume I (The IIIrd Symposium on Lycia 07-10 November 2005, Antalya), 19-29. Antalya, Ege Yayınları.

153

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:44 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East Arslan, N. 2003. Antiocheia (Pisidya) kazılarında bulunan seramikler. Olba 10, 209-238. Arthur, P. 2007. Form, function and technology in pottery production from Late Antiquity to the Early Middle Ages. In L. A. Lavan, E. Zanini and A. Sarantis (eds.), Technology in Transition A.D. 300-650. Late Antique Archaeology 4, 159-186. Leiden, Brill. Ateş, G. 2001. Die Eastern Sigillata B aus einer frühkaiserzeitlichen Auffüllungsschicht in Aizanoi. Archäologischer Anzeiger, 319-332. Atık, N. 1995. Die Keramik aus den Südthermen von Perge. Istanbuler Mitteilungen Beiheft 40. Tübingen, Wasmuth. Atık, N. 1998. Çanak çömlek buluntuları. In J. Inan, Toroslar’da bir antik kent. Eine antike Stadt im Taurusgebirge. Lyrbe?-Seleukeia?, 92-103. Istanbul, Arkeoloji ve Sanat Yayınları. Aupert, P. 1976. Rapport sur les travaux de la mission de l’École française à Amathonte en 1975. Bulletin de correspondance hellénique 100, 907-941. Aupert, P. 1980. Objets de la vie quotidienne à Argos en 585 ap. J.-C. In Études Argiennes. Bulletin de correspondance hellénique Supplément 6, 395-457. Paris, De Boccard. Aupert, P. 1986. 2. Thermes A et rue du Théâtre. Bulletin de correspondance hellénique 110, 767-772. Aupert, P. and Hermary, A. 1980. Rapport sur les travaux de la Mission de l’École française à Amathonte en 1979. Bulletin de correspondance hellénique 104, 805822. Autret, C., Yağcı, R. and Rauh, N. K. 2010. Soli/ Pompeiopolis’te LRA 1 Amphora Fırını Alanı/An LRA 1 amphora kiln site at Soli/Pompeiopolis. News of Archaeology from Anatolia’s Mediterranean Areas 8, 203-207. Avigad, N. 1976. Archaeological discoveries in the Jewish Quarter of Jerusalem. The Second Temple. The Israel Museum Catalogue 144. Jerusalem, Israel Exploration Society. Badre, L. 1991. Le sondage stratigraphique de Shabwa, 1976-1981. Syria 68, 229-314. Bailey, D. M. 1993. Hellenistic and Roman pottery. In G. B. Waywell and J. J. Wilkes, Excavations at Sparta: The Roman Stoa, 1988-1991. Preliminary report, part 1. The Annual of the British School at Athens 88, 221249. Bailey, D. M. 1998: Pottery, lamps and glass of the Late Roman and Early Arab periods. Excavations at ElAshmunein 5. London, British Museum Press. Bald-Romano, I. 1994. A Hellenistic deposit from Corinth. Evidence for interim activity (146-44 B.C.). Hesperia 63, 57-104. Baldwin Bowsky, M. W. 2011. Colonia Iulia Nobilis Cnosus, the first 100 years: the evidence of Italian Sigillata stamps. In R. J. Sweetman (ed.), Roman colonies in the first century of their foundation, 117134. Oxford, Oxbow Books.

Baldwin Bowsky, M. W. 2014. Stamps on Italian Sigillata and the renaissance of Aptera, Crete. Hesperia 83, 503568. Ballet, P. 2002. Les productions céramiques d’Égypte à la période hellénistique. Les indices de l’hellénisation. In F. Blondé, P. Ballet and J.-F. Salles (eds.), Céramiques hellénistiques et romaines. Productions et diffusion en Méditerranée orientale (Chypre, Égypte et côte syro-palestinienne). Travaux de la Maison de l’Orient Méditerranéen 35, 85-96. Lyon, Maison de l’Orient. Baly, T. J. C. 1962. Pottery. In H. Dunscombe Colt (ed.), Excavations at Nessana (Auja Hafir, Palestine), 270303. Princeton, Princeton University Press. Bang, P. F. 2008. The Roman Bazaar. A comparative study of trade and markets in a tributary empire. Cambridge, Cambridge University Press. Bar-Nathan, R. and Adato, M. 1986. Byzantine pottery (Stratum 5). In L. I. Levine and E. Netzer, Excavations at Caesarea Maritima 1975, 1976, 1979 – Final report. Qedem 21, 132-136. Jerusalem, Hebrew University of Jerusalem. Begley, V. 1993. New investigations at the port of Arikamedu. Journal of Roman Archaeology 6, 93-108. Begley, V. 1996. The ancient port of Arikamedu. New excavations and researches 1989-1992. Volume one. Mémoires Archéologiques 22. Pondichéry, École française d’Extrême-Orient. Begley, V. and De Puma, R. D. (eds.) 1991. Rome and India. The ancient sea trade. Madison, University of Wisconsin Press. Ben-Arieh, R. 1997. The Roman, Byzantine and Umayyad pottery. In Y. Hirschfeld, The Roman baths of Hammat Gader. Final report, 347-381. Jerusalem, Israel Exploration Society. Berlin, A. M. 1992. Hellenistic and Roman pottery, preliminary report, 1990. In R. L. Vann (ed.), Caesarea Papers. Straton’s Tower, Herod’s Harbour, and Roman and Byzantine Caesarea. Including the papers given at a symposium held at the University of Maryland, The Smithsonian Institution, and the Jewish Community Center of Greater Washington on 25-28 March, 1988. Journal of Roman Archaeology Supplementary Series 5, 112-128. Ann Arbor, Journal of Roman Archaeology. Berlin, A. M. 1993. Italian cooking vessels and cuisine from Tel Anafa. Israel Exploration Journal 43, 35-44. Berlin, A. M. 1997. The plain wares. In S. C. Herbert (ed.), Tel Anafa II,i: the Hellenistic and Roman pottery. Journal of Roman Archaeology Supplementary Series 10, 1-244. Ann Arbor, Journal of Roman Archaeology. Berlin, A. M. 2000. Appendix 2: a stamped sigillata bowl base from Horvat ‘Eleq. In Y. Hirschfeld, Ramat Hanadiv excavations. Final report of the 1984-1998 seasons, 458-459, 746. Jerusalem, Israel Exploration Society. Berlin, A. M. 2006. Gamla I: the pottery of the Second Temple Period. The Shmarya Gutmann excavations, 1976-1989. Israel Antiquities Authority Reports 29. Jerusalem, Israel Antiquities Authority.

154

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:44 UTC All use subject to https://about.jstor.org/terms

Bibliography Berlin, A. M. 2012. Book review of Jackson and Tidmarsh 2011. Bryn Mawr Classical Review (DOI: http://bmcr. brynmawr.edu/2012/2012-10-09.html) (Accessed 22 January 2015). Berlin, A. M. and Heath, S. 2005. Appendix. Ceramic observations, Hellenistic through Late Roman. In S. E. Alcock, A. M. Berlin, A. B. Harrison, S. Heath, N. Spencer and D. L. Stone, Pylos Regional Archaeological Project, part 7. Hesperia 74, 194-204. Berndt, M. 2003. Funde aus dem Survey auf der Halbinsel von Milet (1992-1999). Kaiserzeitliche und frühbyzantinische Keramik. Internationale Archäologie 79. Rahden, Marie Leidorf. Berns, C., Von Hesberg, H., Vandeput, L. and Waelkens, M. (eds.) 2002. Patris und Imperium. Kulturelle und politische Identität in den Städt der römischen Provinzen Kleinasiens in der frühen Kaiserzeit. Kolloquium Köln, November 1998. Bulletin Antieke Beschaving Supplement 8. Leuven, Peeters. Berti, F. 2000. I bolli italici di Iasos. Rei Cretariae Romanae Fautorum Acta 36, 497-502. Abingdon, Rei Cretariae Romanae Fautores. Berti, F., Carandini, A., Fabbricotti, E., Gasparri, C., Gianelli, M., Moriconi, M. P., Palma, B., Panella, C., Picozzi, M. G., Ricci, A. and Tatti, M. 1970. Le Terme del Nuotatore. Scavo dell’ambiente I. Studi Miscellanei 16. Ostia 2. Rome, De Luca. Bes, P. M. 2007. A chronological and geographical study of the distribution and consumption of tablewares in the Roman east. Unpublished PhD thesis, University of Leuven. Bes, P. M. 2010. Ελληνιστική και ρωμαϊκή κεραμική από την αρχαιολογική επιφανειακή έρευνα στα Σαγαίικα της ΒΔ Πελοποννήσου. In Κ. Παπαγιαννόπουλος (ed.), Σαγαίικα. Μεταμορφώσεις ενός τοπίου (Sagaiika: The Changing Face of a Landscape). Ιστορία Δυτικής Αχαΐας 4, 75-82. Patras, Δήμος Μόβρης/Ινστιτούτο Τοπικής Ιστορίας (Ι.Τ.Ι.) and Ομάδα Αρχαιολογίας του Τοπίου (ΟΜ.ΑΡ.Τ.). Bes, P. M. Accepted a. A rising tide lifts all boats? Italian Sigillata from Boeotia in the central Greece landscape. In E. Farinetti (ed.), Romanized Landscapes. Leiden, Leiden University Press. Bes, P. M. Accepted b. A world within a world. First impressions of Hellenistic to Late Roman ceramic patterns in Boeotia. In D. Zhuravlev and U. Schlotzhauer (eds.), Late Hellenistic and Roman tableware in the Black Sea, Eastern Mediterranean and in the West (150 BC-250 AD). Berlin, Deutsches Archäologisches Institut. Bes, P. M. Forthcoming a. That’s the limit!? Patterning of African and Phocaean Red Slip Wares in central Greece. Bes, P. M. Forthcoming b. African tablewares and amphorae at Sagalassos: quantitative and qualitative observations. Bes, P. M. and Poblome, J. 2006. The ICRATES Platform. A new look at old data. Integrating published evidence

on early Imperial tablewares from Greece. In Malfitana et al. 2006, 141-165. Bes, P. M. and Poblome, J. 2008. (Not) see the wood for the trees? 19,000+ sherds of tablewares and what we can do with them. Rei Cretariae Romanae Fautorum Acta 40, 505-514. Bonn, Rei Cretariae Romanae Fautores. Bes, P. M. and Poblome, J. 2009. African Red Slip Ware on the move: the effects of Bonifay’s Études for the Roman East. In J. H. Humphrey (ed.), Studies on Roman pottery of the provinces of Africa Proconsularis and Byzacena (Tunisia). Hommage à Michel Bonifay. Journal of Roman Archaeology Supplementary Volume 76, 73-91. Portsmouth, Journal of Roman Archaeology. Bes, P. M. and Poblome, J. Accepted. The (Late) Hellenistic to Late Roman pottery from urban Thespiae. In J. L. Bintliff and A. M. Snodgrass (eds.), The urban survey of the city of Thespiae. Cambridge, McDonald Institute for Archaeological Research. Bes, P. M., Poblome, J. and Bintliff, J. L. 2006. Puzzling over pottery. Thespiae, Tanagra and methodological approaches toward surface ceramics. In Malfitana et al. 2006, 339-345. Bes, P. M., Poblome, J., Malfitana, D. and Bintliff, J. L. 2011. Late Roman tablewares from the survey at ancient Tanagra. Finding the wider perspective. In R. Attoui (ed.), When did Antiquity end? Archaeological case studies in three continents. The proceedings of an international seminar held at the University of Trento on April 29-30, 2005 on Late Antique societies, religion, pottery and trade in Germania, Northern Africa, Greece, and Asia Minor. British Archaeological Reports International Series 2268, 129-136. Oxford, Archaeopress. Beyll, D. 1993. Terra Sigillata aus der Marienkirche in Ephesos. Erste Zwischenbilanz. Österreichischen Archäologisches Institutes Berichte und Materialien 5, 5-45. Biers, J. C. 1985. The Great Bath on the Lechaion Road. Corinth 17. Princeton, American School of Classical Studies at Athens. Blackman, D. J. and Branigan, K. 1975. An archaeological survey on the south coast of Crete, between the Ayiofarango and Chrisostomos. The Annual of the British School at Athens 70, 17-36. Blanton, R. E. 2000. Ceramic tabulations. In R. E. Blanton, Hellenistic, Roman and Byzantine settlement patterns of the coast lands of Western Rough Cilicia. British Archaeological Reports International Series 879, 7889. Oxford, Archaeopress. Boardman, J. 1989. The finds. In M. Balance, J. Boardman, S. Corbett and M. S. F. Hood, Excavations in Chios 1952-1955. Byzantine Emporio. British School of Archaeology at Athens Supplement 20, 86-142. London, Thames & Hudson. Böhlendorf-Arslan, B. 2007. Stratified Byzantine pottery from the city wall in the southwestern sector of Amorium. In B. Böhlendorf-Arslan, A. O. Uysal and J. Witte-Orr (eds.), Çanak. Late Antique and Medieval pottery and tiles in Mediterranean archaeological

155

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:44 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East contexts. Proceedings of the First International Symposium on Late Antique, Byzantine, Seljuk and Ottoman pottery and tiles in archaeological context (Çanakkale, 1-3 June 2005). BYZAS 7, 273-294. Istanbul, Ege Yayınları. Bommeljé, S. and Vroom, J. C. 1995. “Deserted and untilled lands”: Aetolia in Roman Times. Pharos. Journal of the Netherlands Institute at Athens 3, 67130. Bonifay, M. 1996. Annexe 1. Les sigillées tardives. In C. Aubert, Rapport préliminaire des fouilles de l’IFAPO dans le centre ville de Beyrouth, Place des Martyrs. Bulletin d’Archéologie et d’Architecture Libanaises 1, 85-89. Bonifay, M. 2002. La céramique africaine aux époques vandale et Byzantine. Antiquité Tardive 10, 177-195. Bonifay, M. 2003. La céramique africaine, un indice du développement économique? Antiquité Tardive 11, 113-128. Bonifay, M. 2004. Études sur la céramique romaine tardive d’Afrique. British Archaeological Reports International Series 1301. Oxford, Archaeopress. Bonifay, M. 2005. Observations sur la diffusion des céramiques africaines en Méditerrannée orientale durant l’Antiquité tardive. In F. Baratte, V. Déroche, C. Jolivet-Lévy and B. Pitarakis (eds.), Mélanges Jean-Pierre Sodini. Travaux et Mémoires 15, 565-581. Paris, Association des amis du Centre d’Histoire et Civilisation de Byzance. Bonifay, M. 2007. Ceramic production in Africa during Late Antiquity: continuity and change. In L. A. Lavan, E. Zanini and A. Sarantis (eds.), Technology in Transition A.D. 300-650. Late Antique Archaeology 4, 143-158. Leiden, Brill. Bonifay, M. Forthcoming. Can we speak of pottery and amphora ‘import substitution’ in inland regions of Roman Africa? In D. J. Mattingly (ed.), TRANSSAHARANS: Human Mobility and Identity, Trade, State Formation and Mobile Technologies across the Sahara (1000 BC-AD 1500). Volume II: Trade. Cambridge. Bonifay, M. and Leffy, R. 2002. Les céramiques du remplissage de la citerne du Sarapéion à Alexandrie. In J.-Y. Empereur (ed.), Alexandrina 2. Études alexandrines 6, 39-84. Cairo, Institut Français d’Archéologie Orientale. Bonifay, M., Capelli, C. and Brun, C. 2012. Pour une approche intégrée archéologique, pétrographique et géochimique des sigillées africaines. In M. Cavalieri (ed.), INDVSTRIA APIVM. L’archéologie: une démarche singulière, des pratiques multiples. Hommages à Raymond Brulet. FERVET OPVS 1, 41-62. Louvain-la-Neuve, Presses Universitaires de Louvain. Bounegru, O. and Erdemgil, S. 1998. Terra-SigillataProduktion in den Werkstätten von Pergamon-Ketiostal – Vorläufiger Bericht. Istanbuler Mitteilungen 48, 263277.

Bowman, A. 1996. Post-Roman imports in Britain and Ireland: a maritime perspective. In K. R. Dark (ed.), External contacts and the economy of Late Roman and post-Roman Britain, 97-108. Woodbridge, Boydell Press. Braund, D. 2007. Black Sea grain for Athens? From Herodotus to Demosthenes. In V. Gabrielsen and J. Lund (eds.), The Black Sea in Antiquity. Regional and interregional economic exchanges. Black Sea Studies 6, 39-68. Aarhus, Aarhus University Press. Brock, J. K. 1949. Pottery. In G. Mackworth Young and J. K. Brock, Excavations in Siphnos, The Annual of the British School at Athens 44, 31-74. Brughmans, T. 2010. Connecting the dots: towards archaeological network analysis. Oxford Journal of Archaeology 29.3, 277-303. Brun, J.-P. and Brunet, M. 1997. Une huilerie du premier siècle avant J.-C. dans le Quartier du théâtre à Délos. Bulletin de correspondance hellénique 121, 573-615. Bruneau, Ph. 1970. La vaisselle. In Ph. Bruneau, C. Vatin, U. Bezerra de Meneses, G. Donnay, E. Levy, A. Bovon, G. Siebert, V. R. Grace, M. SavvatianouPetropoulakou, E. Lyding Will and T. Hackens, L’îlot de la maison des comédiens. Exploration archéologique de Délos Fascicule 27, 239-262. Paris, De Boccard. Bruneau, Ph. 1991. La céramique pergaménienne à reliefs appliqués de Délos. Bulletin de correspondance hellénique 115, 597-666. Burkhalter, F. 1987. La céramique hellénistique et romaine du sanctuaire d’Aphrodite à Amathonte (fouilles 1976 à 1982). Bulletin de correspondance hellénique 111, 353-395. Burragato, F., Di Nezza, M., Ferrazzoli, A. F. and Ricci, M. 2007. Late Roman 1 amphora types produced at Elaiussa Sebaste. In M. Bonifay and J.-C. Tréglia (eds.), LRCW 2. Late Roman Coarse Wares, Cooking Wares and Amphorae in the Mediterranean. Archaeology and Archaeometry. British Archaeological Reports International Series 1662, 689-700. Oxford, Archaeopress. Butcher, K. 2003. Roman Syria and the Near East. London, British Museum Press. Calderon, R. 1999. The pottery. In Y. Hirschfeld, The Early Byzantine monastery at Khirbet ed-Deir in the Judaean Desert: the excavations in 1981-1987. Qedem 38, 135147. Jerusalem, Hebrew University of Jerusalem. Calderon, R. 2000. Appendix I: Byzantine pottery from South Sinaï. In Dahari 2000, 183-231. Camardo, D. 2002. Appendice I. Primo inquadramento dei materiali ceramici da Hephaestia dall’età imperiale al tardo antico. In E. Greco, Hephaestia. Annuario della Scuola Archeologica di Atene e delle Missioni italiane in Oriente 80 Series 3.2, Part 2, 1002-1006. Campbell, E. 1996. The archaeological evidence for external contacts: imports, trade and economy in Celtic Britain A.D. 400-800. In K. R. Dark (ed.), External contacts and the economy of Late Roman and postRoman Britain, 83-96. Woodbridge, Boydell Press.

156

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:44 UTC All use subject to https://about.jstor.org/terms

Bibliography Carandini A., Saguì, L., Tortorella, S. and Tortorici, E. 1981. Ceramica africana. In A. Carandini, L. Anselmino, C. Pavolini, L. Saguì, S. Tortorella and E. Tortorici, Atlante delle forme ceramiche I. Ceramica fine romana nel bacino Mediterraneo (medio e tardo impero). Enciclopedia dell’arte antica. Classica e orientale, 9-183. Rome, Istituto della Enciclopedia Italiana. Carandini, A. and Panella, C. (eds.) 1973. Le Terme del Nuotatore. Scavo degli ambienti III, V, VI, VII e di un saggio nell’area SO. Studi Miscellanei 21. Ostia 3. Rome, De Luca. Carandini, A. and Panella, C. (eds.) 1977. Le Terme del Nuotatore. Scavo dell’ambiente XVI e dell’area XXV. Studi Miscellanei 23. Ostia 4. Rome, De Luca. Carmi, I., Eldar-Nir, I., Nir, Y., Werker, E., Stager, L. E. and Johnson, B. L. 1994. The dating of ancient waterwells by archaeological and 14C methods: comparative study of ceramic and wood. Israel Exploration Journal 44, 184-200. Casal, J. M. 1949. Fouilles de Virampatnam-Arikamedu. Rapport de l’Inde et de l’Occident aux environs de l’ère chrétienne. Paris, Klincksieck. Catling, H. W. 1972. An Early Byzantine pottery factory at Dhiorios in Cyprus. Levant 4, 1-82. Cavalier, L. 2007. Horrea d’Andriakè et Patara: un nouveau type d’édifice fonctionnel en Lycie à l’époque impériale. Revue des Études Anciennes 109, 51-65. Cherry, J. F., Davis, J. L. and Mantzourani, E. 1991. Landscape Archaeology as long-term history. Northern Keos in the Cycladic Islands from earliest settlement until modern times. Monumenta Archaeologica 16. Los Angeles, UCLA Institute of Archaeology. Christensen, A. P. 1986. Les céramiques romaines et byzantines. In A. P. Christensen, R. Thomsen and G. Ploug, Hama. Fouilles et recherches 1931-1938 III 3. The Graeco-Roman objects of clay, the coins and the necropolis. Nationalsmuseets Skrifter 10, 7-31. Copenhagen, Nationalmuseet. Civelek, A. 2010. Red-slip ware from Tralles excavations. Colloquium Anatolicum 9, 169-191. Clamer, Chr. 1997. La documentation matérielle romaine et Byzantine. I. La céramique de la période romaine ancienne, la vaiselle de pierre et les stucs. In Chr. Clamer, Fouilles archéologiques de ‘Aïn ez-Zâra/ Callirrhoé villégiature hérodienne. Bibliothèque Archéologique et Historique Tome 147, 63-80. Beirut, Institut Français d’Archéologique du Proche-Orient. Clamer, Chr. and Magness, J. 1997. La documentation matérielle romaine et Byzantine. II. La céramique romaine tardive de la période Byzantine ancienne. In Chr. Clamer, Fouilles archéologiques de ‘Aïn ezZâra/Callirrhoé villégiature hérodienne. Bibliothèque Archéologique et Historique Tome 147, 81-89. Beirut, Institut Français d’Archéologique du Proche-Orient. Clarke, G. 2011. Preface. In Jackson and Tidmarsh 2011, xxx-xxxi. Coldstream, J. N. 1972. Deposits of pottery from the settlement. In J. N. Coldstream and G. L. Huxley

(eds.), Kythera. Excavations and studies conducted by the University of Pennsylvania Museum and the British School at Athens. London, Faber & Faber. Coldstream, J. N. 1973. The pottery. In J. N. Coldstream, Knossos. The sanctuary of Demeter. The British School at Athens Supplement 8, 18-55. London, Thames & Hudson. Comfort, H. C. 1929. Arretine signatures found in the excavations in the theatre district of Corinth. American Journal of Archaeology 33, 484-501. Comfort, H. C. 1948. Imported western terra Sigillata. In F. O. Waagé (ed.), Antioch on-the-Orontes 4: part one, ceramics and Islamic coins, 61-77. Princeton, Princeton University Press. Comfort, H. C. 1958. Imported pottery and glass from Timna’. In R. LeBaron Bowen jr. and F. P. Albright (eds.), Archaeological discoveries in South Arabia, 199-212. Baltimore, John Hopkins Press. Comfort, H. C. 1960. Some imported pottery at Khor Rori (Dhofar). Bulletin of the American School of Oriental Research 160, 15-20. Comfort, H. C. 1991. Terra Sigillata at Arikamedu. In Begley and De Puma, 134-150. Comfort, H. C. and Waagé, F. O. 1936. Selected pottery from Beth Shan (Roman date). Palestine Exploration Fund Quarterly Statement 68, 221-224. Connelly, J. B. 2005. Excavations on Geronisos Island: second report, the Central South Complex. Report of the Department of Antiquities, Cyprus, 149-181. Conze, A. 1903. Die Kleinfunde aus Pergamon, Berlin, Verlag der Königlichen Preussischen Akademie der Wissenschaften. Cox, D. H. 1949. The Greek and Roman pottery. The excavations at Dura-Europos conducted by Yale University and the French Academy of inscriptions and letters. Final Report 4 Part 1, Fascicle 2. New Haven, Yale University Press. Croissant, F. 1971. 2. Sondages. Bulletin de correspondance hellénique 95, 748-755. Crowfoot, G. M. 1957. Roman and later wares, 3. Terra Sigillata general list, 4. Late Roman A, B and C wares. In Crowfoot et al. 1957, 306-361. Crowfoot, J. W., Crowfoot, G. M. and Kenyon, K. M. 1957. The Objects from Samaria. Samaria-Sebaste 3. London, Palestine Exloration Fund. Dagron, G. 1985. II. Un tarif des sportules à payer aux curiosi du port de Séleucie de Piérie (VIe siècle). In G. Dagron and D. Feissel, Inscriptions inédites du Musée d’Antiochie. Travaux et Mémoires 9, 435-455. Dahari, U. 2000. Monastic settlements in South Sinaï in the Byzantine period, The Archaeological remains. Israel Antiquities Authority Reports 9. Jerusalem, Israel Antiquities Authority. 2004. Damaskus, Zitadelle. Archäologischer Anzeiger, 363-364. Dar, Sh. (ed.) 1999. Sumaqa. A Roman and Byzantine Jewish village on Mount Carmel, Israel. British Archaeological Reports International Series 815. Oxford, Archaeopress.

157

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:44 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East Daszewski, W.A. 1990. Excavations at Marina elAlamein: 1987-1988. Mitteilungen des Deutschen Archäologischen Instituts, Abteilung Kairo 46, 15-51. Daszewski, W.A. 1995. Cypriot sigillata in Marina elAlamein. In H. Meyza and J. Mlynarczyk (eds.), “Hellenistic and Roman pottery in the Eastern Mediterranean – Advances in scientific studies” Acts of the 2 Nieborów pottery workshop, Nieborów, 18-20 december 1993, 27-39. Warsaw, Polish Academy of Sciences. Daszkiewicz, M. and Baranowski, M. 2011. The potential of macroscopic identification of laboratory-defined provenance groups. The case of so-called Pergamenian Sigillata from Delos, Greece. Études et Travaux 24, 42-65. Daszkiewicz, M. and Schneider, G. 2011. Laboratory analysis of so-called Pergamenian Sigillata from Delos, Greece. Études et Travaux 24, 78-91. Dauphin, C. and Kingsley, S. A. 2003. Ceramic evidence for the rise and fall of a Late Antique ecclesiastical estate at Shelomi in Phoenicia Maritima. In G. Claudio Bottini, L. di Segni and L. Daniel Chrupcala (eds.), One land-many cultures. Archaeological studies in honour of Stanislao Loffreda OFM. Studium Biblicum Franciscanum Collectio Maior 41, 61-74. Jerusalem, Franciscing Printing Press. Davies, J. K. 2005. Linear and nonlinear flow models for ancient economies. In J. G. Manning and I. Morris (eds.), The ancient economy. Evidence and models, 127-156. Stanford, Stanford University Press. De Callataÿ, F. 2005. The Graeco-Roman economy in the super long-run: lead, copper, and shipwrecks. Journal of Roman Archaeology 18, 361-372. De Callataÿ, F. 2014. Foreword. In F. de Callataÿ (ed.), Quantifying the Greco-Roman economy and beyond. Pragmateiai 27, 5-11. Bari, Edipuglia. De Contenson, H. 1963. Les fouilles à Axoum en 1958. Rapport préliminaire. Annales d’Éthiopie 5, 3-16. De Ligt, L. 1993. Fairs and markets in the Roman Empire: economic and social aspects of periodic trade in a pre-industrial society. Dutch Monographs on Ancient History and Archaeology 11. Amsterdam, Gieben. De Tommaso, G. 2001. Ceramica a pareti sottili. In A. DiVita (ed.), Gortina V.3. Lo scavo del pretorio (19891995). Volume V.3 t. I. I materiali. Monografie della Scuola Archeologica italiana di Atene e delle Missioni italiane in Oriente 12, 21-24. Padua, Aldo Ausilio. Debevoise, N. C. 1934. Parthian Pottery from Seleucia on the Tigris. University of Michigan Studies, Humanistic Series 32. Ann Arbor, University of Michigan Press. Decker, M. J. 2001. Food for an empire: wine and oil production in North Syria. In S. A. Kingsley and M. J. Decker (eds.), Economy and exchange in the East Mediterranean during Late Antiquity. Proceedings of a conference at Somerville College, Oxford, 29th May, 1999, 69-86. Oxford, Oxbow Books. Degryse, P., Poblome, J., Bounegru, O. and Viaene, W. 2001. Archaeometry of Eastern Sigillata C from Pergamon: a reconnaissance Study. Rei Cretariae

Romanae Fautorum Acta 37, 115-117. Abingdon, Rei Cretariae Romanae Fautores. Dello Preite, A. 1984 [1988]. Le importazioni di ceramica fine a Gortina e a Crete tra il IV e il VII sec. d.C. Annuario della Scuola Archeologica di Atene e delle Missioni Italiane in Oriente 72 (46), 177-198. Dello Preite, A. 1997. Sigillata africana, Sigillata “Late Roman C” (“Phocaean Red Slip Ware”), Sigillata cipriota tarda (“Cypriot Red Slip Ware”), Altre ceramiche fini tarde. In A. DiVita and A. Martin (eds.), Gortina II. Pretorio. Il materiale degli scavi colini 1970-1977. Monografie della Scuola Archeologica italiana di Atene e delle Missioni italiane in Oriente 7, 132-200. Padua, Aldo Ausilio. Dello Preite, A. 2004. Ceramica Sigillata. In A. DiVita (ed.), Gortina 6. Scavi 1979-1982. Monografie della Scuola Archeologica italiana di Atene e delle Missioni italiane in Oriente 14, 426-436. Padua, Aldo Ausilio. Delougaz, P. 1960. The objects. In P. Delougaz and R. C. Haines, A Byzantine church at Khirbet al-Karak. The University of Chicago Oriental Institute Publications 85, 30-52. Chicago: The University of Chicago Press. Demangel, R. and Laumonier, A., 1923. Fouilles de Notion (1921). Première partie. Topographie, architecture, épigraphie. Bulletin de correspondance hellénique 47, 353-386. Demesticha, S. and Michaelides, S. 2001. The excavation of a Late Roman 1 amphora kiln in Paphos. In Villeneuve and Watson (eds.) 2001, 289-296. Déroche, V. and Pétridis, P. 1992. 4. Agora romaine et “Thermes du sud”. Bulletin de correspondance hellénique 116, 709-711. Deubner O. 1957. Die griechischen Scherben von Babylon. In F. Wetzel, E. Schmidt and A. Mallwitz, Das Babylon der Spätzeit. Ausgrabungen der Deutschen Orient-Gesellschaft in Babylon 8. Wissenschaftliche Veröffentlichungen der Deutschen Orient-Gesellschaft 62, 51-58. Berlin, Gebrüder Mann Verlag. Diederichs, C. 1980. Céramiques hellénistiques, romaines et byzantines. Salamine de Chypre 9. Paris, Maison Orient Mediterranéen. Dinahet-Couilloud Le, M.-Th. 1997. Une famille de notables tyriens à Délos. Bulletin de correspondance hellénique 121, 617-666. Doğer, L., Byzantine ceramics: excavations at Smyrna agora (1997-98 and 2002-03). In B. BöhlendorfArslan, A. O. Uysal and J. Witte-Orr (eds.), Çanak. Late Antique and Medieval pottery and tiles in Mediterranean archaeological contexts. Proceedings of the First International Symposium on Late Antique, Byzantine, Seljuk and Ottoman pottery and tiles in archaeological context (Çanakkale, 1-3 June 2005). BYZAS 7, 97-122. Istanbul, Ege Yayınları. Domżalski, K. 2002. 13. Preliminary information on Terra Sigillata and Red Slip Wares from Chhîm. In T. Waliszewski and R. Ortali-Tarazi, Village romain et byzantin à Chhîm-Marjiyat. Rapport préliminaire (1996-2002). Bulletin d’Archéologie et d’Architecture Libanaises 6, 77-85.

158

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:44 UTC All use subject to https://about.jstor.org/terms

Bibliography Domżalski, K. 2012. At the crossroads of trade routes: terra sigillata, red slip wares and related fine pottery from the Polish Excavations in Ptolemais (2002-2009). In J. Żelazowski (ed.), Ptolemais in Cyrenaica. Studies in Memory of Tomasz Mikocki, 317-343. Warsaw, University of Warsaw, Institute of Archaology. Dothan, M. 1983. Hammath Tiberias. Early Synagogues and the Hellenistic and Roman Remains. Jerusalem, Israel Exploration Society. Dothan, M. and Freedman, D. N. 1967. Ashdod I. The First Season of Excavation 1962. ‘Atiqot VII. Jerusalem, Department of Antiquities and Museums. Dragendorff, H. 1897. Zur Terrasigillataindustrie in Griechenland, Kleinasien, Südrussland und Aegypten. Bonner Jahrbücher des Rheinischen Landesmuseums in Bonn 101, 140-152. Drexhage, H.-W. 2007. Wirtschaftspolitik und Wirtschaft in der römischen Provinz Asia in der Zeit von Augustus bis zum Regierungsantritt Diokletians. Asia Minor Studien 59. Bonn, Rudolf Habelt. Du Plat Taylor, J. 1980. Excavations at Ayios Philon, The Ancient Carpasia Part 1. The Classical to Roman periods. Report of the Department of Antiquities, Cyprus, 152-216. Du Plat Taylor, J. and Megaw, A. H. S. 1981. Excavations at Ayios Philon, The Ancient Carpasia Part 2. The Early Christian buildings. Report of the Department of Antiquities, Cyprus, 209-250. Duman, B. 2014. Red slip pottery from Laodicea/Phrygia. Rei Cretariae Romanae Fautorum Acta 43, 9-17. Abingdon, Rei Cretariae Romanae Fautores. Dunand, M. and Duru, R. 1962. Céramique. In M. Dunand and R. Duru, Oumm el-‘Amed. Une ville de l’epoque hellénistique aux échelles de Tyr. Études et Documents d’Archéologie Tome 4, 197-211. Paris, Librairie d’Amérique et d’Orient. Dunham, D. 1957. The royal cemeteries of Kush 4: Royal tombs at Meroë and Barkal. Boston, Museum of Fine Arts. Durliat, J. and Guillou, A. 1984. Le tarif d’Abydos, Bulletin de correspondance hellénique 108, 581-98. Egloff, M. 1977. Kellia. La potterie copte. Quatre siècles d’artisanat et d’échanges en basse-Egypte Tome 1. Recherches suisses d’archéologie copte Volume III. Genève, Georg. Eilers, C. 2002. Roman patrons of Greek cities. Oxford, Oxford University Press. Eiring, L. J. 2004. The earliest Italian sigillata at Knossos. In Poblome et al. 2004, 71-72. Eisenmenger, U. 2003. Late Roman pottery in Limyra (Lycia). In C. Abadie-Reynal (ed.), Les céramiques an Anatolie aux époques hellénistique et romaine. Actes de la Table Ronde d’Istanbul, 22-24 mai 1996. Varia Anatolica 15, 193-196. Paris, De Boccard. Eiwanger, J. 1981. Keramik und Kleinfunde aus der Damokratia-Basilika in Demetrias Teil I & II. Demetrias IV. Bonn, Rudolf Habelt. Élaigne, S. 1998. Alexandrie, Étude préliminaire d’un contexte céramique du Haut-Empire. In J.-Y. Empereur

(ed.), Alexandrina 1. Études alexandrines 1, 75-114. Cairo, Institut Français d’Archéologie Orientale. Élaigne, S. 1999. Le passage des vernis noirs aux vernis rouges. Contribution à l’étude des céramiques fines de Méditerranée orientale, à la lumière du corpus alexandrin. Topoi 9, 219-228. Élaigne, S. 2000. Fine ware from the Late Hellenistic (second half of the first century BC), Augustan and Tiberian deposits of the French excavations in Alexandria (Egypt). Rei Cretariae Romanae Fautorum Acta 43, 19-30. Abingdon, Rei Cretariae Romanae Fautores. Élaigne, S. 2004. L’apport italique de sigillées en Egypte au début du Haut Empire: le cas d’Alexandrie et de Coptos. In Poblome et al. 2004, 133-144. Empereur, J.-Y. (ed.) 1998. Commerce et artisanat dans l’Alexandrie héllenistique et romaine. Actes de la Table Ronde organisée par le CNRS, le Laboratoire de Céramologie de Lyon et l’ÉFA, Athènes, 10-12 octobre 1988. Bulletin de correspondance hellénique suppléments 33. Paris, École française d’Athènes. Empereur, J.-Y. and Picon, M. 1986. A propos d’un nouvel atelier de “Late Roman C”. Figlina 7, 143-146. Empereur, J.-Y. and Picon, M. 1989. Les régions de production d’amphores impériales en Méditerranée orientale. In Amphores romaines et histoire économique: dix ans de recherche. Collection de l’École française de Rome 114, 223-248. Rome, École française de Rome. Enden, van der, M., Poblome, J. and Bes, P. M. 2014. From Hellenistic to Roman imperial in Pisidian tableware: the genesis of Sagalassos Red Slip Ware. In H. Meyza (ed.), Late Hellenistic to mediaeval fine wares of the Aegean coast of Anatolia. Their production, imitation and use. Travaux de L’Institut des Cultures Méditerranéennes et Orientales de L’Académie Polonaises des Sciences Tome 1, 81-93. Warsaw, Éditions Neriton. Enden, van der, M., Poblome, J. and Bes, P. M. In press. Sagalassian mastoi in an eastern Mediterranean context. In S. Drougou (ed.), 9th International Scientific Meeting on Hellenistic Pottery, Thessaloniki 2012. Enden, van der, M. and Bes, P. M. Forthcoming. The Hellenistic and Roman pottery from Koroneia. Preliminary observations. Erdkamp, P. 2005. The grain market in the Roman Empire. A social, political and economic study. Cambridge, Cambridge University Press. Erol, D. 2007. Smyrna Devlet Agora buluntusu terra sigillataları. In B. Böhlendorf-Arslan, A. O. Uysal and J. Witte-Orr (eds.), Çanak. Late Antique and Medieval pottery and tiles in Mediterranean archaeological contexts. Proceedings of the First International Symposium on Late Antique, Byzantine, Seljuk and Ottoman pottery and tiles in archaeological context (Çanakkale, 1-3 June 2005). BYZAS 7, 73-95. Istanbul, Ege Yayınları. Etienne, R. and Braun, J.-P. 1986. La céramique. In R. Etienne and J.-P. Braun, Ténos I. Le sanctuaire de

159

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:44 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East Poseidon et d’Amphitrite. Bibliothèque des Écoles Françaises d’Athènes et de Rome Fascicule 263, 201231. Paris, De Boccard. Ettlinger, E. 1987. How was Arretine Ware sold? Rei Cretariae Romanae Fautorum Acta 25/26, 5-19. München, Rei Cretariae Romanae Fautores. Ettlinger, E., Hedinger, B., Hoffmann, B., Kenrick, P. M., Pucci, G., Roth-Rubi, K., Schneider, G., Von Schnurbein, S., Wells, C. M. and ZabehlickyScheffenegger, S. 1990. Conspectus Formarum Terrae Sigillatae Italico Modo Confectae. Materialien zur römisch-germanischen Keramik 10. Bonn, Rudolf Habelt. Evans, J. A. S. 1996: The age of Justinian. The circumstances of imperial power. London, Routledge. Faiers, J. 2005. Late Roman pottery at Amarna and related studies. Excavation Memoir 72. London, Egypt Exploration Society. Feig, N. 1985. Pottery, glass and coins from Magen. Bulletin of the American School of Oriental Research 258, 33-40. Felten, F., Hiller, S., Reinholdt, C., Gauss, W. and Smetana, R. 2003. Ägina-Kolonna 2002. Vorbericht über die Grabungen des Instituts für Klassische Archäologie der Universität Salzburg. Jahreshefte des Österreichischen Archäologisches Institutes in Wien 72, 41-65. Fenn, N. 2008. Thin-Walled pottery from an Early Imperial complex at Priene/Ionia. Rei Cretariae Romanae Fautorum Acta 40, 249-254. Bonn, Rei Cretariae Romanae Fautores. Fentress, E. and Perkins, P. 1988. Counting African Red Slip Ware. In A. Mastino (ed.), L’Africa Romana: Atti del V Convegno di studio Sassari, 11-13 dicembre 1987, 205-214. Sassari, Università degli studi di Sassari. Fentress, E., Fontana, S., Bruce Hitchner, R. and Perkins, P. 2004. Accounting for ARSW: fineware and sites in Sicily and Africa. In S. E. Alcock and J. F. Cherry (eds.), Side-by-side survey. Comparative regional studies in the Mediterranean World, 147-162. Oxford, Oxbow Books. Ferrandes, A. F. 2013. Le ceramiche fini. In M. Almonte, Cossyra II. Ricognizione topografica/Storia di un paesaggio mediterraneo. Tübinger Archäologische Forschungen 11, 377-408. Rahden, Marie Leidorf. Ferrazzoli, A. F. 2003a. Catalogo dei reperti ceramici. In E. E. Schneider (ed.), Elaiussa Sebaste 2: un porto tra Oriente e Occidente Volume 1. Bibliotheca Archaeologica 37, 144-179. Rome, L’Erma di Bretschneider. Ferrazzoli, A. F. 2003b. Instrumentum domesticum. Tipologia dei reperti ceramici e aspetti della produzioni e della circolazione dei materiali. In E. E. Schneider (ed.), Elaiussa Sebaste 2: un porto tra Oriente e Occidente Volume 1. Bibliotheca Archaeologica 37, 649-707. Rome, L’Erma di Bretschneider. Fırat, N. 2000. So-called “Cypriot Red Slip Ware” from the habitation quarter at Perge (Pamphylia). Rei Cretariae Romanae Fautorum Acta 36, 35-38. Abingdon, Rei Cretariae Romanae Fautores.

Fırat, N. 2003. Roman period ceramics. In H. Abbasoğlu and W. Martini (eds.), Die Akropolis von Perge 1. Survey und Sondagen 1994-1997, 123-130. Mainz, Philipp von Zabern. Fischer-Genz, B., Gerber, Y. and Hamel, H. (eds.) 2014. Roman pottery in the Near East. Local production and regional trade. Proceedings of the Round Table held in Berlin, 19-20 February 2010. Roman and Late Antique Mediterranean Pottery 3. Oxford, Archaeopress. Fisher, C. S. 1938. Tombs. The southwest cemetery. In C. H. Kraeling (ed.), Gerasa, City of the Decapolis, 549-571. New Haven, American Schools of Oriental Research. Forsell, R. 1996. The Roman period. In. B. Wells (in collaboration with C. Runnels), The Berbati-Limnes archaeological survey 1988-1990. Acta Instituti Atheniensis Regni Sueciae Series in 4° XLIV, 285343. Stockholm, Svenska Institutet i Athen. Forster, G. 2001. The Roman period. In J. N. Coldstream, J. Eiring and G. Forster, Knossos pottery handbook. Greek and Roman. British School at Athens Studies 7, 137-167. London, British School at Athens. Forster, G. 2005. The Roman city of Knossos, Colonia Julia Nobilis – the nature of the imported fine wares. In M. Berg Briese and L. E. Vaag (eds.), Trade relations in the Eastern Mediterranean from the Late Hellenistic period to Late Antiquity: the ceramic evidence. Acts from a Ph.D.-seminar for young scholars, Sandbjerg Manorhouse, 12-15 February 1998. Halicarnassian Studies 3, 124-131. Odense, University Press of Southern Denmark. Foss, C. 1996. Cities, fortresses and villages of Byzantine Asia Minor. Variorum Collected Studies Series 538. Aldershot, Variorum. Frankel, R., Getzov, N., Aviam, M. and Degani, A. 2001. Settlement dynamics and regional Diversity in ancient Upper Galilee. Archaeological Survey of Upper Galilee. Israel Antiquities Authority Reports 14. Jerusalem, Israel Antiquities Authority. Freed, J. 1998. Pottery report. In C. M. Wells, M. Carroll, J. Freed and D. Godden, The construction of decumanus VI N and the economy of the early colony of Carthage. In J. T. Peña, J. J. Rossiter, A. I. Wilson, C. Wells, M. Carroll, J. Freed and D. Godden, Carthage Papers. The early colony’s economy, water supply, a public bath, and the mobilization of state olive oil. Journal of Roman Archaeology Supplementary Series 28, 18-63. Portsmouth, Journal of Roman Archaeology. Frost, F. J. and Hadjidaki, E. 1990. Excavations at the harbor of Phalasarna in Crete: the 1988 season. Hesperia 59, 513-527. Fulford, M. G. 1984. The red-slipped wares. In M. G. Fulford and D. P. S. Peacock, The Avenue du President Habib Bourguiba, Salammbo: the pottery and other ceramic objects from the site. Excavations at Carthage: the British Mission 1.2, 48-115. Sheffield, University of Sheffield.

160

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:44 UTC All use subject to https://about.jstor.org/terms

Bibliography Fulford, M. G. 1987. Economic interdependence among urban communities of the Roman Mediterranean. World Archaeology 19, 58-75. Fulford, M. G. and Timby, J. 1994. The fine wares. In M. G. Fulford and D. P. S. Peacock, The circular harbour, north side. The pottery. Excavations at Carthage. The British Mission 2.2, 1-21. Oxford, Oxford University Press. Fulford, M. G. and Tomber, R. S. (eds.) 1994. The finewares and lamps. Excavations at Sabratha 19481951. Volume II: the finds, part 2. Society for Libyan Studies Monograph 3. Tripoli, Society for Libyan Studies. Fülle, G. 1997. The internal organization of the Arretine Terra Sigillata industry: problems of evidence and interpretation. The Journal of Roman Studies 87, 111155. Garnett, D. 2011. Appendix II. Geochemical characteristics of imported Hellenistic pottery from Jebel Khalid, Antioch, and Pella. In Jackson and Tidmarsh 2011, 527-543. Gassner, V. 1997. Das Südtor der Tetragonos-Agora. Keramik und Kleinfunde. Forschungen in Ephesos 13/1/1). Vienna, Verlag der Österreichischen Akademie der Wissenschaften. Gebara, Ch. 1982. Remarques sur la sigillée orientale d’après les fouilles de Khan Khaldé (Heldua). In Archéologie au Levant. Recueil à la mémoire de Roger Saidah. Collection de la Maison de l’Orient Méditerranéen 12. Serie Archéologique 9, 409-417. Lyon, Maison de l’Orient. Gichon, M. 1993. Die Keramik des Kastells, Katalog der Kastell-Keramikfunde. In M. Gichon, En Boqeq. Ausgrabungen in einer Oase am Toten Meer. Band I. Geograpie und Geschichte der Oase das spätrömischbyzantinische Kastell, 129-387. Mainz, Philipp von Zabern. Guidice, E. and Narbone, G. 2000. Terre sigillata ellenistico-romane. In F. Guidice, E. Guidice and G. Guidice, Paphos, Garrison’s Camp. VIIa Campagna. Report of the Department of Antiquities, Cyprus, 270275. Guidice, E. and Narbone, G. 2001. Terre Sigillate ellenistico-romane. In F. Guidice, G. Guidice and E. Guidice, Paphos, Garrison’s Camp. VIIIa campagna. Report of the Department of Antiquities, Cyprus, 266274. Guidice, E., Guzzetta, G., Malfitana, D., Narbone, G., Oliveri, R.A., Privitera, A., Puppo, P., Rizzone, V. G. 1999. Catalogo dei pezzi selezionati dal Q. 102. In F. Guidice and G. Guidice, Pafos, Garrison’s Camp. VIa Campagna. Report of the Department of Antiquities, Cyprus, 283-301. Guidice, E., Malfitana, D., Narbone, C. and Privitera, A. 1996. Terre sigillate ellenistico-romane e lucerne. Catalogo. In F. Guidice, Paphos, Garrison’s camp. Campagna 1991. Report of the Department of Antiquities, Cyprus, 212-236.

Given, M. and Knapp, A. B. 2003. The Sydney Cyprus Survey Project: social approaches to regional archaeological survey. Monumenta Archaeologica 21. Los Angeles, Cotsen Institute of Archaeology. Goethert, K.-P. 1985. Leihgaben und Neuerwerbungen der Original- und Abguß-Sammlung in den Jahren 1983 – 1984. In F. G. Maier, Alt-Paphos auf Cypern. Ausgrabungen zur Geschichte von Stadt und Heiligtum 1966-1984. Trierer Winckelmannsprogramme Heft 6, 55-65. Mainz, Philipp von Zabern. Goldman, H. (ed.) 1950. Excavations at Gözlü Kule, Tarsus. Volume I text and plates. The Hellenistic and Roman periods. Princeton, Princeton University Press. Gomez, B., Rautman, M. L., Neff, H., Glascock, G. D. 1996. Clays used in the manufacture of Cypriot red slip pottery and related ceramics. Report of the Department of Antiquities, Cyprus, 69-82. Goren, Y. and Fabian, P. 2008. The Oboda Potter’s workshop reconsidered. Journal of Roman Archaeology 21, 340-351. Greco, E., Savelli, S., Consoli, V., Camponetti, G., Lucchese, C., Berutti, S., Golin, M., Guastella, P., Carando, E., Privitera, S., Polosa, S., Dibenedetto, A., Vitti, P., Voza, O., Papi, E., Cerri, L., Farinetti, E. and Sebastiani, R. 2003. Hephaestia 2003. Annuario della Scuola Archeologica di Atene e delle Missioni italiane in Oriente 81 Series 3.3, Part 2, 1023-1099. Gregory, T. E. 1985. An Early Byzantine complex at Akra Sophia near Corinth. Hesperia 54, 411-128. Gregory, T. E. 1993. The Hexamilion and the fortress. Isthmia V. Princeton, Princeton University Press. Gregory, T. E. 1995. The Roman Bath at Isthmia. Preliminary report 1972-1992. Hesperia 64, 279-313. Grigoropoulos, D. 2009. Kaiserzeitliche und spätantike Keramik aus Attika in der Sammlung des Deutschen Archäologischen Instituts Athen. Mitteilungen des Deutschen Archäologischen Instituts, Athenische Abteilung 124, 403-485. Gschwind, M. 2002. Hellenistische Tradition contra Italische Mode: Ein frühkaiserzeitlichen Keramikkomplex aus den türkischen Rettungsgrabungen in Zeugma am mittleren Euphrat. Damaszener Mitteilungen 13, 321359. Gunneweg, J., Perlman, I. and Yellin, J. 1983. The provenience, typology and chronology of Eastern Terra Sigillata. Qedem 17. Jerusalem, Hebrew University of Jerusalem. Hagn, T. 2003. Italischer Import oder lokal-regionale “Imitation”? Sigillata aus der Gymnasium-TycheionGrabung von Aigeira. Rei Cretariae Romanae Fautorum Acta 38, 265-269. Abingdon, Rei Cretariae Romanae Fautores. Handberg, S., Stone, P. J. and Hjarl Petersen, J. 2012. Uncommon tastes: the consumption of Campana A pottery in the southern Levant and the Black Sea region. Acta Hyperborea 13, 51-80. Hansen, S. L. 2003. The Roman and Late Roman fine wares from the Danish excavations of a Late Roman villa in Halikarnassos. In C. Abadie-Reynal (ed.), Les

161

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:44 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East céramiques an Anatolie aux époques hellénistique et romaine. Actes de la Table Ronde d’Istanbul, 22-24 mai 1996. Varia Anatolica 15, 197-201. Paris, De Boccard. Harper, R. P. 1980. Athis – Neocaesareia – Qasrin – Dibsi Faraj. In J. Cl. Margueron (ed.), Le moyen Euphrate. Zone de contacts et d’échanges. Actes du Colloque de Strasbourg 10-12 mars 1977. Travaux du Centre de Recherches sur le Proche-Orient et la Grèce Antiques 5, 327-348. Leiden, Brill. Harper, R. P. 1995. The pottery. In R. P. Harper, Upper Zohar. An Early Byzantine fort in Palaestina Tertia. Final report of excavations in 1985-1986. British Academy Monographs in Archaeology 9, 21-33. Oxford, Oxford University Press for the British School of Archaeology in Jerusalem. Harris, W. V. 2011. Rome’s imperial economy: twelve essays. Oxford, Oxford University Press. Harrison, G. W. M. and Hayden, B. J. 2005. Roman pottery summary. In B. J. Hayden, The Vrokastro regional survey project sites and pottery: Reports on the Vrokastro area, eastern Crete. Volume 3. University Museum Monograph 123, 57-79. Pennsylvania, University of Pennsylvania. Hartal, M. 2009. Finds from the northwestern suburb. In M. Hartal, Paneas Volume IV. The Aqueduct and the northern Suburbs. Israel Antiquities Authority Reports 40, 161-178. Jerusalem, Israel Antiquities Authority. Hatcher, H., Hedges, R. E. M., Pollard, A. M., and Kenrick, P. M. 1980. Analysis of Hellenistic and Roman fine pottery from Benghazi. Archaeometry 22, 133-151. Hawthorne, J. W. J. 1997. Post processual economics: the role of African Red Slip Ware vessel volume in Mediterranean demography. In K. Meadows, C. Lemke and J. Heron (eds.), TRAC 96: Proceedings of the sixth annual Theoretical Roman Archaeology Conference, Sheffield 1996, 29-37. Oxford, Oxbow Books. Hayes, J. W. 1967. “Cypriot Sigillata”. Report of the Department of Antiquities, Cyprus, 65-77. Hayes, J. W. 1971. Four Early Roman groups from Knossos. The Annual of the British School at Athens 66, 249-275. Hayes, J. W. 1972: Late Roman Pottery. London, The British School at Rome. Hayes, J. W. 1973a. Roman Pottery from the South Stoa at Corinth. Hesperia 42, 416-470. Hayes, J. W. 1973b. The pottery. In J. Boardman and J. W. Hayes, Excavations at Tocra 1963-1965. The Archaic deposits II and later deposits. British School at Athens Supplementary Volume 10, 108-119. London, British School of Archaeology at Athens. Hayes, J. W. 1976a. Pottery: stratified groups and typology. In J. H. Humphrey (ed.), Excavations at Carthage 1975. Conducted by the University of Michigan Volume 1, 47-108. Tunis, Cérès Publications. Hayes, J. W. 1976b [1978]. Appendix two. Notes on the Palace pottery. In R. G. Goodchild, J. G. Pedley and D. White (ed. J. H. Humphrey), Apollonia, the port of Cyrene. Excavations by the University of Michigan

1965-1967. Supplements to Libya Antiqua 4, 260-265. Tripoli, The Department of Antiquities. Hayes, J. W. 1977. Pottery report 1976. In J. H. Humphrey (ed.), Excavations at Carthage 1976. Conducted by the University of Michigan Volume 4, 23-98. Ann Arbor, University of Michigan Press. Hayes, J. W. 1978. Appendix 1: selected deposits (continued). In J. H. Humphrey (ed.), Excavations at Carthage 1975. Conducted by the University of Michigan Volume 2, 113-118. Ann Arbor, American Schools of Oriental Research. Hayes, J. W. 1980. A supplement to Late Roman pottery. London, The British School at Rome. Hayes, J. W. 1983. The Villa Dionysos excavations, Knossos: the pottery. The Annual of the British School at Rome 78, 97-169. Hayes, J. W. 1985a. Sigillate orientale. In Atlante della forme ceramiche II: ceramica fine romana nel bacino mediterraneo (tardo ellenismo e primo imperio). Enciclopedia dell’arte antica. Classica e orientale, 1-96. Rome, Istituto della Enciclopedia Italiana. Hayes, J. W. 1985b. Hellenistic to Byzantine fine wares and derivatives in the Jerusalem corpus. In A. D. Tushingham (ed.), Excavations in Jerusalem 19611967 1, 179-194. Toronto, Royal Ontario Museum. Hayes, J. W. 1991. Paphos III. The Hellenistic and Roman pottery. Nicosia, The Department of Antiquities. Hayes, J. W. 1992. Excavations at Saraçhane in Istanbul 2. The pottery. Lawrenceville, Princeton University Press. Hayes, J. W. 1995a. An early Roman well group from the Troia excavations 1992. Studia Troica 5, 185-196. Hayes, J. W. 1995b. Pottery from the construction levels of the upper cavea. In G. B. Waywell and J. J. Wilkes, Excavations at the ancient theatre of Sparta 1992-4: preliminary report. The Annual of the British School at Athens 90, 449-451. Hayes, J. W. 1996a. Recent work on Roman imported and local pottery from the Athenian Agora and the Isthmian sanctuary. In M. Herford-Koch, U. Mandel and U. Schädler (eds.), Hellenistische und kaiserzeitliche Keramik des östlichen Mittelmeergebietes. Kolloquium Frankfurt 24-25 April 1995, 7-17. Frankfurt, Archäologisches Institut der Johann Wolfgang GoetheUniversität. Hayes, J. W. 1996b. The pottery. In S. E. Sidebotham and W. Z. Wendrich (eds.), Berenike 1995. Preliminary report of the 1995 excavations at Berenike (Egyptian Red Sea coast) and the survey of the Eastern Desert. CNWS Publications Special Series 2, 147-178. Leiden, Centre of Non-Western Studies. Hayes, J. W. 1997. Handbook of Mediterranean Roman pottery. London, University of Oklahoma Press. Hayes, J. W. 1998a. The pottery assemblage of Late Hellenistic date. In E. R. Gebhard and F. P. Hemans, University of Chicago excavations at Isthmia: 2. Hesperia 67, 60-62. Hayes, J. W. 1998b. The pottery assemblage of Roman date. In E. R. Gebhard, F. P. Hemans and J. W. Hayes,

162

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:44 UTC All use subject to https://about.jstor.org/terms

Bibliography University of Chicago excavations at Isthmia 1989: 3. Hesperia 67, 444-454. Hayes, J. W. 2000a. Roman pottery from the Sanctuary. In J. W. Shaw and M. C. Shaw (eds.), The Greek sanctuary. Kommos 4, 312-320. Princeton, Princeton University Press. Hayes, J. W. 2000b. From Rome to Beirut and beyond: Asia Minor and eastern Mediterranean trade connections. Rei Cretariae Romanae Fautorum Acta 36, 285-297. Abingdon, Rei Cretariae Romanae Fautores. Hayes, J. W. 2001. Late Roman fine wares and their successors: a Mediterranean Byzantine perspective (with reference to the Syro-Jordanian situation). In Villeneuve and Watson (eds.) 2001, 275-282. Hayes, J. W. 2003. Hellenistic and Roman pottery deposits from the ‘Saranda Kolones’ castle site at Paphos. The Annual of the British School at Athens 98, 447-516. Hayes, J. W. 2005. Late Hellenistic and Roman pottery in the eastern Mediterranean – an overview of recent developments. In M. Berg Briese and L. E. Vaag (eds.), Trade relations in the Eastern Mediterranean from the Late Hellenistic period to Late Antiquity: the ceramic evidence. Acts from a Ph.D.-seminar for young scholars, Sandbjerg Manorhouse, 12-15 February 1998. Halicarnassian Studies 3, 11-26. Odense, University Press of Southern Denmark. Hayes, J. W. 2007. Pottery. In A. H. S. Megaw†, Kourion. Excavations in the episcopal precinct. Dumbarton Oaks Studies 38, 435-475. Washington D.C., Dumbarton Oaks Research Library and Collection. Hayes, J. W. 2008a. Roman pottery. Fine ware imports: typology. The Athenian Agora 32. Princeton, The American School of Classical Studies at Athens. Hayes, J. W. 2008b. The Roman and Late Roman fine wares. In K. Prag (ed.), Discoveries in Hellenistic to Ottoman Jerusalem centenary volume: Kathleen M. Kenyon 1906-1978. Levant Supplementary Series 6. Excavations by K. M. Kenyon in Jerusalem 1961-1967 5, 379-387. Oxford, Oxbow Books. Hayes, J. W. 2008c. Tablewares, functional ceramics and ritual pots: creating a typology of Roman-period Athenian products. In S. Vlizos (ed.), Athens during the Roman period. Recent discoveries, new evidence. Mouseio Benaki Supplement 4, 439-447. Athens, Benaki Museum. Hayes, J. W. and Harlaut, C. 2002. Ptolemaic and Roman pottery deposits from Alexandria. In J.-Y. Empereur (ed.), Alexandrina 2. Études alexandrines 6, 99-131. Cairo, Institut Français d’Archéologie Orientale. Heath, S. 2011. Trading at the edge: pottery, coins, and household objects at Dura-Europos. In J. Y. Chi and S. Heath, Edge of empires. Pagans, Jews, and Christians at Roman Dura-Europos, 63-73. Princeton, Princeton University Press. Heberdey, R. 1906. Forschungen in Ephesos. Forschungen in Ephesos 1. Vienna, A. Holder. Heimberg, U. 1982. Terra Sigillata. In U. Heimberg, Die Keramik des Kabirions. Das Kabirenheiligtum bei Theben Band 3, 53-54. Berlin, De Gruyter.

Hellström, P. 1965. Labraunda. Swedish excavations and researches Volume 2 Part 1. Pottery of Classical and later date. Terracotta lamps and glass. Acta Instituti Atheniensis Regni Sueciae, Series 4°, 5, 2:1. Lund, Swedish Institute in Athens. Henrich, P. 2002. Studien zur römischen Keramik im Hauran – Untersuchungen am Beispiel eines Fundkomplexes aus Qanawat/Südsyrien. Damaszener Mitteilungen 13, 245-315. Hepding, H. 1952. Eine hellenistische Töpferwerkstatt in Pergamon. Nachrichten der Gießener Hochschulgesellschaft 21, 49-60. Herbert, S. C. and Berlin, A. M. 2003. The excavation: occupation history and ceramic assemblages. In S. C. Herbert and A. M. Berlin, Excavations at Coptos (Qift) in Upper Egypt 1987-1992. Journal of Roman Archaeology Supplementary Series 53, 12-131. Ann Arbor, Journal of Roman Archaeology. Hitchner, R. B. 1993. Olive production and the Roman economy: the case for intensive growth in the Roman Empire. In M.-C. Amouretti and J.-P. Brun (eds.), La production du vin et l’huile en Méditerranée. Bulletin de correspondance hellénique Supplément 26, 499508. Athens, École française d’Athènes. Hitchner, R. B. 2005. “The advantages of wealth and luxury”. The case for economic growth in the Roman Empire. In J. G. Manning and I. Morris (eds.), The ancient economy. Evidence and models, 207-222. Stanford, Stanford University Press. Hjohlman, J., Penttinen, A. and Wells, B. 2005. Pyrgouthi. A rural site in the Berbati Valley from the Early Iron Age to Late Antiquity. Excavations by the Swedish Institute at Athens 1995 and 1997. Acta Instituti Atheniensis Regni Sueciae, Series 4°, 52. Stockholm, Swedish Institute in Athens. Hochuli-Geysel, A. 1977. Kleinasiatische glasierte Reliefkeramik (50 v. Chr. bis 50 n. Chr.) und ihre oberitalische Nachahmungen. Acta Bernensia 7. Bern, Stämpfli Verlag. Hochuli-Geysel, A. 2002. La céramique à glaçure plombifère d’Asie Mineure et du bassin méditerranéen oriental (du Ier s. av. J.-C. au Ier s. ap. J.-C. In F. Blondé, P. Ballet and J.-F. Salles (eds.), Céramiques hellénistiques et romaines. Productions et diffusion en Méditerranée orientale (Chypre, Égypte et côte syro-palestinienne). Travaux de la Maison de l’Orient Méditerranéen 35, 303-319. Lyon, Maison de l’Orient. Hood, M. S. F. 1967. Some ancient sites in south-west Crete. The Annual of the British School at Athens 62, 47-56. Hood, M. S. F. 1970. Isles of refuge in the Early Byzantine period. The Annual of the British School at Athens 65, 37-45. Höpken, C., Schimmer, F. and Sieler, M. 2008. Ausgewählte Keramik aus dem Heiligtum des Iupiter Dolichenus und der Folgebesiedlung auf dem Dülük Baba Tepesi. In E. Winter (ed.), ΠΑΤΡΙΣ ΠΑΝΤΡΟΦΟΣ ΚΟΜΜΑΓΗΝΗ. Neue Funde und Forschungen zwischen Taurus und

163

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:44 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East Euphrat. Asia Minor Studien 60, 137-163. Bonn, Rudolf Habelt. Hopkins, K. 1980. Taxes and trade in the Roman Empire (200 B.C.-A.D. 400). The Journal of Roman Studies 70, 101-125. Hopkins, K. 1983. Introduction. In P. Garnsey, K. Hopkins and C. R. Whittaker (eds.), Trade in the ancient economy, ix-xxv. London, Chatto & Windus. Hornblower, S. and Spawforth, A. J. S. (eds.) 2003. The Oxford Classical Dictionary. Oxford, Oxford University Press. Horsfield, G. and Horsfield, A. 1942. Sela-Petra, the Rock of Edom and Nabatene 4. The Quarterly of the Department of Antiquities in Palestine 9, 105-204. Hübner, G. 1996. Die römische Keramik von Patras: Voraussetzungen und Möglichkeiten der Annäherung im Rahmen der Stadtgeschichte. In M. Herford-Koch, U. Mandel and U. Schädler (eds.), Hellenistische und kaiserzeitliche Keramik des östlichen Mittelmeergebietes. Kolloquium Frankfurt 24-25 April 1995, 1-5. Frankfurt, Archäologisches Institut der Johann Wolfgang Goethe-Universität. Hübner, G. 2003. Patras: Kreuzweg zwischen Ost und West. Die Sigillata-waren aus dem Fundgut χ 14, Korinthou 288/Kanari. Rei Cretariae Romanae Fautorum Acta 38, 257-264. Abingdon, Rei Cretariae Romanae Fautores. Iacomi, V. 2010. Some notes on Late-Antique oil and wine production in Rough Cilicia (Isauria) on the light of epigraphic sources: funerary inscriptions from Korykos, LR 1 amphorae production in Elaiussa Sebaste and the Abydos Tariff. In. Ü. Aydınoğlu and A. Kaan Şenol (eds.), Olive oil and wine production in Anatolia during Antiquity. Symposium proceedings 0608 November 2008, Mersin, Turkey, 19-32. Istanbul, Ege Yayınları. Illife, J. H. 1938. Sigillata wares in the Near East. The Quarterly of the Department of Antiquities in Palestine 6, 4-53. Iro, D., Schwaiger, H. and Waldner, A. 2009. Die Grabungen der Jahres 2005 in der Süd- und Nordhalle der Kuretenstrasse. Ausgewählte Befunde und Funde. In S. Ladstätter (ed.), Neue Forschungen zur Kuretenstraße von Ephesos. Akten des Symposiums für Hilke Thür vom 13. Dezember 2006 an der Österreichischen Akademie der Wissenschaften. Österreichische Akademie der Wissenschaften Philosophisch-historische Klasse Denkschriften 382. Archäologische Forschungen 15, 53-87. Vienna, Verlag der Österreichischen Akademie der Wissenschaften. Ivantchik, A. I. 2002. Un puits de l’époque paléochrétienne sur l’agora d’Argos. Bulletin de correspondance hellénique 126, 333-409. Jackson, H. and Tidmarsh, J. 2011. The pottery. Jebel Khalid on the Euphrates 3. Mediterranean Archaeology Supplement 7. Sydney, Meditarch. Japp, S. 2003. Sigillata und dünnwandige hartware aus Pergamon. Bemerkungen zu westlichen imported und stilistischer Einflussnahme. Rei Cretariae Romanae

Fautorum Acta 38, 243-246. Abingdon, Rei Cretariae Romanae Fautores. Japp, S. 2005. Die lokale Keramikproduktion von Kibyra. Rei Cretariae Romanae Fautorum Acta 39, 237-241. Abingdon, Rei Cretariae Romanae Fautores. Japp, S. 2007. Late Roman, Byzantine and Ottoman pottery from Alexandria Troas. In B. Böhlendorf-Arslan, A. O. Uysal and J. Witte-Orr (eds.), Çanak. Late Antique and Medieval pottery and tiles in Mediterranean archaeological contexts. Proceedings of the First International Symposium on Late Antique, Byzantine, Seljuk and Ottoman pottery and tiles in archaeological context (Çanakkale, 1-3 June 2005). BYZAS 7, 55-71. Istanbul, Ege Yayınları. Japp, S. 2009a. Archäometrisch-archäologische Untersuchungen an Keramik aus Pergamon und Umgebung. Istanbuler Mitteilungen 59, 193-268. Japp, S. 2009b. The local pottery production of Kibyra. Anatolian Studies 59, 95-129. Johansen, Ch. F. 1971. Les terres sigillées orientales. In A. P. Christensen and Ch. F. Johansen, Hama. Fouilles et recherches 1931-1938 3.2. Les poteries hellénistiques et les terres sigillées orientales. Nationalmuseets Skrifter 8, 55-208. Copenhagen, Nationalmuseet. Johnson, W. R. 1979. Roman pottery. In D. S. Whitcomb and J. H. Johnson, Quseir al-Qadim 1978: preliminary report, 67-103. Princeton, The American Research Center in Egypt. Johnson, B. L. 1988. The pottery. In G. D. Weinberg (ed.), Excavations at Jalame. Site of a glass factory in Late Roman Palestine, 137-226. Missouri, University of Missouri Press. Johnson, B. L. 2008a. Imported pottery of the Roman and Late Roman periods. Ashkelon 2. Winona Lake, Harvard Semitic Museum. Johnson, B. L. 2008b. The pottery. In J. Patrich, Archaeological excavations at Caesarea Maritima. Areas CC, KK and NN Final Report. Volume I: the objects, 19-206. Jerusalem, Israel Exploration Society. Jones, D. 2006. The bankers of Puteoli. Finance, trade and industry in the Roman world. Stroud, Tempus. Jones, F. F.1950. The pottery. In Goldman 1950, 149-296. Jones, F. F. 1969 [1971]. Sherds from Kululu. Anadolu/ Anatolia 13, 89-96. Jones, F. F. and Goell, T. 1963. Keramikfunde aus hellenistischer-römischer Zeit. In F. K. Dörner and T. Goell, Arsameia am Nymphaios. Die Ausgrabungen im Hierothesion des Mithradates Kallinikos von 19531956. Istanbuler Forschungen 23, 234-243. Berlin, Gebrüder Mann Verlag. Jongman, W. M. 2002. The Roman economy: from cities to empire. In L. de Blois and J. Rich (eds.), The transformation of economic life under the Roman Empire. Proceedings of the second workshop of the international network Impact of Empire (Roman Empire, c. 200 B.C. – A.D. 476). Nottingham, July 4-7, 2001, 28-47. Leiden, Brill. Jurišić, M. 2000. Ancient shipwrecks of the Adriatic. Maritime transport during the first and second

164

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:44 UTC All use subject to https://about.jstor.org/terms

Bibliography centuries AD. British Archaeological Reports International Series 828. Oxford, Archaeopress. Kadous, E. Z. H. 1988. Die Terra Sigillata in Alexandria. Untersuchungen zur Typologie der westlichen und östlichen Terra Sigillata des Hellenismus und der frühen Kaiserzeit. Unpublished PhD thesis, University of Trier. Kantor, H. J. 1958. The pottery. In L. S. Braidwood, H. Frankfort and C. W. McEwan, Soundings at Tell Fakhariyah. The University of Chicago Oriental Institute Publications 79, 21-41. Chicago, The University of Chicago Press. Kaptijn, E. 2009. Life on the watershed. Reconstructing subsistence in a steppe region using archaeological survey: a diachronic perspective on habitation in the Jordan Valley. Leiden, Sidestone. Karageorghis, V. 1986. 2. Musée regional de Limassol. Bulletin de correspondance hellénique 110, 828-837. Karagiorgou, O. 2001. LR2: a container for the military annona on the Danubian border? In S. A. Kingsley and M. J. Decker (eds.), Economy and exchange in the East Mediterranean during Late Antiquity. Proceedings of a conference at Somerville College, Oxford, 29th May, 1999, 129-166. Oxford, Oxbow Books. Karambinis, M. and Bes, P. M. Forthcoming. Late Roman pottery from the island of Skyros, Greece (a surface survey assemblage). Karivieri, A. 2003. Roman pottery. In J. Forsén and B. Forsén, The Asea Valley survey. An Arcadian mountain valley from the Palaeolithic period until modern times. Acta Instituti Atheniensis Regni Sueciae, Series 4°, 51, 275-289. Stockholm, Swedish Institute in Athens. Kazenwadel, B. 1995. Lidar Höyük: die hellenistische und römische Keramik. Unpublished PhD thesis, University of Heidelberg. Keay, S. 2012. The port system of imperial Rome. In S. Keay (ed.), Rome, Portus and the Mediterranean. Archaeological Monographs of the British School at Rome 21, 33-67. London, British School at Rome. Kenkel, F. 2007. The Cypriot Red Slip Ware and its derivatives from Pednelissos in Pisidia. In B. Böhlendorf-Arslan, A. O. Uysal and J. Witte-Orr (eds.), Çanak. Late Antique and Medieval pottery and tiles in Mediterranean archaeological contexts. Proceedings of the First International Symposium on Late Antique, Byzantine, Seljuk and Ottoman pottery and tiles in archaeological context (Çanakkale, 1-3 June 2005). BYZAS 7, 131-145. Istanbul, Ege Yayınları. Kenrick, P. M. 1978. Hellenistic and Roman fine wares. In J. Matthers, Tell Rifa’at 1977: preliminary report of an archaeological survey. Iraq 40, 147-156. Kenrick, P. M. 1981. Fine wares of the Hellenistic and Roman periods. In J. Matthers (ed.), The river Qoueiq, northern Syria, and its catchment. Studies arising from the Tell Rifa’at survey 1977-79 Part ii. British Archaeological Reports International Series 98 ii, 439458. Oxford, Archaeopress.

Kenrick, P. M. 1985. Excavations at Sidi Khrebish Benghazi (Berenice) 3,1. The fine pottery. Supplements to Libya Antiqua 5. Tripoli, Society for Libyan Studies. Kenrick, P. M. 1987. Part 3. Hellenistic and Roman fine wares. In D. White (ed.), The extramural sanctuary of Demeter and Persephone at Cyrene, Libya. Final Reports 3. University Museum Monograph 66, 1-18. Philadelphia, University of Pennsylvania Museum of Archaeology and Anthropology. Kenrick, P. M. 2000. Fine wares from the city wall section at Bait Nawashi (Area XLII). Archäologischer Anzeiger, 235-265. Kenrick, P. M. 2004. Signatures on Italian Sigillata: a new perspective. In Poblome et al. 2004, 253-262. Kenrick, P. M. 2006. Thirty six thousand pieces of signed Italian Sigillata: what next? In Malfitana et al. 2006, 65-71. Kenyon, K. M. 1957. Roman and later wares. 1. Terra sigillata, 2. Stratified groups. In Crowfoot et al. 1957, 281-306. Khairy, N. I. 1990. Terra Sigillata. In N. I. Khairy, The 1981 Petra excavations Volume 1. Abhandlungen des Deutschen Palästinavereins 13, 32-34. Wiesbaden, Harrassowitz. Kingsley, S. A. 1999. The Sumaqa pottery assemblage: classification and quantification, In Dar 1999, 263-329. Kingsley, S. A. 2001. The economic impact of the Palestinian wine trade in Late Antiquity. In S. A. Kingsley and M. J. Decker (eds.), Economy and exchange in the East Mediterranean during Late Antiquity. Proceedings of a conference at Somerville College, Oxford, 29th May, 1999, 44-68. Oxford, Oxbow Books. Kingsley, S. A. 2003. Late Antique trade: research methodologies & field practices. In L. A. Lavan and W. Bowden (eds.), Theory and practice in Late Antique archaeology. Late Antique Archaeology 1, 113-138. Leiden, Brill. Klynne, A. 2006. Consumption of Italian Sigillata: yesterday, today and tomorrow. In Malfitana et al. 2006, 167-174. Kögler, P. 2010. Feinkeramik aus Knidos vom mittleren Hellenismus bis in die mittlere Kaiserzeit (ca. 200 v.Chr. bis 150 n.Chr.). Wiesbaden, Reichert Verlag. Kondoleon, C. (ed.) 2000: Antioch. The lost ancient city. Princeton, Princeton University Press. Konrad, M. 1992. Flavische und spätantike Bebauung unter der Basilika B von Resafa-Sergiupolis. Damaszener Mitteilungen 6, 312-359. Konrad, M. 1996. Frühkaiserzeitliche Befestigungen an der Strata Diocletiana? Neue Kleinfunde des I. Jahrhunderts n. Chr. aus Nordsyrien. Damaszener Mitteilungen 9, 163-180. Konrad, M. 2001. Der spätrömische Limes in Syrien. Archäologische Untersuchungen an den Grenzkastellen von Sura, Tetrapyrgium, Cholle und in Resafa. Resafa 5. Mainz, Philipp von Zabern. Korkut, T. 2007. Die spätantike und frühbyzantinische Keramik aus Patara. In B. Böhlendorf-Arslan, A. O. Uysal and J. Witte-Orr (eds.), Çanak. Late Antique

165

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:44 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East and Medieval pottery and tiles in Mediterranean archaeological contexts. Proceedings of the First International Symposium on Late Antique, Byzantine, Seljuk and Ottoman pottery and tiles in archaeological context (Çanakkale, 1-3 June 2005). BYZAS 7, 147168. Istanbul, Ege Yayınları. Koutsoukou, A. and Najjar, M. 1997. The pottery. In A. Koutsoukou, K. W. Russell, M. Najjar and A. Momani, The Great Temple of Amman. The excavations. American Center of Oriental Research Occasional Papers 3, 55-118. Amman, American Center of Oriental Research. Kozal, E. 2001. Studies in Roman Ilion: the lower city. Stratified Domestic Assemblages. Studia Troica 11, 309-342. Kramer, N. 2004. Gindaros. Geschichte und Archäologie einer Siedlung im nordwestlichen Syrien von hellenistischer bis in frühbyzantinischer Zeit. Internationale Archäologie 41. Rahden, Marie Leidorf. Kramer, N. 2005. Ergebnisse des Keramiksurveys 2002/03. In D. Wannagat, Neue Forschungen in Uzuncaburç 2001-2001. Das Zeus-Olbios-Heiligtum und die Stadt Diokaisareia. Archäologischer Anzeiger, 125-128. Kramer, N. 2012. Keramik und Kleinfunde aus Diokaesareia. Diokaesareia in Kilikien. Ergebnisse des Surveys 2001-2006 Band 1. Berlin, De Gruyter. Kübler, K. 1931. Spätantike Stempelkeramik. Mitteilungen des Deutschen Archäologischen Instituts, Athenische Abteilung 56, 75-86. Ladstätter, S. 1998. Eine African Red Slip Schale mit Reliefverzierung aus Ephesos. Jahreshefte des Österreichischen Archäologisches Institutes in Wien 67, Beiblatt, 197-214. Ladstätter, S. (in Zusammenarbeit mit Bezecky, T., CzurdaRuth, B., Giuliani, A. and Rogl, C.) 2003. Keramik. In C. Lang-Auinger (ed.), Hanghaus 1 in Ephesos. Funde und Ausstattung, 22-85. Vienna, Verlag der Österreichischen Akademie der Wissenschaften. Ladstätter, S. 2005. Keramik. In H. Thür (ed.), Hanghaus 2 in Ephesos. Die Wohneinheit 4. Baubefund Ausstatung Funde. Forschungen in Ephesos 8/6, 230-358. Vienna, Verlag der Österreichischen Akademie der Wissenschaften. Ladstätter, S. 2007. Mode or politischen Manifest? Überlegungen zur Übernahme römischen Formenguts in der frühkaiserzeitlichen Keramik von Ephesos. In Meyer (ed.) 2007, 203-219. Ladstätter, S. 2008. Römische, spätantike und byzantinische Keramik. In M. Steskal and M. la Torre (eds.), Das Vediusgymnasium in Ephesos. Archäologie und Baubefund. Forschungen in Ephesos 14/1, 97-189. Vienna, Verlag der Österreichischen Akademie der Wissenschaften. Ladstätter, S. (ed.) 2009. Neue Forschungen zur Kuretenstraße von Ephesos. Akten des Symposiums für Hilke Thür vom 13. Dezember 2006 an der Österreichischen Akademie der Wissenschaften. Österreichische Akademie der Wissenschaften Philosophisch-historische Klasse Denkschriften 382.

Archäologische Forschungen 15. Vienna, Verlag der Österreichischen Akademie der Wissenschaften. Ladstätter, S. 2010a. Keramik. In F. Krinzinger (ed.), Hanghaus 2 in Ephesos. Die Wohneinheiten 1 und 2. Baubefund, Ausstatung, Funde. Forschungen in Ephesos 8/8, 172-279. Vienna, Verlag der Österreichischen Akademie der Wissenschaften. Ladstätter, S. 2010b. Keramik. In M. Steskal, Das Prytaneion in Ephesos. Forschungen in Ephesos 9/4, 85-172. Vienna, Verlag der Österreichischen Akademie der Wissenschaften. Ladstätter, S. and High, N. M. 2010. Katalog. In M. Steskal, Das Prytaneion in Ephesos. Forschungen in Ephesos 9/4, 103-172. Vienna, Verlag der Österreichischen Akademie der Wissenschaften. Ladstätter, S. and Sauer, R. 2002. Late Roman C Ware in Ephesos. The significance of imported and local production by petrological and mineralogical methods. In V. Kilikoglou, A. Hein and Y. Maniatis (eds.), Modern trends in scientific studies on ancient ceramics. Papers presented at the Fifth European Meeting on Ancient Ceramics, Athens 1999. British Archaeological Reports International Series 1011, 323334. Oxford, Archaeopress. Ladstätter, S. and Sauer, R. 2005. Late Roman C-Ware und lokale spätantike Feinware aus Ephesos. In F. Krinzinger (ed.), Spätantike und mittelalterliche Keramik aus Ephesos. Österreichische Akademie der Wissenschaften Philosophisch-historische Klasse Denkschriften, 332. Band. (Archäologische Forschungen 13, 143-201. Vienna, Verlag der Österreichischen Akademie der Wissenschaften. Lamb, W. and Pryce, F. N. 1940. Sigillata from Lesbos. The Journal of Hellenic Studies 60, 96-98. Landgraf, J. 1980. Keisan’s Byzantine pottery. In J. Briend and J.-B. Humbert (eds.), Tell Keisan, 1971-1976: Une cité phénicienne en Galilée, 51-99. Fribourg, Éditions universitaires. Langlotz, E. 1970. Beobachtungen in Phokaia. Archäologischer Anzeiger, 377-385. Lauffray, J. 1940. Une fouille au pied de l’acropole de Byblos. Bulletin du Musée de Beyrouth 4, 7-36. Lawson, J. 1996. The Roman pottery, In W. Cavanagh, J. Crouwel, R. W. V. Catling and G. Shipley, Continuity and change in a Greek rural landscape. The Laconia survey Volume 2. Archaeological data. British School at Athens Supplementary Volume 27, 111-123. London, British School of Archaeology at Athens. Leidwanger, J., Knappett, C., Arnaud, P., Arthur, P., Blake, E., Broodbank, C., Brughmans, T., Evans, T., Graham, S., Greene, E. S., Kowalzig, B., Mills, B., Rivers, R., Tartaron, T. F. and van de Noort, R. 2014. A manifesto for the study of ancient Mediterranean maritime networks. Antiquity 342 (DOI: http://journal.antiquity. ac.uk/projgall/leidwanger342) (Accessed 23 January 2015). Lenoir, M. 2003. Sa’neh ou le desert des tartares: un camp oublié du Limes Arabicus. Syria 80, 139-150.

166

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:44 UTC All use subject to https://about.jstor.org/terms

Bibliography Levick, B. 1967. Roman Colonies in Southern Asia Minor. Oxford, Oxford University Press. Lewit, T. 2011. Dynamics of fineware production and trade in the 1st to 7th centuries AD: the puzzle of supraregional exporters. Journal of Roman Archaeology 24, 313-332. Liebeschuetz, J. H. W. G. 1997. Cities, taxes and the accommodation of the barbarians. In W. Pohl (ed.), Kingdoms of the Empire. The Integration of Barbarians in Late Antiquity. The Transformation of the Roman world 1, 135-151. Leiden, Brill. Liebeschuetz, J. H. W. G. 2001. The decline and fall of the Roman city. Oxford, Oxford University Press. Lippolis, E. 2001. Terra Sigillata orientale. In A. DiVita (ed.), Gortina V.3. Lo scavo del pretorio (1989-1995). Volume V.3 t. I. I materiali. Monografie della Scuola Archeologica italiana di Atene e delle Missioni italiane in Oriente 12, 26-35. Padua, Aldo Ausilio. Lloyd, S. 1954. Sultantepe, Part 2. Anatolian Studies 4, 101-110. Loeschke, S. 1912. Sigillata-töpfereien in Tschandarli. Mitteilungen des Deutschen Archäologischen Instituts, Athenische Abteilung 37, 342-407. Loffreda, S. 1974. Cafarnao II. La ceramica. Pubblicazioni dello Studium Biblicum Franciscanum 19. Jerusalem, Franciscan Printing Press. Loffreda, S. 2008a. Tipologie e contesti stratigrafici della ceramica (1968-2003). Cafarnao VI. Jerusalem, Edizioni Terra Santa. Loffreda, S. 2008b. Documentazione grafica della ceramica (1968-2003). Cafarnao VII. Jerusalem, Edizioni Terra Santa. Loffreda, S. 2008c. Documentazione fotografica degli oggetti (1968-2003). Cafarnao VIII. Jerusalem, Edizioni Terra Santa. Love, I. C. 1969. Ceramics. In K. Lehmann and P. Williams-Lehmann (eds.), The Hieron. Samothrace 3.2, 143-236. Princeton, Princeton University Press. Lund, J. 1992. Centuries of darkness? A ceramic sidelight on Cyprus and the eastern Mediterranean between A.D. 200 and A.D. 350. In P. Aström (ed.), Acta Cypria Part 2. Acts of an International Congress on Cypriote archaeology held in Göteborg on 22-24 August 1991. Studies in Mediterranean Archaeology and Literature Pocket-book 117, 193-213. Jonsered, Paul Åströms Förlag. Lund, J. 1993. Pottery of the Classical, Hellenistic and Roman periods. In L. W. Sørensen and D. W. Rupp (eds.), The land of the Paphian Aphrodite Volume 2. The Canadian Palaipaphos survey project, Artifact and ecofactual studies. Studies in Mediterranean Archaeology 104:2, 79-155. Göteborg, Paul Åströms Förlag. Lund, J. 1995. A fresh look at the Roman and Late Roman fine wares from the Danish excavations at Hama, Syria. In H. Meyza and J. Mlynarczyk (eds.), “Hellenistic and Roman pottery in the Eastern Mediterranean – Advances in scientific studies” Acts of the 2 Nieborów pottery workshop, Nieborów, 18-20 december 1993, 135-161. Warsaw, Polish Academy of Sciences.

Lund, J. 1996a. Fine wares from the Hellenistic through the Late Antique periods. In L. W. Sørensen, Preliminary report of the Danish archaeological excavations to Panayia Ematousa, Aradippou 1993 and 1994. Report of the Department of Antiquities, Cyprus, 144-155. Lund, J. 1996b. From archaeology to history? Reflections on the chronological distribution of ceramic fine wares in south western and southern Asia Minor from the 1st to the 7th c. AD. In M. Herford-Koch, U. Mandel and U. Schädler (eds.), Hellenistische und kaiserzeitliche Keramik des östlichen Mittelmeergebietes. Kolloquium Frankfurt 24-25 April 1995, 105-125. Frankfurt, Archäologisches Institut der Johann Wolfgang GoetheUniversität. Lund, J. 1997. The distribution of Cypriot Sigillata as evidence of sea-trade involving Cyprus. In S. Swiny, R. L. Hohlfelder and H. W. Swiny (eds.), Res Maritimae. Cyprus and the Eastern Mediterranean from Prehistory to Late Antiquity. Proceedings of the second international symposium “Cities on the sea”. Nicosia, Cyprus, October 18-22, 1994. Cyprus American Archaeological Research Institute Monograph Series 1, 201-215. Atlanta, Scholars Press. Lund, J. 1998. The ceramic finewares from the Late Classical to the Late Antique period found in 1995 and 1996. In L. W. Sørensen, Third preliminary report of the Danish archaeological excavations at Panayia Ematousa, Aradippou, Cyprus. Proceedings of the Danish Institute at Athens 2, 332-348. Aarhus, Aarhus University Press. Lund, J. 1999. Trade patterns in the Levant from ca. 100 BC to AD 200 – as reflected by the distribution of ceramic fine wares in Cyprus. Münstersche Beiträge zur antiken Handelsgeschichte 18, 1-22. Lund, J. 2002. The ontogenesis of Cypriot Sigillata. In A. Rathje, M. Nielsen and B. Bundgaard Rasmussen (eds.), Pots for the living, pots for the dead. Acta Hyperborea 9, 185-223. Copenhagen, Museum Tusculanum Press. Lund, J. 2003. Eastern Sigillata B: a ceramic fine ware industry in the political and commercial landscape of the Eastern Mediterranean. In C. Abadie-Reynal (ed.), Les céramiques an Anatolie aux époques hellénistique et romaine. Actes de la Table Ronde d’Istanbul, 2224 mai 1996. Varia Anatolica 15, 125-136. Paris, De Boccard. Lund, J. 2004. Italian-made fine wares and cooking wares in the Eastern Mediterranean before the time of Augustus. In Poblome et al. 2004, 3-15. Lund, J. 2005. An economy of consumption: the Eastern Sigillata A industry in the Late Hellenistic period. In Z. H. Archibald, J. K. Davies and V. Gabrielsen (eds.), Making, moving and managing. The new world of ancient economies, 323-31 BC, 233-252. Oxford, Oxbow Books. Lund, J. 2006a. Ceramic fine wares from the 4th century BC to the 7th century AD. In: L. W. Sørensen and K. Winther-Jacobsen (eds.), Panayia Ematousa 1. Monographs of the Danish Institute at Athens 6,1, 182230. Athens, Danish Institute at Athens.

167

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:44 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East Lund, J. 2006b. On the circulation of goods in Hellenistic and Early Roman Cyprus: the ceramic evidence. In L. W. Sørensen and K. Winther-Jacobsen (eds.), Panayia Ematousa 2. Political, cultural, ethnic and social relations in Cyprus. Approaches to regional studies. Monographs of the Danish Institute at Athens 6,2, 3149. Athens, Danish Institute at Athens. Lund, J. 2006c. Writing long-term history with potsherds: problems – prospects. In Malfitana et al. 2006, 213227. Lund, J., Malfitana, D. and Poblome, J. 2006. Rhosica vasa mandavi (Cic., Att. 6.1.13). Towards the identification of a major ceramic tableware industry of the Eastern Mediterranean: Eastern Sigillata A. Archeologia Classica 57, 491-507. Mackensen, M. 1984. Eine befestigte spätantike Anlage vor den Stadtmauern von Resafa. Ausgrabungen and spätantike Kleinfunde eines Surveys in Umland von Resafa-Sergiupolis. Resafa 1. Mainz, Philipp von Zabern. Mackensen, M. 1993. Die spatäntiken Sigillata- und Lampentöpfereien von El Mahrine (Nordtunesien): Studien zur nordafrikanischen Feinkeramik des 4. bis 7. Jahrhunderts. Münchener Beiträge zur Vor- und Frühgeschichte 50. Munich, Verlags C. H. Beck. Mackensen, M. 1998a. Centres of African Red Slip Ware production in Tunisia from the late 5th to the 7th centuries. In L. Saguì (ed.), Ceramica in Italia: VIVII secolo. Atti del Convegno in Onore di John W. Hayes, Roma, 11-13 maggio 1995, 23-39. Archeologia medievale Biblioteca 14. Florence, All’insegna del giglio. Mackensen, M. 1998b. New evidence for Central Tunisian red slip ware with stamped decoration (ARS style D). Journal of Roman Archaeology 11, 355-370. Mackensen, M. 2003a. Production of 3rd century sigillata A/C (C) or ‘el-Aouja’ ware and its transition to sigillata C with appliqué decoration in Central Tunisia. Rei Cretariae Romanae Fautorum Acta 38, 279-286. Abingdon, Rei Cretariae Romanae Fautores. Mackensen, M. 2003b. Datierung und Provenienz einer spätantiken figürlichen Punze für nordafrikanischen Sigillata. Zur spätphase der Feinkeramikproduktion in Sidi Marzouk Tounsi (Zentraltunesien). Bayerischer Vorgeschichtsblätter 68, 101-108. Mackensen, M. 2006. The study of 3rd century African red slip ware based on the evidence from Tunisia. In Malfitana et al. 2006, 105-124. Mackensen, M. and Schneider, G. 2002. Production centres of African red slip ware (3rd-7th c.) in northern and central Tunisia: archaeological provenance and reference groups based on chemical analyses. Journal of Roman Archaeology 15, 121-158. Mackensen, M. and Schneider, G. 2006. Production centres of African Red Slip ware (2nd-3rd c.) in northern and central Tunisia: archaeological provenance and reference groups based on chemical analyses. Journal of Roman Archaeology 19, 163-190.

MacMullen, R. 2000. Romanization in the time of Augustus. New Haven, Yale University Press. Magnelli, A. 2001. Iscrizioni. In A. DiVita (ed.), Gortina V.3. Lo scavo del pretorio (1989-1995). Volume V.3 t. I. I materiali. Monografie della Scuola Archeologica italiana di Atene e delle Missioni italiane in Oriente 12, 626-654. Padua, Aldo Ausilio. Magness, J. 1990. Late Roman and Byzantine pottery, preliminary report, 1990. In R. L. Vann (ed.) Caesarea Papers. Straton’s Tower, Herod’s Harbour, and Roman and Byzantine Caesarea. Including the papers given at a symposium held at the University of Maryland, The Smithsonian Institution, and the Jewish Community Center of Greater Washington on 25-28 March, 1988. Journal of Roman Archaeology Supplementary Series 5, 129-153. Ann Arbor, Journal of Roman Archaeology. Magness, J. 1999. Redating the forts at Ein Boqeq, Upper Zohar, and other sites in SE Judaea, and the implications for the nature of the Limes Palaestinae. In J. H. Humphrey (ed.), The Roman and Byzantine Near East Volume II. Journal of Roman Archaeology Supplementary Series 31, 189-206. Portsmouth, Journal of Roman Archaeology. Magness, J. 2005. The Roman legionary pottery. In B. Arubas and H. Goldfus (eds.), Excavations on the site of the Jerusalem International Convention Center (Binyanei Ha’uma): a settlement of the late first to second Temple period, the tenth legion’s kiln works, and a Byzantine monastic complex. The pottery and other small finds. Journal of Roman Archaeology Supplementary Series 60, 69-191. Portsmouth, Journal of Roman Archaeology. Majcherek, G. and Taha, A. 2004. Roman and Byzantine layers at Umm el-Tlel: ceramics and other finds. Syria 81, 229-248. Malamidou, V. 2005. Roman pottery in context. Fine and coarse wares from five sites in north-eastern Greece. British Archaeological Reports International Series 1386. Oxford, Archaeopress. Malfitana, D. 1992. Ceramica romana. Catalogo. In F. Guidice, Paphos, Garrison’s camp, Campagna 1988. Report of the Department of Antiquities, Cyprus, 229237. Malfitana, D. 1993. Terree varia sigillate. Catalogo. In F. Guidice, Paphos, Garrison’s camp. Campagna 1989. Report of the Department of Antiquities, Cyprus, 297323. Malfitana, D. 1994. Terre sigillata ellenistico-romane e lucerne. Catalogo. In F. Guidice, Paphos, Garrison’s camp. Campagna 1990. Report of the Department of Antiquities, Cyprus, 244-255. Malfitana, D. 2002. Eastern terra sigillata wares in the Eastern Mediterranean. Notes on an initial quantitative analysis. In F. Blondé, P. Ballet and J.-F. Salles (eds.), Céramiques hellénistiques et romaines. Productions et diffusion en Méditerranée orientale (Chypre, Égypte et côte syro-palestinienne). Travaux de la Maison de l’Orient Méditerranéen 35, 133-157. Lyon, Maison de l’Orient.

168

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:44 UTC All use subject to https://about.jstor.org/terms

Bibliography Malfitana, D. 2004. The importation of stamped Italian Sigillata to Cyprus. In Poblome et al. 2004, 109-115. Malfitana, D., Guidice, E. and Narbone, G. 2004. I materiali. In F. Guidice, Pafos, Garrison’s camp. IXa Campagna. Report of the Department of Antiquities, Cyprus, 282-312. Malfitana, D., Poblome, J. and Lund, J. 2005. Eastern Sigillata A in Italy. A Socio-Economic Evaluation. Bulletin Antieke Beschaving 80, 199-212. Malfitana, D., Poblome, J. and Lund, J. (eds.) 2006. Old pottery in a new century. Innovating perspectives on Roman pottery studies. Atti del Convegno Internazionale di Studi Catania, 22-24 Aprile 2004. Monografie dell’Istituto per i Beni Archeologici e Monumentali, Consiglio Nazionale delle Richerche 1. Catania, L’Erma di Bretschneider. Marabini-Moevs, M. T. 1973. The Roman thin-walled pottery from Cosa (1948-1954). Memoirs of the American Academy in Rome 32. Rome, American Academy in Rome. Marabini-Moevs, M. T. 2006. Cosa VI: The Italian Sigillata. Memoirs of the American Academy in Rome Supplementary Volume III. Rome, The University of Michigan Press. Marangou-Lerat, A. 2002. La production céramique des ateliers crétois à l’époque romaine. In F. Blondé, P. Ballet and J.-F. Salles (eds.), Céramiques hellénistiques et romaines. Productions et diffusion en Méditerranée orientale (Chypre, Égypte et côte syro-palestinienne). Travaux de la Maison de l’Orient Méditerranéen 35, 67-71. Lyon, Maison de l’Orient. Marquié, S. 2002. La circulation des sigillées d’époque impériale au sud de Chypre. In F. Blondé, P. Ballet and J.-F. Salles (eds.), Céramiques hellénistiques et romaines. Productions et diffusion en Méditerranée orientale (Chypre, Égypte et côte syro-palestinienne). Travaux de la Maison de l’Orient Méditerranéen 35, 289-301. Lyon, Maison de l’Orient. Marquié, S. 2004. Un depot de la deuxième moitié du Ier s. de notre ère à Kition-Kathari (Chypre). In J. Eiring and J. Lund (eds.), Transport amphorae and trade in the Eastern Mediterranean. Acts of the International Colloquium at the Danish Institute at Athens, September 26-29, 2002. Monographs of the Danish Institute at Athens 5, 251-262. Athens, Danish Institute at Athens. Martelli, F. 1976. Terra sigillata arretina e tardo-italica a Cirene. In Quaderni di Archeologia della Libia 8 (Cirene e la Grecia), 431-434. Martin, A. 1995. Die Arbeit an der Kaiserzeitlichen Keramik. In U. Sinn, G. Ladstätter, A. Martin and T. Völling, Bericht über das Forschungsprojekt ,Olympia während der römischen Kaiserzeit und in der Spätantike’ IV Die Arbeiten im Jahr 1995. Nikephoros 8, 174-179. Martin, A. 1997a. Sigillata italica, Sigillata orientale A, Sigillata orientale B, Sigillata orientale C, Ceramica cnidia a rilievo. In A. DiVita and A. Martin (eds.), Gortina II. Pretorio. Il materiale degli scavi colini 1970-1977. Monografie della Scuola Archeologica

italiana di Atene e delle Missioni italiane in Oriente 7, 125-131, 223-224. Padua, Aldo Ausilio. Martin, A. 1997b. Roman and late antique fine wares from Olympia. Rei Cretariae Romanae Fautorum Acta 35, 211-216. Abingdon, Rei Cretariae Romanae Fautores. Martin, A. 1999. African Red Slip Ware. In D. Soren and N. Soren, A Roman villa and a Late Roman infant cemetery: excavation at Poggio Gramignano Lugnano in Teverina, 237-246. Rome, L’Erma di Bretschneider. Martin, A. 2001. The South-West Building at Olympia: the ceramic and glass evidence for its dating. In J.-Y. Marc and J.-C. Moretti (eds.), Constructions publiques et programmes edilitaires en Grèce entre le IIe siècle av. J.-C. et le Ier siècle ap. J.-C. Actes du Colloque Organisé par l’Ecole française d’Athènes et le CNRS, Athènes 14-17 mai 1995. Bulletin de correspondance hellénique Supplément 39, 313-322. Athens, École française d’Athènes. Martin, A. 2003. Western Sigillata in the Hanghäuser at Ephesos. Rei Cretariae Romanae Fautorum Acta 38, 247-249. Abingdon, Rei Cretariae Romanae Fautores. Martin, A. 2004. Italian sigillata in the East. Olympia: a case study. In Poblome et al. 2004, 67-70. Martin, A. 2006. Italian Sigillata in the East: two different models of supply (Ephesos and Olympia). In Malfitana et al. 2006, 175-187. Martin, A. 2008. Pottery from Schedia near Alexandria. Rei Cretariae Romanae Fautorum Acta 40, 263-269. Bonn, Rei Cretariae Romanae Fautores. Marty, J. 1993. Three pottery deposits and the history of Roman Isthmia. In T. E. Gregory (ed.), The Corinthia in the Roman period. Including papers given at a symposium held at Ohio State University on 7-9 March, 1991. Journal of Roman Archaeology Supplementary Series 8, 115-129. Ann Arbor, Journal of Roman Archaeology. Mattingly, D. J. 1995. Olive presses in roman Africa: technical evolution or stagnation? In M. Khanoussi, P. Ruggeri and C. Vismara (eds.), L’Africa romana, XI, 577-595. Sassari, Università degli studi di Sassari. Mayet, F. 1975. Les céramiques à parois fines dans la peninsula Ibérique. Publications de Centre Pierre Paris 1. Paris, De Boccard. Mayet, F. and Picon, M. 1986. Une sigillée phocéenne tardive (“Late Roman C Ware”) et sa diffusion en Occident. Figlina 7, 129-142. Mazzocchin, S. 2008. Missione archeologica italiana a Tyana/Kemerhisar (Turchia). I materiali provenienti dallo scavo delle Terme romane: dati preliminari. Rei Cretariae Romanae Fautorum Acta 40, 255-261. Bonn, Rei Cretariae Romanae Fautores. McClellan, M. C. 1999. Hellenistic and Roman pottery from ‘Ain Dara. In E. C. Stone and P. E. Zimansky, The Iron Age settlement at ‘Ain Dara, Syria. Survey and surroundings. British Archaeological Reports International Series 786, 19-22. Oxford, Archaeopress. McCormick, M. 2002. Origins of the European economy: communications and commerce, AD 300-900. Cambridge, Cambridge University Press.

169

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:44 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East Mee, C. and Forbes, H. (eds.) 1997. A rough and rocky place: landscape and settlement history of the Methana Peninsula, Greece. Monographs in Archaeology & Oriental Studies. Liverpool, Liverpool University Press. Meriç, R. 2002. Späthellenistisch-römische Keramik und Kleinfunde aus einem Schachtbrunnen am Staatsmarkt in Ephesos. Forschungen in Ephesos 9/3. Vienna, Verlag der Österreichischen Akademie der Wissenschaften. Metzger, I. R. 1969. Die Hellenistische Keramik in Eretria. Eretria Ausgrabungen und Forschungen 2. Bern, Francke Verlag. Metzger, I. R. 1993. Späthellenistischer und römische Epoche. In P. Ducrey, I. I. Metzger and K. Reber, Le Quartier de la Maison aux Mosaïques. Eretria Fouilles et recherches 8, 126-143. Lausanne, Payot. Meyer, M. (ed.) 2007. Neue Zeiten – Neue Sitten. Zu Rezeption und Integration römischen und italischen Kulturguts in Kleinasien. Wiener Forschungen zur Archäologie 12. Vienna, Phoibos Verlag. Meyer-Schlichtmann, C. 1988. Die pergamenischen Sigillata aus der Stadtgrabung von Pergamon. Mitte 2. JH v. Chr. – Mitte 2. JH n. Chr. Pergamenische Forschungen 6. Berlin, De Gruyter. Meyers, E. B. and Meyers, C.L. 2013: The pottery from ancient Sepphoris. Sepphoris Excavation Reports 1. Winona Lake, Eisenbrauns. Meyers, E. M., Strange, J. F. and Meyers, C. L. 1982. Second preliminary report on the 1981 excavations at en-Nabratein. Bulletin of the American School of Oriental Research 246, 35-54. Meyza, H. 1995. Cypriot Sigillata and Cypriot Red Slip Ware: problems of origin and continuity. In H. Meyza and J. Mlynarczyk (eds.), “Hellenistic and Roman pottery in the Eastern Mediterranean – Advances in scientific studies” Acts of the 2 Nieborów pottery workshop, Nieborów, 18-20 december 1993, 179-202. Warsaw, Polish Academy of Sciences. Meyza, H. 2005. Stamped Late Hellenistic and Early Roman fine wares found at Kom el-Dikka in Alexandria before 1987. Études et Travaux 20, 121-136. Meyza, H. 2007. Cypriot Red Slip Ware: studies on a Late Roman Levantine fine ware. Nea Paphos 5. Warsaw, Centre d’archéologie méditerranéenne de l’Académie polonaise des sciences. Meyza, H. and Papuci-Wladyka, E. 1999. Nea Paphos, Cyprus: pottery from cistern STR 1/96-97. In Centenary of Mediterranean Archaeology at the Jagiellonian University 1897-1997, 75-92. Krakow, Jagiellonian University. Mieliwodzki, P. 1995. Terra sigillata from two seasons at Tell Atrib. Études et Travaux 17, 122-131. Millett, P. 2001. Productive to some purpose? The problem of ancient economic growth. In D. J. Mattingly and J. Salmon (eds.), Economies beyond agriculture in the Classical world, 17-48. London, Routledge. Mitchell, S. 1999. Greek epigraphy and social change. A study of the Romanization of south-west Asia Minor in the third century A.D. In XI Congresso Internazionale

di Epigrafia Greca e Latina, Roma, 18-24 settembre 1997, Atti II, 419-433. Rome, Edizioni Quasar. Mitsopoulos-Leon, V. 1975. Gefässe der Ostsigillata A von einem gesunkenen Schiff aus dem Golf van Antalya. Germania 53, 101-212. Mitsopoulos-Leon, V. 1991. Die Basilika am Staatsmarkt in Ephesos. Kleinfunde 1. Teil: Keramik hellenistischer und römischer Zeit. Forschungen in Ephesos 9/2/2). Vienna, Schindler. Mlynarczyk, J. 2001. Pottery Report. In A. Segal, J. Mlynarczyk and M. Burdajemicz, Hippos [Sussita], second season of excavations July 2001, 8-10. Haifa, Zinman Institute of Archaeology. Mlynarczyk, J. 2002. Pottery report, Sussita 2002. In A. Segal, J. Mlynarczyk, M. Burdajemicz and M. Schuler, Hippos. Third season of excavations July 2002, 37-59. Haifa, Zinman Institute of Archaeology. Mlynarczyk, J. 2003. Pottery report. In A. Segal, J. Mlynarczyk, M. Burdajemicz, M. Schuler and M. Eisenberg, Hippos-Sussita. Fourth season of excavations June-July 2003, 50-88. Haifa, Zinman Institute of Archaeology. Mlynarczyk, J. 2004. Pottery report (Hippos-Sussita 2004). In A. Segal, J. Mlynarczyk, M. Burdajemicz, M. Schuler and M. Eisenberg, Hippos-Sussita. Fifth season of excavations (September - October 2004) and summary of all five seasons (2000-2004), 140-170 Haifa, Zinman Institute of Archaeology. Mlynarczyk, J. 2005a. Pottery report. In A. Segal, J. Mlynarczyk, M. Burdajemicz, M. Schuler and M. Eisenberg, Hippos-Sussita. Sixth season of excavations (July 2005), 113-139. Haifa, Zinman Institute of Archaeology. Mlynarczyk, J. 2005b. The “Pink Powdery Ware” at Yeronisos. A local west Cypriot ware of the late Hellenistic period. Études et Travaux 20, 137-149. Mlynarczyk, J. 2006. Pottery report. In A. Segal, J. Mlynarczyk, M. Burdajemicz, M. Schuler and M. Eisenberg, Hippos-Sussita. Seventh season of excavations (July 2006), 91-126. Haifa, Zinman Institute of Archaeology. Mlynarczyk, J. 2007. Pottery report. In A. Segal, J. Mlynarczyk, M. Burdajemicz, M. Schuler and M. Eisenberg, Hippos-Sussita. Seventh season of excavations (July 2007), 105-154. Haifa, Zinman Institute of Archaeology. Mlynarczyk, J. 2008. Pottery report. In A. Segal, J. Mlynarczyk, M. Burdajemicz, M. Schuler and M. Eisenberg, Hippos-Sussita. Seventh season of excavations (June-July 2008), 59-109. Haifa, Zinman Institute of Archaeology. Mlynarczyk, J. 2010. Sigillatae (ESA, CS) and “pseudosigillata” (PPW) at Geronisos: an overview of forms and comparison of repertoire. Report of the Department of Antiquities, Cyprus, 349-364. Mommsen, H. and Japp, S. 2009. Neutronenaktivierungsanalyse von 161 Keramikproben aus Pergamon und Fundorten der Region. Istanbuler Mitteilungen 59, 269-286.

170

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:44 UTC All use subject to https://about.jstor.org/terms

Bibliography Monacchi, D. 1988. Ceramiche sigillata. In A. DiVita (ed.), Gortina 1. Monografie della Scuola Archaeologica di Atene e delle Missioni italiane in Oriente 3, 199-220. Rome, L’Erma di Bretschneider. Montet, P. 1928. Objets trouvés hors du temple et des tombeaux. In P. Montet, Byblos et l’Égypte. Quatre campagnes de fouilles a Gebeil 1921 - 1922 - 1923 – 1924. Bibliothèque Archéologique et Historique 11. Paris, Paul Geuthner. Morel, J.-P. 1976. Céramiques d’Italie et céramiques hellénistiques (150-30 av. J.-C.). In P. Zanker (ed.), Hellenismus in Mittelitalien. Kolloquium in Göttingen vom 5. bis 9. Juni 1974, Erster Teil, 471-501. Göttingen, Vandenhoeck & Ruprecht. Morel, J.-P. 1981. Céramique campanienne: les formes. Rome, École française. Morel, J.-P. 1986. Céramiques à vernis noir d’Italie trouvées à Délos. Bulletin de correspondance hellénique 110, 461-493. Morrisson, C. and Sodini, J.-P. 2002. The sixth-century economy. In A. E. Laiou (ed.), The economic history of Byzantium from the seventh through the fifteenth century Volume 2, 171-220. Washington DC, Dumbarton Oaks Research Library. Müller, C. Forthcoming. The Roman fate of Thespiai (171 BC-4th c. AD). In J. L. Bintliff and A. M. Snodgrass (eds.), The urban survey of the city of Thespiae. Cambridge, Cambridge University Press. Munro, J. A. R. and Tubbs, H. A. 1891. Excavations in Cyprus, 1890. The Journal of Hellenic Studies 12, 59198. Murphy, E. A. and Poblome, J. 2011. Producing pottery vs. producing models: interpreting workshop organization at the Potters’ Quarter of Sagalassos. In M. L. Lawall and J. Lund (eds.), Pottery in the archaeological record: Greece and beyond. Acts of the International Colloquium held at the Danish and Canadian Institutes in Athens, June 20-22, 2008. Gösta Enbom Monographs 1, 30-36. Aarhus, Aarhus University Press. Murphy, E. A. and Poblome, J. 2012. Technical and social considerations of tools from Roman-period ceramic workshops at Sagalassos (southwest Turkey): not just tools of the trade? Journal of Field Archaeology 25.2, 197-217. Negev, A. 1972. Nabatean sigillata. Révue Biblique 79, 381-398. Negev, A. 1974. The Nabatean potter’s workshop at Oboda. Rei Cretariae Romanae Fautorum Acta Supplementa 1. Bonn, Rei Cretariae Romanae Fautores. Negev, A. 1986. The late Hellenistic and early Roman pottery of Nabatean Oboda. Qedem 22. Jerusalem, The Institute of Archaeology, the Hebrew University. Negev, A. and Sivan, R. 1977. The pottery of the Nabataean necropolis at Mampsis. Rei Cretariae Romanae Fautorum Acta 17-18, 109-131. S.l., Rei Cretariae Romanae Fautores. Neuru, L. and Soren, D. 1987. Significant pottery and lamps from the temple. In D. Soren (ed.), The sanctuary of Apollo Hylates at Kourion, Cyprus, 232-262. Tucson, University of Arizona Press.

Newson, P. 2014. Pottery of the ‘Land of Carchemish’ Project and the northern Euphrates. In Fischer-Genz et al. 2014, 3-20. Nicolaou, K. 1976. The historical topography of Kition. Studies in Mediterranean Archaeology Volume 43. Göteborg, Paul Åströms Förlag. Nixon, L., Moody, J. and Rackham, O. 1988. Archaeological survey in Sphakia, Crete. Echos du Monde Classique/ Classical Views 32, 259-273. Nixon, L., Moody, J., Niniou-Kindeli, V., Price, S. and Rackham, O. 1990. Archaeological survey at Sphakia, Crete. In Echos du Monde Classique/Classical Views 34, 213-220. Nixon, L., Moody, J., Price, S. and Rackham, O. 1989. Archaeological survey in Sphakia, Crete. Echos du Monde Classique/Classical Views 33, 210-215. Ober, J. 1987. Pottery and miscellaneous artifacts from fortified sites in northern and western Attica. Hesperia 56, 197-227. Oikonomou-Laniado, A. 2005. Argos paléochrétienne. Contribution à l’étude du Péloponnèse Byzantin. British Archaeologial Reports International Series 1173. Oxford, Archaeopress Olcese, G. 2004. Italian terra sigillata in Rome and the Rome area; production, distribution and laboratory analysis. In Poblome et al. 2004, 279-298. Orssaud, D. 1980. La céramique. In J.-P. Sodini, G. Tate, B. Bavant, S. Bavant, J.-L. Biscop and D. Orssaud, Déhès (Syrie du Nord) Campagnes I-III (1976-1978). Recherches sur l’habitat rural. Syria 57, 234-266. Orssaud, D. 1991. Annexe. La céramique. In J. Lauffray, Halabiyya-Zenobia. Place forte du limes oriental et la haute-Mésopotamie au VIe siècle. Tome II. L’architecture publique, religieuse, privée et funéraire. Institut Français d’Archéologie du Proche-Orient 138, 260-275. Paris, Paul Geuthner. Ortisi, S. (mit einem Beitrag von M. Mackensen). 1999. Ein frühkaiserzeitlicher Keramikkomplex aus einem Schöpfbrunnen in Insula E 117 West in Karthago. Mitteilungen des Deutschen Archäologischen Instituts, Römische Abteilung 106, 439-493. Outschar, U. 1997. Die keramischen Funde aus dem und bene dem Sarkophag westlich des “AndroklosHeroons”. In H. Thür (ed.), “…und verschönerte die Stadt…” Ein ephesischer Priester des Kaiserkultus in seinem Umfeld. Österreichisches Archäologisches Institut Sonderschriften 27, 27-40. Vienna, Verlag der Österreichischen Akademie der Wissenschaften. Outschar, U. 2000. Keramik macht Baugeschichte. Die Begrenzung der Insula Hanghaus 2 in Ephesos: Die Evidenz des keramischen Fundmaterials unter dem Pflaster der sog. Hanghausstrasse. Römische Historische Mitteilungen 42, 107-169. Oxé, A. 1927. Terra sigillata aus dem Kerameikos. Mitteilungen des Deutschen Archäologischen Instituts, Athenische Abteilung 52, 213-224. Oxé, A., Comfort, H. C. and Kenrick, P. M. 2000. Corpus Vasorum Arretinorum. Antiquitas 41. Bonn, Rudolf Habelt.

171

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:44 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East Panero, E. 2008. Scambi commerciali tra Occidente e Oriente: la circolazione della Sigillata Gallica a Creta. Rei Cretariae Romanae Fautorum Acta 40, 243-247. Abingdon, Rei Cretariae Romanae Fautores. Papacostas, T. 2001. The economy of Late Antique Cyprus. In S. A. Kingsley and M. J. Decker (eds.), Economy and exchange in the East Mediterranean during Late Antiquity. Proceedings of a conference at Somerville College, Oxford, 29th May, 1999, 107-128. Oxford, Oxbow Books. Papaioannou, M. 2010. East meets west: the pottery evidence from Abdera. In http://151.12.58.75/archeologia/bao_ document/articoli/5_PAPAIOANNOU.pdf (Accessed 20 February 2013). Papanikola-Bakirtzis, D. and Kousoulakou, T. (eds.) 2010. KEPAMIKH YΣΤΕΡΗΣ APXAIOTHTAΣ AΠO EΛΛAΔIKO XΩPO (3oς-7oς αι. μ.X.). Eπιστημονική Συνάντηση. Θεσσαλoνίκη, 12-16 Noνεμβρίου 2006. Publications of the Archaeological Institute of Macedonian and Thracian Studies 8. Thessaloniki, Aristotle University of Thessaloniki. Parker, A. J. 1992. Ancient shipwrecks of the Mediterranean and the Roman provinces. British Archaeological Reports International Series 580. Oxford, Archaeopress. Parker, S. T. 1994. A late Roman soldier’s grave by the Dead Sea. Annual of the Department of Antiquities of Jordan 38, 385-394. Parmegiani, N. 1987. The eastern sigillata in Tell Barri/ Kahat. Mesopotamia 22, 113-128. Peacock, D. P. S. 1982a. Pottery in the Roman world: an ethnoarchaeological approach. Harlow, Longman. Peacock, D. P. S. 1982b. The giant fine-ware producers. In Peacock 1982a, 114-128. Peacock, D. P. S. 2007. Pottery from the survey. In D. P. S. Peacock and L. Blue (eds.), The ancient Red Sea port of Adulis, Eritrea. Results of the Eritro-British Expedition 2004-5, 79-108. Oxford, Oxbow Books. Peacock, D. P. S. and Williams, D. F. 1986. Amphorae and the Roman economy. London, Longman. Peacock, D. P. S., Bejaoui, F. and Ben Lazreg, N. 1990. Roman pottery production in central Tunisia. Journal of Roman Archaeology 3, 59-84. Peeters, D., Bes, P. M., Poblome, J., Hayes, J. W. and Bintliff, J. L. In press. Making use of time and space – using African Red Slip Ware as an indicator of economic activity in Mid- and Late Roman Thespiae and Tanagra. In J. Hilditch, A. Kotsonas, C. BeestmanKruijshaar, M. Revello-Lami, S. Ruckl and S. Ximeri (eds.), Island, mainland, coastland and hinterland: ceramic perspectives on connectivity in the ancient Mediterranean. Proceedings of the Conference held at the University of Amsterdam, 1-3 February 2013, Amsterdam. Peignard, A. 1997. La vaisselle de la Maison des Sceaux, Délos. In Δ’ΕΠΙΣΤΗΜΟΝΙΚΗ ΣΥΝΑΝΤΗΣΗ ΓΙΑ ΤΗΝ ΕΛΛΗΝΙΣΤΙΚΗ ΚΕΡΑΜΙΚΗ. ΧΡΟΝΟΛΟΓΙΚΑ ΠΡΟΒΛΗΜΑΤΑ ΚΛΕΙΣΤΑ ΣΥΝOΛΑ ΕΡΓΑΣΤΗΡΙΑ. ΠΡΑΚΤΙΚΑ. Κείμενα, 308-316. Athens.

Peignard-Giros, A. 2000. Habitudes alimentaires grecques et romaines à Délos à l’époque hellénistique: le témoignage de la céramique. In J.-M. Luce (ed.), Paysage et alimentation dans le monde grec. Pallas 52, 209-220. Toulouse, Presses Universitaires du Mirail. Peignard-Giros, A. 2003. Présentation des céramiques issues des fouilles menées en 1998-1999 sur l’acropole lycienne de Xanthos. Anatolia Antiqua 11, 215-221. Peña, J. T. 2007a. Roman pottery in the archaeological record. New York, Cambridge University Press. Peña, J. T. (with an appendix by J. McCaw) 2007b. The quantitative analysis of Roman pottery. General problems, the methods employed at the Palatine East, and the supply of African Sigillata to Rome. In E. Papi (ed.), Supplying Rome and the empire. The proceedings of an international seminar held at SienaCertosa di Pontignano on May 2-4, 2004 on Rome, the provinces, production and distribution. Journal of Roman Archaeology Supplementary Series 69, 153171. Portsmouth, Journal of Roman Archaeology. Perdrizet P. 1908. Monuments figures: petits bronzes, terres-cuites, antiquités diverses. Fouilles de Delphes 5.1. Paris, Albert Fontemoing. Pétridis, P. 2007. Relations between pottery workshops in the Greek mainland during the Early Byzantine period. In B. Böhlendorf-Arslan, A. O. Uysal and J. WitteOrr (eds.), Çanak. Late Antique and Medieval pottery and tiles in Mediterranean archaeological contexts. Proceedings of the First International Symposium on Late Antique, Byzantine, Seljuk and Ottoman pottery and tiles in archaeological context (Çanakkale, 1-3 June 2005). BYZAS 7, 43-54. Istanbul, Ege Yayınları. Pétridis, P. 2010. La céramique protobyzantine de Delphes: une production et son contexte. Fouilles de Delphes 5. Monuments Figurés 4. Paris, École française d’Athènes. Pettegrew, D. K. 2007. The busy countryside of Late Roman Corinth. Interpreting ceramic data produced by regional archaeological survey. Hesperia 76, 743-784. Pharr, C. 1969. The Theodosian Code and Novels and the Sirmondian Constitutions. Corpus Iuris Romani Volume 1. New York, Princeton University Press. Pickersgill, C. and Roberts, P. 2003. New light on Roman Sparta: Roman pottery from the Sparta theatre and stoa. The Annual of the British School at Athens 98, 549-597. Pieri, D. 2005. Le commerce du vin oriental à l’époque byzantine (Ve-VIIe siècle). Le temoignage des amphores en Gaule. Bibliothèque Archéologique et Historique 174. Beirut, Institut Français d’Archéologique du Proche-Orient. Ploug, G. 1985. Minor objects of terracotta. In G. Ploug, Hama. Fouilles et recherches 1931-1938 3 1. The Graeco-Roman town. Nationalsmuseets Skrifter 9, 228-230. Copenhagen, Nationalmuseet. Poblome, J. 1999. Sagalassos Red Slip Ware. Typology and chronology. Studies in Eastern Mediterranean Archaeology 2. Turnhout, Brepols.

172

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:44 UTC All use subject to https://about.jstor.org/terms

Bibliography Poblome, J. 2004a. Comparing ordinary crafts. Textile and pottery production in Roman Asia Minor. Journal of the Economic and Social History of the Orient 47, 491506. Poblome, J. 2004b. Italian Sigillata in the Eastern Mediterranean. In Poblome et al. 2004, 17-30. Poblome, J. 2006a. Mixed feelings on Greece and Asia Minor in the third century AD. In Malfitana et al. 2006, 189-212. Poblome, J. 2006b. Made in Sagalassos. Modelling regional potential and constraints. In S. Menchelli and M. Pasquinucci (eds.), Territorio e produzioni ceramiche. Paesaggi, economia e società in età romana. Atti del Convegno Internazionale Pisa 20-22 ottobre 2005. Instrumenta 2, 355-363. Pisa, Pisa University Press. Poblome, J. Forthcoming. The potters of ancient Sagalassos revisited. In M. Flohr and A. I. Wilson (eds.), Beyond marginality. Craftsmen, traders and the socioeconomic history of urban communities in the Roman world. Oxford, Oxford University Press. Poblome, J. and Brulet, R. 2005. Production mechanisms of sigillata manufactories. When east meets west. In M. Berg Briese and L. E. Vaag (eds.), Trade relations in the Eastern Mediterranean from the Late Hellenistic period to Late Antiquity: the ceramic evidence. Acts from a Ph.D.-seminar for young scholars, Sandbjerg Manorhouse, 12-15 February 1998. Halicarnassian Studies 3, 27-36. Odense, University Press of Southern Denmark. Poblome, J. and Fırat, N. 2011. Late Roman D. A matter of open(ing) or closed horizons? In M. A. Cau Ontiveros, P. Reynolds and M. Bonifay (eds.), Late Roman fine wares. Solving problems of typology and chronology. A review of the evidence, debate and new contexts. Late Roman Fine Wares 1. Roman and Late Antique Mediterranean Pottery 1, 49-55. Oxford, Archaeopress. Poblome, J. and Waelkens, M. 2003. Sagalassos and Alexandria. Exchange in the Eastern Mediterranean. In C. Abadie-Reynal (ed.), Les céramiques an Anatolie aux époques hellénistique et romaine. Actes de la Table Ronde d’Istanbul, 22-24 mai 1996. Varia Anatolica 15, 179-191. Paris, De Boccard. Poblome, J. and Zelle, M. 2002. The table ware boom. A socio-economic perspective from western Asia Minor. In Berns et al. 2002, 275-287. Poblome, J., Bes, P. M. and Lauwers, V. 2007. Winning hearts, minds and stomachs? Artefactual or artificial evidence for Romanization. In Meyer (ed.) 2007, 221232. Poblome, J., Bounegru, O., Degryse, P., Viaene, W., Waelkens, M. and Erdemgil, S., 2001b. The sigillata manufactories of Pergamon and Sagalassos. Journal of Roman Archaeology 14, 143-165. Poblome, J., Degryse, P., Cottica, D. and N. Fırat, 2001a. A new Early Byzantine production centre in western Asia Minor. A petrographical and geochemical study of Red Slip Ware from Hierapolis, Perge and Sagalassos. Rei Cretariae Romanae Fautorum Acta 37, 119-126. Abingdon, Rei Cretariae Romanae Fautores.

Poblome, J., Malfitana, D. and Lund, J. 2006. A concluding dilemma: Sisyphos versus Daidalos. In Malfitana et al. 2006, 557-579. Poblome, J., Malfitana, D. and Lund, J. 2013. Pottery, Roman Empire. In R. S. Bagnall, K. Brodersen, C. B. Champion, A. Erskine and S. R. Huebner (eds.), The Encyclopedia of Ancient History, 5482-5489. Malden, Wiley-Blackwell. Poblome, J., Talloen, P., Brulet, R. and Waelkens, M. (eds.) 2004. Early Italian Sigillata. The chronological framework and trade patterns. Proceedings of the first international ROCT-congress. Leuven, May 7 and 8, 1999. Bulletin Antieke Beschaving Supplement 10. Leuven, Peeters. Pritchard, J. B. 1952-1954. The excavation at Herodian Jericho, 1951. The Annual of the American Schools of Oriental Research 32-33, 1-58. Quercia, A., Johnston, A., Bevan, A., Conolly, J. and Tsaravopoulos, A. 2011. Roman pottery from an intensive survey of Antikythera, Greece. The Annual of the British School at Athens 106, 47-98. Raab, H. A. 2001. Rural settlement in Hellenistic and Roman Crete. The Akrotiri Peninsula. British Archaeological Reports International Series 984. Oxford, Archaeopress. Rauh, N. K. 1993. The sacred bonds of commerce. Religion, economy and trade society at Hellenistic Roman Delos, 166-87 B.C. Amsterdam, J. C. Gieben. Rautman, M. L. 1995a. Two Late Roman wells at Sardis. The Annual of the American Schools of Oriental Research 53, 37-84. Rautman, M. L. 1995b. Neutron activity analysis of Cypriot and related ceramics at the University of Missouri. In H. Meyza and J. Mlynarczyk (eds.), “Hellenistic and Roman pottery in the Eastern Mediterranean – Advances in scientific studies” Acts of the 2 Nieborów pottery workshop, Nieborów, 18-20 december 1993, 331-349. Warsaw, Polish Academy of Sciences. Rautman, M. L. 2003. A Cypriot village of Late Antiquity. Kalavasos-Kopetra in the Vasilikos valley. Journal of Roman Archaeology Supplementary Series 52. Ann Arbor, Journal of Roman Archaeology Rautman, M. L., Gomez, B., Neff, H. and Glascock, M. D. 1993. Neutron Activation Analysis of Late Roman ceramics from Kalavasos-Kopetra and the environs of the Vasilikos Valley. Report of the Department of Antiquities, Cyprus, 233-264. Rautman, M. L., Neff, H. and Glascock, M. D. 2003. Appendix 3. Compositional study of ceramics from Kopetra. In Rautman 2003, 267-271. Regev, D. 2007. A new approach on the origin of Eastern Sigillata A. Topoi Supplément 8, 189-200. Rendini, P. 2004. Ceramiche fini e comuni, ceramica sigillata. In A. DiVita (ed.), Gortina 6. Scavi 19791982. Monografie della Scuola Archeologica italiana di Atene e delle Missioni italiane in Oriente 14, 341, 401-406. Padua, Aldo Ausilio. Reynolds, P. 1995. Trade in the western Mediterranean, AD 400-700: the ceramic evidence. British

173

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:44 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East Archaeological Reports International Series 604. Oxford, Archaeopress. Reynolds, P. 1997. Appendix 1b. Ceramic studies. A first century AD pottery assemblage from Lepcis Magna. In H. Walda, S.-A. Ashton, P. Reynolds, J. Sidell, I. Welsby Söstrom and K. Wilkinson, The 1996 excavations at Lepcis Magna. Libyan Studies 28, 49-63. Reynolds, P. 1997-1998. Pottery production and economic exchange in second century Berytus: some preliminary observations of ceramic trends from quantified ceramic deposits from the AUB-Leverhulme excavations in Beirut. Berytus 43, 35-110. Reynolds, P. 2002. The pottery. In W. Bowden, R. Hodges and K. Lako, Roman and late-antique Butrint: excavations and survey 2000-2001. Journal of Roman Archaeology 15, 221-227. Reynolds, P. 2004a. Italian fine wares in first century AD Berytus: the assemblage from the Cistern Deposit BEY 006 12300/12237. In Poblome et al. 2004, 117-131. Reynolds, P. 2004b. The Roman pottery from the Triconch Palace. In R. Hodges, W. Bowden and K. Lako, Byzantine Butrint: excavations and Surveys 1994-99, 224-269. Oxford, Oxbow Books. Reynolds, P. 2005a. Hispania in the Late Roman Mediterranean: ceramics and trade. In K. Bowes and M. Kulikowski (eds.), Hispania in Late Antiquity. Current Perspectives. The Medieval and Early Modern Iberian World 24, 369-486. Leiden, Brill. Reynolds, P. 2005b. Levantine amphorae from Cilicia to Gaza: a typology and analysis of regional production trends from the 1st to 6th centuries. In J. M. Gurt i Esparraguera, J. Buxeda i Garrigós and M. A. Cau Ontiveros (eds.), LRCW I. Late Roman Coarse Wares, Cooking Ware, and Amphorae in the Mediterranean: Archaeology and Archaeometry. British Archaeological Reports International Series 1340, 563-612. Oxford, Archaeopress. Reynolds, P. 2010. Trade networks of the east, 3rd to 7th centuries: the view from Beirut (Lebanon) and Butrint (Albania) (fine wares, amphorae and kitchen wares). In S. Menchelli, S. Santoro, M. Pasquinucci and G. Giuducci (eds.), LRCW 3. Late Roman Coarse Wares, Cooking Wares and Amphorae in the Mediterranean. Archaeology and Archaeometry. Comparison between western and eastern Mediterranean. British Archaeological Reports International Series 2185, 1, 89-114. Oxford, Archaeopress. Reynolds, P. 2011. Fine wares from Beirut contexts, c. 450 to c. 600. In M. A. Cau Ontiveros, P. Reynolds and M. Bonifay (eds.), Late Roman fine wares. Solving problems of typology and chronology. A review of the evidence, debate and new contexts. Late Roman Fine Wares 1. Roman and Late Antique Mediterranean Pottery 1, 207-230. Oxford, Archaeopress. Reynolds, P., Hernandez, D. R. and Çondi, D. 2008. Excavations in the Roman Forum of Buthrotum (Butrint): first to third century pottery assemblages and trade. Rei Cretariae Romanae Fautorum Acta 40, 7188. Bonn, Rei Cretariae Romanae Fautores.

Rhodes, P. J. and Osborne, R. (eds.) 2003. Greek historical inscriptions 404-323 BC. Oxford, Oxford University Press. Ricci, M. 2007. Elaiussa Sebaste: context, production and commerce. In B. Böhlendorf-Arslan, A. O. Uysal and J. Witte-Orr (eds.), Çanak. Late Antique and Medieval pottery and tiles in Mediterranean archaeological contexts. Proceedings of the First International Symposium on Late Antique, Byzantine, Seljuk and Ottoman pottery and tiles in archaeological context (Çanakkale, 1-3 June 2005). BYZAS 7, 169-180. Istanbul, Ege Yayınları. Rickman, G. E. 1980. The corn supply of ancient Rome. Oxford, Oxford University Press. Rife, J. L. 2012. The Roman and Byzantine graves and human remains. Isthmia 9. Princeton, Princeton University Press. Rife, J. L., Moore Morison, M., Barbet, A., Dunn, R. K., Ubelaker, D. H. and Monier, F. 2007. Life and death at a port in Roman Greece. The Kenchreai Cemetery Project, 2002-2006. Hesperia 76, 143-181. Riley, J. A. 1975. The pottery from the first session of excavation in the Caesarea Hippodrome. Bulletin of the American Schools of Oriental Research 218, 25-63. Riley, J. A. 1979. The coarse pottery from Benghazi. In J. A. Lloyd (ed.), Excavations at Sidi Khrebish, Benghazi (Berenice), Volume 2, 91-497. Tripoli, Society for Libyan Studies. Riley, J. A. 1981. The pottery from the cisterns 1977.1, 1977.2, 1977.3. In J. H. Humphrey (ed.), Excavations at Carthage 1977, conducted by the University of Michigan Volume 6, 85-124. Ann Arbor, University of Michigan Press. Rizzo, M. A. 2001. Terra Sigillata italica, Terra Sigillata cipriota (“Cypriot Sigillata”), Terra Sigillata macedonica, Terra Sigillata africana, Terra Sigillata LRC, Terra Sigillata color crema, Ceramica cnidia a rilievo. In A. DiVita (ed.), Gortina V.3. Lo scavo del pretorio (1989-1995). Volume V.3 t. I. I materiali. Monografie della Scuola Archeologica Italiana di Atene e delle Missioni Italiane in Oriente 12, 36-67, 72-73. Padua, Aldo Ausilio. Rizzo, M. A. 2004. Ceramica sigillata. In A. DiVita (ed.), Gortina 6. Scavi 1979-1982. Monografie della Scuola Archeologica italiana di Atene e delle Missioni italiane in Oriente 14, 184-198, 326-336. Padua, Aldo Ausilio. Robinson, H. S. 1959. Pottery of the Roman period. Chronology. The Athenian Agora 5. Princeton, The American School of Classical Studies at Athens. Robinson, H. S. 1969. A sanctuary and cemetery in western Corinth. Hesperia 38, 1-35. Rodziewicz, M. 1976. Alexandrie I. La céramique romaine tardive d’Alexandrie. Warsaw, Éditions scientifiques de Pologne. Rodziewicz, M. 1994. Chronology and topography. In K. Grzymski, J. Anderson, J. W. Hayes, M. Abd elSamie, A. el-Tabai and O. Hamza, Canadian-Egyptian excavations at Tell al-Farama (Pelusium) west: spring 1993. In Sociétés urbaines en Égypte et au Soudan. Cahiers de Recherches de l’Institut de Papyrologie et 174

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:44 UTC All use subject to https://about.jstor.org/terms

Bibliography d’Egyptologie de Lille no. 16, 154-160. Lille, Institut de Papyrologie et d’Egyptologie de Lille. Römer, C. 1995. Die rot engobierte Keramik im Unteren Habur-Gebiet. In H. Meyza and J. Mlynarczyk (eds.), “Hellenistic and Roman pottery in the Eastern Mediterranean – Advances in scientific studies” Acts of the 2 Nieborów pottery workshop, Nieborów, 18-20 december 1993, 351-363. Warsaw, Polish Academy of Sciences. Römer-Strehl, C. 2000. Die Keramik. In M. Novák, A. Oettel and C. Witzel, Der parthisch-römische Friedhof von Tall Šēh Hamad/Magdala Teil I. Berichte der Ausgrabung Tall Šēh Hamad/Dūr-Katlimmu 5, 121147. Berlin, Harrassowitz Verlag. Rosenthal, R. 1978. The Roman and Byzantine pottery. In E. Stern, Excavations at Tel Mevorakh (1973-1976). Part 1: From the Iron Age to the Roman period. Qedem 9, 14-19. Jerusalem, Hebrew University of Jerusalem Institute of Archaeology. Rosenthal-Heginbottom, R. 1988. The pottery. In Y. Tsafrir, J. Patrich, R. Rosenthal-Heginbottom, I. Hershkovitz and Y. D. Nevo, Excavations at Rehovot-in-the-Negev Volume I: The northern Church. Qedem 25, 78-96. Jerusalem, Hebrew University of Jerusalem - Institute of Archaeology. Rosenthal-Heginbottom, R. 1995. Imported Hellenistic and Roman pottery. In E. Stern, Excavations at Dor, final report volume 1B. Areas A and C: the finds. Qedem Reports 2, 183-233, 257-271. Jerusalem, Hebrew University of Jerusalem - Institute of Archaeology. Rosenthal-Heginbottom, R. 2005. The 1968 excavations. In B. Arubas and H. Goldfus (eds.), Excavations on the site of the Jerusalem International Convention Center (Binyanei Ha’uma): a settlement of the late first to second Temple period, the tenth legion’s kiln works, and a Byzantine monastic complex. The pottery and other small finds. Journal of Roman Archaeology Supplementary Series 60, 229-282. Portsmouth, Journal of Roman Archaeology. Rossiter J. J. and Freed, J. 1991. Canadian-Turkish Excavations at Domuztepe, Cilicia (1989). Echos du Monde Classique/Classical Views 35, 145-174. Roth, R. E. 2007. Styling Romanisation. Pottery and society in central Italy. Cambridge, Cambridge University Press. Rotroff, S. I. 1997a. Hellenistic pottery: Athenian and imported wheelmade table ware and related material. Part 1: Text. The Athenian Agora Volume 29. Princeton, Princeton University Press. Rotroff, S. I. 1997b. From Greek to Roman in Athenian ceramics. In M. C. Hoff and S. I. Rotroff (eds.), The Romanization of Athens. Proceedings of an international conference held at Lincoln, Nebraska (April 1996). Oxbow Monographs 94, 97-116. Oxford, Oxbow Books Rotroff, S. I. and Oliver jr., A. 2003. The Hellenistic pottery from Sardis: the finds through 1994. Archaeological Exploration of Sardis Monograph 12. Cambridge MA, Harvard University Press.

Rowe, A. H. 1999. A Late Roman streetscape in Nea Pafos. Report of the Department of Antiquities, Cyprus, 269277. Rudolph, W. W. 1979. Excavations at Porto Cheli and vicinity preliminary report V: the early Byzantine remains. Hesperia 48, 294-324. Rutten, K. 2007. The Roman fine wares of ed-Dur (Umm al-Qaiwain U.A.E.) and their distribution in the Persian Gulf and the Indian Ocean. Arabian Archaeology and Epigraphy 18, 8-24. Sackett, L. H. 1992. Knossos from Greek City to Roman colony. Excavations at the Unexplored Mansion 2. British School of Archaeology at Athens Supplement 21. Oxford, Thames & Hudson. Safrai, Z. 1994. The economy of Roman Palestine. London, Routledge. Sakka, N. 2001. Classical, Hellenistic, and Roman pottery. In M. B. Cosmopoulos, The rural history of ancient Greek city-states. The Oropos Survey Project. British Archaeological Report International Series 1001, 6470. Oxford, Archaeopress. Salles, J.-F. 1993. La citerne 198. In J.-F. Salles (ed.), Les niveaux hellénistiques. Kition-Bamboula 4, 261-284. Paris, Éditions Recherche sur les Civilisations. Salles, J.-F. 1995. Céramiques hellénistiques de KitionBamboula. In H. Meyza and J. Mlynarczyk (eds.), “Hellenistic and Roman pottery in the Eastern Mediterranean – Advances in scientific studies” Acts of the 2 Nieborów pottery workshop, Nieborów, 18-20 december 1993, 397-414. Warsaw, Polish Academy of Sciences. Sanders, G. D. R. 1999. A Late Roman bath at Corinth. Excavations in the Panayia field, 1995-1996. Hesperia 68, 441-80. Sanders, I. F. 1976. Settlement in the Hellenistic and Roman periods on the Plain of the Mesara, Crete. The Annual of the British School at Athens 71, 131-137. Sazanov, S. 2000. Les ensembles clos de Chersonese de la fin du VIe-troisième quart du VIIe siècles: la chronologie de la céramique. In M. Kazanski and V. Soupault (eds.), Les sites archéologiques en Crimée et au Caucase durant l’Antiquité tardive et le haut Moyen Age. Colloquia Pontica 5, 123-149. Leiden, Brill. Schäfer, J. 1962. Terra Sigillata aus Pergamon. Archäologischer Anzeiger, 777-802. Schauer, C. 2001. Ein späthellenistischer Bau in Olympia (Grabungen des DAI 1987-1991). In J.-Y. Marc and J.-C. Moretti (eds.), Constructions publiques et programmes edilitaires en Grèce entre le IIe siècle av. J.-C. et le Ier siècle ap. J.-C. Actes du Colloque Organisé par l’Ecole française d’Athènes et le CNRS, Athènes 14-17 mai 1995. Bulletin de correspondance hellénique Supplément 39, 323-328. Athens, École française d’Athènes. Schauer, C. 2003. Kaiserzeitliche Keramik der westpeloponnes (Grabungen des deutschen archäologischen Instituts in Olympia. Rei Cretariae Romanae Fautorum Acta 38, 271-278. Abingdon, Rei Cretariae Romanae Fautores.

175

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:44 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East Schauer, C. 2008. Das neugefundenen hellenistische Bad in Olympia. Datierende Keramik aus Bauzeit, Umbauund Zerstörungsschicht. Rei Cretariae Romanae Fautorum Acta 40, 227-235. Abingdon, Rei Cretariae Romanae Fautores. Schmid, S. G. 1999. Decline or prosperity at Roman Eretria? Industry, purple dye works, public buildings, and gravestones. Journal of Roman Archaeology 12, 273-293. Schmid, S. G. 2000. Sullan debris from Eretria (Greece)? Rei Cretariae Romanae Fautorum Acta 36, 169-180. Abingdon, Rei Cretariae Romanae Fautores. Schneider, C. 1996. Die Importkeramik. In A. Bignasca, N. Desse-Berset, R. Fellmann Brogli, R. Glutz, S. Karg, D. Keller, B. Kolb, C. Kramar, M. Peter, S. G. Schmid, C. Schneider, R. A. Stucky, J. Studer and I. Zanoni, Petra Ez-Zantur I. Ergebnisse der SchweizerischLiechtensteinischen Ausgrabungen 1988-1992. Terra Archaeologica 2, 129-150. Mainz, Philipp von Zabern. Schneider, G. 1995. Roman red and black slipped pottery from NE-Syria and Jordan. First results of chemical analysis. In H. Meyza and J. Mlynarczyk (eds.), “Hellenistic and Roman pottery in the Eastern Mediterranean – Advances in scientific studies” Acts of the 2 Nieborów pottery workshop, Nieborów, 18-20 december 1993, 415-422. Warsaw, Polish Academy of Sciences. Schneider, G. 1996a. Chemische und mineralogische Untersuchungen von Keramik der hellenistischen bis frühislamischen Zeit in Nordost-Syrien. In K. Bartl and S. R. Hauser (eds.), Continuity and change in Northern Mesopotamia from the Hellenistic to the Early Islamic period. Berliner Beiträge zum Vorderen Orient 17, 127-136. Berlin, Dietrich Reimer Verlag. Schneider, G. 1996b. Chemical grouping of Roman Terra Sigillata finds from Turkey, Jordan and Syria. In I. Demirci, A. M. Özer and G. D. Summers (eds.), Archaeometry 94 – Proceedings of the 29th International Symposium on Archaeometry, Ankara 9-14 May 1994, 189-196. Ankara, Tübitak. Schneider, G. 2000. Chemical and mineralogical studies of Late Hellenistic to Byzantine pottery production in the eastern Mediterranean. Rei Cretariae Romanae Fautorum Acta 36, 525-536. Abingdon, Rei Cretariae Romanae Fautores. Schneider, G. and Daszkiewicz, M. 2006. Chemical analysis of Italian Sigillata from Italy and from the Northern Provinces. In Malfitana et al. 2006, 537-543. Schneider, G. and Japp, S. 2009. Röntgenfluoreszenzanalysen von 115 Keramikproben aus Pergamon, Çandarlı, Elaia und Atarneus (Türkei). Istanbuler Mitteilungen 59, 287-306. Sedov, A. V. 1992. New archaeological and epigraphical material from Qana (South Arabia). Arabian Archaeology and Epigraphy 3, 110-137. Sedov, A. V. 2007. The port of Qana’ and the incense trade. In D. P. S. Peacock and D. F. Williams (eds.), Food for the gods: new light on the ancient incense trade, 71111. Oxford, Oxbow Books.

Sève, M. 1980. Un puit argien du Haut-Empire. In Études Argiennes. Bulletin de correspondance hellénique Supplément 6, 295-321. Sevin, V. and Derin, Z. 1989. A fortified site to the east of Malatya. In D. H. French and C. S. Lightfoot (eds.), The eastern frontier of the Roman Empire. Proceedings of a colloquium held at Ankara in September 1988. British Institute of Archaeology at Ankara Monograph 11. British Archaeological Reports International Series 553ii, 437-460. Oxford, Archaeopress. Shajan, K. P., Cherian, P. J., Tomber, R. S. and Selvakumar, V. 2008. The external connections of early historic Pattanam, India: the ceramic evidence. Antiquity 315 (DOI: http://antiquity.ac.uk/projgall/tomber/index. html) (Accessed 20 January 2015). Sherwin-White, A. N. 1976. Rome, Pamphylia and Cilicia, 133-70 B.C. The Journal of Roman Studies 66, 1-14. Sieler, M. 2004. Späthellenistische, römische und spätantike Feinkeramik aus Petra – Surveymaterial der Naturhistorischen Gesellschaft Nürnberg. Damaszener Mitteilungen 14, 91-166. Sieler, M. 2008. Egyptian Red Slip Ware A and its production at the site of the Late Roman fort at Nag elHagar/Upper Egypt. Rei Cretariae Romanae Fautorum Acta 40, 271-278. Abingdon, Rei Cretariae Romanae Fautores. Silberstein, N. 2000. Hellenistic and Roman pottery. In Y. Hirschfeld, Ramat Hanadiv excavations. Final report of the 1984-1998 seasons, 420-469. Jerusalem, Israel Exploration Society. Sinn, U., Ladstätter, G. and Martin, A. 1993. Bericht über das Forschungsprojekt, Olympia während der römischen Kaiserzeit und in der Spätantike’ II. Die Arbeiten im Jahr 1993. Nikephoros 6, 153-158. Slane, K. W. 1986. Two deposits from the early Roman cellar building, Corinth. Hesperia 55, 271-318. Slane, K. W. 1987. Italian Sigillata imported to Corinth. Rei Cretariae Romanae Fautorum Acta 25/26, 189205. München, Rei Cretariae Romanae Fautores. Slane, K. W. 1989. Corinthian ceramic imports: the changing pattern of provincial trade in the first and second centuries A.D. In S. Walker and A. Cameron (eds.), The Greek Renaissance in the Greek Empire. Papers from the tenth British Museum Classical Colloquium. Bulletin of the Institute of Classical Studies Supplement 55, 219-225. London, Institute of Classical Studies. Slane, K. W. 1990. The Sanctuary of Demeter and Kore. The Roman pottery and lamps. Corinth 18.2. Princeton, The American School of Classical Studies at Athens. Slane, K. W. 1991a. Observations on Mediterranean amphoras and tablewares found in India. In Begley and De Puma (eds.) 1991a, 204-215. Slane, K. W. 1991b. Book review of Meyer-Schlichtmann 1988. Gnomon 63, 150-154. Slane, K. W. 1994. Tetrarchic recovery in Corinth. Pottery, lamps and other finds from the Peribolos of Apollo. Hesperia 63, 127-169.

176

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:44 UTC All use subject to https://about.jstor.org/terms

Bibliography Slane, K. W. 1996a. The pottery of Stobi and its outside connections (book review of Anderson-Stojanović 1992). Journal of Roman Archaeology 9, 556-557. Slane, K. W. 1996b. Other ancient ceramics imported from the Mediterranean. In Begley 1996b, 351-368. Slane, K. W. 1997. The fine wares. In S. C. Herbert (ed.), Tel Anafa II,i: the Hellenistic and Roman pottery. Journal of Roman Archaeology Supplementary Series 10, 247406. Ann Arbor, Journal of Roman Archaeology. Slane, K. W. 2003. Corinth’s Roman pottery. Quantification and meaning. In C. K. Williams II and N. Bookidis (eds.), Corinth, the Centenary 1896-1996. Corinth XX, 321-335. Princeton, The American School of Classical Studies at Athens. Slane, K. W. 2004. Corinth: Italian Sigillata and other Italian imports to the early colony. In Poblome et al. 2004, 31-42. Slane, K. W. 2008a. The end of the Sanctuary of Demeter and Kore on Acrocorinth. Hesperia 77, 465-496. Slane, K. W. 2008b. Corinth’s trade with the Adriatic. Rei Cretariae Romanae Fautorum Acta 40, 237-241. Abingdon, Rei Cretariae Romanae Fautores. Slane, K. W. and Magness, J. 2005. Jalame restudied and reinterpreted. Rei Cretariae Romanae Fautorum Acta 39, 257-261. Abingdon, Rei Cretariae Romanae Fautores. Slane, K. W. and Sanders, G.D.R. 2005. Corinth: Late Roman horizons. Hesperia 74, 243-297. Slane, K. W. and Walbank, M. E. H. 2006. Anointing and commemorating the dead: funerary rituals of Roman Corinthians. In Malfitana et al. 2006, 377-387. Slane, K. W., Michael Elam, J., Glascock, M. D. and Neff, H. 1993. Compositional Analysis (NAA) of Eastern Sigillata A and other Wares from Tel Anafa. American Journal of Archaeology 97, 325-326. Slane, K. W., Michael Elam, J., Glascock, M. D. and Neff, H. 1994. Compositional analysis of Eastern Sigillata A and related wares from Tel Anafa (Israel). Journal of Archaeological Science 21, 51-64. Smokotina, A. 2014. The North African Red Slip Ware and amphorae imported into Early Byzantine Bosporus. Rei Cretariae Romanae Fautorum Acta 43, 71-80. Abingdon, Rei Cretariae Romanae Fautores. Sodini, J.-P. 2000. Productions et échanges dans le monde protobyzantin (IVE-VIIE S.): le cas de la céramique. In K. Belke, F. Hild, J. Koder and P. Soustal (eds.), Byzanz als Raum. Zu Methoden und Inhalten der historischen Geographie des östlichen Mittelmeerraumes. Philosophisch-Historische Klasse Denkschriften, 283. Band. Veröffentlichungen der Kommission für die Tabula Imperii Byzantini 7, 181-208. Vienna, Verlag der Österreichischen Akademie der Wissenschaften. Soricelli, G. 1987. Tripolitanian Sigillata: North African or Campanian? Libyan Studies 18, 73-87. Soricelli, G. 2004. La produzione di terra sigillata in Campania. In Poblome et al. 2004, 299-307. Southern P. 2001. The Roman Empire from Severus to Constantine. London, Routledge.

Stephens Crawford, J. 1990. The Byzantine shops at Sardis. Sardis Monographs 9. Cambridge, MA, Harvard University Press. Strommenger, E. 1967. Gefässe aus Uruk von der neubabylonischen Zeit zu den Sassaniden. Ausgrabungen der Deutschen Forschungsgemeinschaft in Uruk-Warka 7. Berlin, Gebrüder Mann Verlag. Stucky, R. A. 1983. Ras Shamra Leukos Limen. Die nachugaritische Besiedlung von Ras Shamra. Mission Archéologique de Ras Shamra 1. Paris, Paul Geuthner. Sutton jr, R. F. 1990. Appendix: ceramics of the Historic period. In J. C. Wright, J. F. Cherry, J. L. Davis, E. Mantzourani and S. B. Sutton, The Nemea Valley Archaeological Project. A preliminary report. Hesperia 59, 646-659. Takaoğlu, T. 2006. New light on the origins of Eastern Sigillata B ware. In Anadolu Arkeolojisine Katkılar. 65. Yaşında Abdullah Yaylalı’ya Sunulan Yazılar, 263270. Tal, O. and Taxel, I. 2010. A re-appraisal of the archaeological findings at Tel Hashash: on the archaeology of the Yarqon estuary from Classical times to Late Antiquity. Palestine Exploration Quarterly 142.2, 95-126. Talbert, R. J. A. (ed.) 2000. Barrington atlas of the Greek and Roman world. Princeton, Princeton University Press. Technau, W. 1929. Griechische Keramik im samischen Heraion. Mitteilungen des Deutschen Archäologischen Instituts, Athenische Abteilung 54, 6-64. Tekkök, B. 1996. The Hellenistic and Roman pottery from Troia: second century B.C. to sixth century A.D. Unpublished PhD thesis, University of MissouriColumbia. Tekkök, B. 2003. Troy from the Julio-Claudian to the Flavian period: its connections with the Mediterranean world. Rei Cretariae Romanae Fautorum Acta 38, 237-242. Abingdon, Rei Cretariae Romanae Fautores. Tekkök, B. and Pernicka, E. 2012. In A. A. Akyol and K. Özdemir (eds.), Türkiye’de Arkeometrinin Ulu Çınarları. Prof. Dr. Ay Melek Özer ve Prof. Dr. Şahinde Demirci’ye Armağan/Two Eminent Contributors to Archaeometry in Turkey. To honour of Prof. Dr. Ay Melek Özer and Prof. Dr. Şahinde Demirci, 345-358. Istanbul, Phoibos Verlag. Tekkök, B., Wallrodt, S., Gündem, C. Y. and Rose, C. B. 2001. Two Roman wells in the lower city of Ilion. Quadrats C29 and w28. Studia Troica 11, 343-382. Thalmann, J.-P., 1978. Tell ‘Arqa (Liban nord). Campagnes I-III (1972-1974). Chantier I. Rapport préliminaire. Syria 55, 1-151. Thoen, H. 2002. Pessinus 2000. Dating the temple area: the evidence of the finds. A preliminary report. Anatolia Antiqua 10, 145-154. Thompson, H. A. 1934. Two centuries of Hellenistic pottery. Hesperia 3, 310-480. Thrümmer, R. 1986. Funde und Datierung. In W. Alzinger, S. Gogos and R. Thrümmer, Aigeira-Hyperesia und die Siedlung Phelloë in Achaia. Österreichische

177

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:44 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East Ausgrabungen auf der Peloponnes 1972-1983 Teil 2: Theater und Umgebung. Klio 68, 57-62. Tidmarsh, J. C. 1990. Excavations in Area XXIII (the Hellenistic period). In P. C. Edwards, S. J. Bourka, K. A. da Costa, J. C. Tidmarsh, A. G. Walmsley and P. M. Watson, Preliminary report on the University of Sydney’s tenth season of excavation at Pella (Tabaqat Fahl) in 1988. Annual of the Department of Antiquities of Jordan 34, 71-76. Tidmarsh, J. C. 2005. Pottery: the imported wares. In G. W. Clarke, Jebel Khalid: the 2004 and 2005 seasons. Mediterranean Archaeology 18, 148-157. Timby, J. R. 1994. Part II: Red-slipped wares. In Fulford and Tomber (eds.) 1994, 67-117. Tobin, J. 2004. Black Cilicia. A study of the Plain of Issus during the Roman and Late Roman periods. British Archaeological Reports International Series 1275). Oxford, Archaeopress. Tölle-Kastenbein, R. 1974. Keramik. In R. TölleKastenbein, Das Kastro Tigani. Die Bauten und Funde griechischer, römischer und byzantinischer Zeit. Samos 14, 138-163. Bonn, Rudolf Habelt. Tomber, R. S. 1988. Pottery from the 1982-3 excavations. In J. H. Humphrey (ed.), The Circus and a Byzantine Cemetery at Carthage I, 437-528. Ann Arbor, University of Michigan Press. Tomber, R. S. 1992. Roman fine wares. In R. M. Harrison, Amorium excavations 1991. The fourth preliminary report. Anatolian Studies 42, 216. Tomber, R. S. 1993a. Quantitative approaches to the investigation of long-distance exchange. Journal of Roman Archaeology 6, 142-166. Tomber, R. S. 1993b. Finewares. In R. M. Harrison, Excavations at Amorium: 1992 interim report. Anatolian Studies 43, 154-155. Tomber, R. S. 1998. The pottery. In S. E. Sidebotham and W. Z. Wendrich (eds.), Berenike 1996. Report of the 1996 excavations at Berenike (Egyptian Red Sea Coast) and the survey of the Eastern Desert. Centre of Non-Western Studies Publications Special Series 3, 163-180. Leiden, Centre of Non-Western Studies. Tomber, R. S. 1999a. Pottery from the sediments of the Inner Harbour (Area I14). In K. Holum, A. Raban and J. Patrich (eds.), Caesarea Papers 2. Herod’s temple, the provincial governor’s praetorium and granaries, the later harbour, a gold coin hoard, and other studies. Journal of Roman Archaeology Supplementary Series 35, 295-322. Ann Arbor, Journal of Roman Archaeology. Tomber, R. S. 1999b. The pottery. In S. E. Sidebotham and W. Z. Wendrich (eds.), Berenike 1997. Report of the 1997 Excavations at Berenike and the Survey of the Egyptian Eastern Desert, including Excavations at Shenshef. Centre of Non-Western Studies Publications Special Series 4, 123-159. Leiden, Centre of NonWestern Studies. Tomber, R. S. 2001. The pottery. In V. A. Maxfield and D. P. S. Peacock, The Roman imperial quarries. Survey and excavation at Mons Porphyrites 1994-1998.

Volume I: topography and quarries. Egypt Exploration Society Excavation Memoir 67, 241-303. London, Egypt Exploration Society. Tomber, R. S. 2002. The ceramic finds from Maroni Petrera. In S. W. Manning, The Late Roman church at Maroni Petrera. Survey and salvage excavations 19901997, and other traces of Roman remains in the lower Maroni Valley, Cyprus, 41-57. Nicosia, A. G. Leventis Foundation. Tomber, R. S. 2004. Polarising and integrating the Late Roman economy: the role of Late Roman Amphorae 1-7. Ancient West & East 3.1, 155-165. Tomber, R. S. 2007a. Pottery from the excavated deposits. In D. P. S. Peacock and V. A. Maxfield, The Roman imperial quarries. Survey and excavation at Mons Porphyrites 1994-1998. Volume II: the excavations. Egypt Exploration Society Excavation Memoir 82, 177-208. London, Egypt Exploration Society Tomber, R. S. 2007b. Roman pottery from the Pattanam 2007 excavations: preliminary report. In P. J. Cherian, V. Selvakumar and K. P. Shajan, Interim Report of the Pattanam Excavations/Explorations 2007. Trivandrum. Tomber, R. S. 2008. Indo-Roman trade: from pots to pepper. London, Duckworth. Tortorella, S. 1986. La ceramica fine da mensa Africana dal IV al VII secolo D.C. In A. Giardina (ed.), Società romana e impero tardoantico volume terzo. Le merci, gli insediamenti, 211-225. Rome, Editori Laterza. Tortorella, S. 1987. La ceramica Africana: un riesame della problematica. In P. Lévêque and J.-P. Morel (eds.), Céramiques hellénistiques et romaines 2. Centre de Recherches d’Histoire Ancienne, 279-327. Paris, Les Belles Lettres. Turnovsky, P. 2005. Late Antique and Byzantine pottery of the church of St. Mary in Ephesos. An introduction. Rei Cretariae Romanae Fautorum Acta 39, 217-224. Abingdon, Rei Cretariae Romanae Fautores. Unterkircher, E. 1983. Terra sigillata aus dem Heraion von Samos. Mitteilungen des Deutschen Archäologischen Instituts, Athenische Abteilung 98, 172-214. Vaag, L. E. 2002. Black and red plain pottery. In L. E. Vaag, V. Nørskov and J. Lund, The Maussolleion at Halikarnassos. 7. The pottery. Ceramic material and other finds from selected contexts. Jutland Archaeological Society Publications 15:7, 28-38. Aarhus, Aarhus University Press. Vaag, L. E. 2003. A closer look at the making of Phocaean Red Slip ware bowls. In C. Abadie-Reynal (ed.), Les céramiques an Anatolie aux époques hellénistique et romaine. Actes de la Table Ronde d’Istanbul, 22-24 mai 1996. Varia Anatolica 15, 203-207. Paris, De Boccard. Vaag, L. E. 2005. Phocaean Red Slip Ware – main and secondary productions. In M. Berg Briese and L. E. Vaag (eds.), Trade relations in the Eastern Mediterranean from the Late Hellenistic period to Late Antiquity: the ceramic evidence. Acts from a Ph.D.seminar for young scholars, Sandbjerg Manorhouse,

178

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:44 UTC All use subject to https://about.jstor.org/terms

Bibliography 12-15 February 1998. Halicarnassian Studies 3, 132138. Odense, University Press of Southern Denmark. Vance Watrous, L. 1982. Lasithi. A history of settlement on a highland plain in Crete. Hesperia Supplement 18. Princeton, The American School of Classical Studies at Athens. Vance Watrous, L., Xatzi-Vallianou, D., Pope, K., Mourtzas, N., Shay, J., Shay, C. T., Bennet, J., Tsoungarakis, D., Angelomati-Tsoungarakis, E., Vallianos, C. and Blitzer, H. 1993. A survey on the western Mesara plain in Crete: preliminary report of the 1984, 1986 and 1987 field seasons. Hesperia 62, 191-248. Vanderhoeven, M. 1989. Les terres sigillées (1966-1972). Fouilles d’Apamée de Syrie 9.1. Brussels, Centre belge de recherches archéologiques à Apamée de Syrie. Vickers, M. 1994. Nabataea, India, Gaul, and Carthage: reflections on Hellenistic and Roman gold vessels and red-gloss pottery. American Journal of Archaeology 98, 231-248. Villeneuve, E. and Watson, P. M. (eds.) 2001. La céramique byzantine et proto-islamique en Syrie-Jordanie (IVeVIIIe siècles apr. J.-C.). Actes du colloque tenu à Amman les 3, 4 et 5 décembre 1994. Bibliothèque Archéologique et Historique 159. Beirut, Institut Français d’Archéologique du Proche-Orient. Vogeikoff-Brogan, N. 2000. Late Hellenistic pottery in Athens: a new deposit and further thoughts on the association of pottery and societal change. Hesperia 69, 293-333. Vogt, Ch. 2004. Echanges et économie proto-byzantines d’Eleutherna d’apres les temoins céramiques. In M. Livadiotti and I. Simiakaki (eds.), Creta romana e protobyzantina, 923-944. Padua, Aldo Ausilio. Vokaer, A. 2013. Pottery production and exchange in late antique Syria (4th-8th c. A.D.). A study based on a selection of imported and local wares. In L. A. Lavan (ed.), Local economies? Production and exchange of inland regions in Late Antiquity. Late Antique Archaeology 10.1, 567-606. Leiden, Brill. Vroom, J. C. 1993. The kastro of Veloukovo (Kallion), a note on the surface finds. Pharos. Journal of the Netherlands Institute at Athens 1, 113-138. Vroom, J. C. 1998. The Late Roman – Early Byzantine finds from the excavations at the eastern city of Limyra. In J. Borchardt, Bericht der Grabungskampagne in Limyra 1997. Kazı Sonuçları Toplantısı 20.2, 143-145. Vroom, J. C. 2004. Late Antique pottery, settlement and trade in the east Mediterranean: a preliminary comparison of ceramics from Limyra (Lycia) and Boeotia. In W. Bowden, L. A. Lavan and C. Machado (eds.), Recent research on the Late Antique countryside. Late Antique Archaeology 2, 281-331. Leiden, Brill. Vroom, J. C. 2005. New light on ‘Dark Age’ pottery: a note of finds from south-western Turkey. Rei Cretariae Romanae Fautorum Acta 39, 249-255. Abingdon, Rei Cretariae Romanae Fautores. Waagé, F. O. 1933. The American excavations in the Athenian Agora, first report: the Roman and Byzantine pottery. Hesperia 2, 279-328.

Waagé, F. O. 1934. The pottery. In G. W. Elderkin (ed.), Antioch-on-the-Orontes 1: The Excavations of 1932, 67-73. Princeton, Princeton University Press. Waagé, F. O. 1937. Vasa Samia. Antiquity 11, 46-55. Waagé, F. O. 1948. Hellenistic and Roman tableware of north Syria. In F. O. Waagé (ed.), Antioch on-theOrontes 4: part one, ceramics and Islamic coins, 1-60. Princeton, Princeton University Press. Wallace-Hadrill, A. 2008. Rome’s cultural revolution. Cambridge, Cambridge University Press. Ward-Perkins, B. 2001. Specialisation, trade, and prosperity: an overview of the economy of the Late Antique eastern Mediterranean. In S. A. Kingsley and M. J. Decker (eds.), Economy and exchange in the East Mediterranean during Late Antiquity. Proceedings of a conference at Somerville College, Oxford, 29th May, 1999, 167-178. Oxford, Oxbow Books. Watson, P. M. 1990. Byzantine remains on the Summit of Tell el-Husn (Area XXXIV). In P. C. Edwards, S. J. Bourka, K. A. da Costa, J. C. Tidmarsh, A. G. Walmsley and P. M. Watson, Preliminary report on the University of Sydney’s tenth season of excavation at Pella (Tabaqat Fahl) in 1988. Annual of the Department of Antiquities of Jordan 34, 76-80. Weber-Hiden, I. 2003. Keramik aus hellenistischer bis frühbyzantinischer Zeit aus Tavium/Büyük Nefes: Bermerkungen und Übersicht über das Begehungsmaterial der Kampagnen 1998-2000 aus drei ausgewählten Bereichen des Stadtgebiedes. Anatolia Antiqua 11, 253-322. Welles, C. B. 1938. Pottery inscriptions. In C. H. Kraeling (ed.), Gerasa, city of the Decapolis, 460-461. New Haven, American Schools of Oriental Research. Wells, C. M. 1990. What is the Conspectus? A note to the reader. In Ettlinger et al. 1990, 1-2. Westholm, A. 1956. Pottery. In: O. Vessberg and A. Westholm, The Hellenistic and Roman periods in Cyprus. The Swedish Cyprus Expedition 4, 3, 53-81. Stockholm, The Swedish Cyprus Expedition. Wheeler, R. E. M., Ghosh, A. and Deva, K. 1946. Arikamedu, an Indo-Roman trading-station on the east coast of India. Ancient India 2, 17-125. Whitcomb, D. S. 1982. Roman ceramics. In D. S. Whitcomb and J. H. Johnson, Quseir al-Qadim 1980: preliminary report. American Research Center in Egypt Reports 7, 51-115. Malibu, Undena Publications. Whittaker, C. R. 1983. Late Roman trade and traders. In P. Garnsey, K. Hopkins and C. R. Whittaker (eds.), Trade in the ancient economy, 163-180. London, Chatto & Windus. Wickham, C. 1988. Marx, Sherlock Holmes, and Late Roman commerce. The Journal of Roman Studies 78, 183-193. Wickham, C. 1998. Overview: production, distribution and demand. In R. Hodges and W. Bowden (eds.), The sixth century. Production, distribution and demand. The transformation of the Roman world 3, 279-292. Leiden, Brill.

179

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:44 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East Wickham, C. 2005. Framing the early Middle Ages. Europe and the Mediterranean, 400-800. Oxford, Oxford University Press. Willet, R. 2012. Red slipped complexity. The sociocultural context of the concept and use of tableware in the Roman east. Unpublished PhD thesis, University of Leuven. Willet, R. and Poblome, J. 2011. The pale red slipped dot. FACTA. A Journal of Roman Material Cultural Studies 5, 101-110. Williams, C. 1977. A Byzantine well-deposit from Anemurium (Rough Cilicia). Anatolian Studies 27, 175-190. Williams, C. 1985. The pottery and glass at Alahan. In M. Gough (ed.), Alahan. An early Christian monastery in southern Turkey. Based on the work of Michael Gough. Pontifical Institute of Mediaeval Studies. Studies and Texts 73, 35-61. Toronto, Pontifical Institute of Mediaeval Studies. Williams, C. 1989. Anemurium. The Roman and early Byzantine pottery. Toronto, Pontifical Institute of Mediaeval Studies. Williams II, C. K. 1980. Corinth excavations, 1979. Hesperia 49, 107-134. Williams II, C. K. and Fisher, J. E. 1975. Corinth, 1974: Forum southwest. Hesperia 44, 1-29. Williams II, C. K. and Fisher, J. E. 1976. Corinth, 1975: Forum southwest. Hesperia 45, 99-137. Williams II, C. K. and Zervos, O. H. 1982. Corinth, 1981: East of the Theater. Hesperia 51, 115-163. Williams II, C. K. and Zervos, O. H. 1983. Corinth, 1982: East of the Theater. Hesperia 52, 1-47. Williams II, C. K. and Zervos, O. H. 1984. Corinth, 1984 - East of the Theater. Hesperia 54, 55-96. Williams II, C. K. and Zervos, O. H. 1986. Corinth, 1985: East of the Theater. Hesperia 55, 129-175. Williams II, C. K. and Zervos, O. H. 1989. Corinth, 1988: East of the Theater. Hesperia 58, 1-50. Williams, C. and Williams, H. 1989. Excavations at Mytilene, 1988. Sixth season. Echos du Monde Classique/Classical Views 33, 167-181. Williams, C. and Williams, H. 1990. Excavations at Mytilene, 1989. Echos du Monde Classique/Classical Views 24, 181-193. Wilson, A. I. 2009. Approaches to quantifying Roman trade. In. A. K. Bowman and A. I. Wilson (eds.), Quantifying the Roman economy. Methods and problems. Oxford Studies on the Roman Economy 1, 213-249. Oxford, Oxford University Press. Wintermeyer, U. 1980. Katalog ausgewählter Keramik und Kleinfunde. In K. Tüchelt, Didyma. Bericht über die Arbeiten der Jahre 1975-1979. Istanbuler Mitteilungen 30, 122-160. Wintermeyer, U. 2004. Die hellenistische und frühkaiserzeitliche Gebrauchskeramik. Auf Grundlage der stratifizierten Fundkeramik aus dem Bereich der heiligen Strasse. Didyma 3, 2. Mainz, Philipp von Zabern.

Winther Jacobsen, K. 2006. Pots for the dead. Pottery and ritual in Cypriote tombs of the Hellenistic and Roman period. In Malfitana et al. 2006, 389-396. Wiseman, J. 1967. Excavations at Corinth, the Gymnasium area, 1966. Hesperia 36, 402-428. Wiseman, J. 1972. Excavations in Corinth, the Gymnasium area, 1969-1970. Hesperia 41, 1-42. Woolley, C. L. and Randall-MacIver, D. 1910. Karanòg, the Romano-Nubian cemetery. Eckley B. Coxe junior Expedition to Nubia: 3/4. Philadelphia, The University Museum. Wrabetz, J. F. 1977. A new Serenus stamping from Sardis and the origins of the Eastern Sigillata B ware. Harvard Studies in Classical Philology 81, 195-197. Wright, K. S. 1980. A Tiberian pottery deposit from Corinth. Hesperia 49, 135-177. Xanthopoulou, M. 2004. Παλαιοχριστιανική κεραμική της αρχαίας Ιτάνου. In M. Livadiotti and I. Simiakaki (eds.), Creta romana e protobyzantina, 1013-1028. Padua, Aldo Ausilio. Yangaki, A. 2005. La céramique des IVe-VIIIe siècles ap. J.-C. d’Éleutherna. Athens, University of Crete. Yener, K. A. 2006. The Amuq Valley Regional Projects, volume 1: surveys in the plain of Antioch and Orontes delta, Turkey, 1995-2002. Oriental Institute Publications 131. Chicago, The Oriental Institute. Yener-Marksteiner, B. 2009. Ausgewählte spätrömischfrühbyzantinischen Keramik aus einem Gebäudekomplex in der Weststadt Limyras. In T. Marksteiner and B. Yener-Marksteiner, Die Grabungen in Sondage 30/36/37 in der Weststadt von Limyra: der archäologische Befund und die Keramik. Jahreshefte des Österreichischen Archäologischen Instituts in Wien 78, 227-252. Yener-Marksteiner, B. 2012. Frühkaiserzeitliche Sigillata der Weststadtgrabung in Limyra 2002-2003. In M. Seyer (ed.), 40 Jahre Grabung Limyra. Akten des internationalen Symposions, Wien, 3.–5. Dezember 2009, 371-386. Vienna, Phoibos Verlag. Yilmaz, Z. 2007. Spätantike Sigillaten aus Priene. In B. Böhlendorf-Arslan, A. O. Uysal and J. Witte-Orr (eds.), Çanak. Late Antique and Medieval pottery and tiles in Mediterranean archaeological contexts. Proceedings of the First International Symposium on Late Antique, Byzantine, Seljuk and Ottoman pottery and tiles in archaeological context (Çanakkale, 1-3 June 2005). BYZAS 7, 123-129. Istanbul, Ege Yayınları. Zabehlicky-Scheffenegger, S. 1995. Subsidiary factories of Italian Sigillata potters: the Ephesian evidence. In H. Koester (ed.), Ephesos. Metropolis of Asia. An interdisciplinary approach to its archaeology, religion, and culture. Harvard Theological Studies 41, 217-228. Valley Forge, Harvard University Press. Zabehlicky-Scheffenegger, S. 1996. Der Italiener in Ephesos. Rei Cretariae Romanae Fautorum Acta 34, 253-271. Abingdon, Rei Cretariae Romanae Fautores. Zabehlicky-Scheffenegger, S. 2004. Italische Sigillata in Ephesos. In Poblome et al. 2004, 73-80.

180

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:44 UTC All use subject to https://about.jstor.org/terms

Bibliography Zachos, G. A. 2010. H Κέα στους ύστερους ρωμαϊκούς χρόνους/Keos in Late Roman context. In PapanikolaBakirtzis and Kousoulakou (eds.) 2010, 782-794. Zahn, R. 1904. Thongeschirr (Scherben von Sigillata gefässen). In T. Wiegand and H. Schrader, Priene. Ergebnisse der Ausgrabungen und Untersuchungen in den Jahren 1895-1898, 430-449. Berlin, Georg Reimer. Zelle, M. 1997. Die Terra Sigillata aus der WesttorNekropole in Assos. Asia Minor Studien 27. Bonn, Rudolf Habelt. Zelle, M. 2004. Funde spätantiker Sigillata in Assos. In B. Liesen and U. Brandl (eds.), Römische Keramik. Herstellung und Handel. Kolloquium Xanten, 15.17.6.2000. Xantener Berichte 13, 77-106. Mainz, Philipp von Zabern.

Zelle, M. 2007. Späthellenistische und frühkaiserzeitliche Keramik in Pednelissos und ihre Aussagekraft zu kulturellen Wandel. In Meyer (ed.) 2007, 195-201. Zoroğlu, L. 1989. Some Roman names on Eastern Sigillata A from Samosata. In D. H. French and C. S. Lightfoot (eds.), The eastern frontier of the Roman Empire. Proceedings of a colloquium held at Ankara in September 1988. British Institute of Archaeology at Ankara Monograph 11. British Archaeological Reports International Series 553ii, 573-579. Oxford, Archaeopress. Zoroğlu, L. 2005. Roman fine wares in Cilicia: an overview. Rei Cretariae Romanae Fautorum Acta 39, 243-248. Abingdon, Rei Cretariae Romanae Fautores.

181

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:44 UTC All use subject to https://about.jstor.org/terms

Archaeopress

Chapter Title: Appendices Book Title: Once upon a Time in the East Book Subtitle: The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East Book Author(s): Philip Bes Published by: Archaeopress. (2015) Stable URL: https://www.jstor.org/stable/j.ctvr43kch.14 JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected]. Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at https://about.jstor.org/terms

Archaeopress is collaborating with JSTOR to digitize, preserve and extend access to Once upon a Time in the East

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:45 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East

Appendix 1a. Map showing all sites and surveys from where data was collected (© Philip Bes/ICRATES Project).

182

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:45 UTC All use subject to https://about.jstor.org/terms

Appendices

Appendix 1b. Map showing sites and surveys otherwise mentioned in the text (© Philip Bes/ICRATES Project).

183

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:45 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East

Appendix 2. Table showing, per region, the sites and surveys from where data was collected (© Philip Bes/ICRATES Project).

184

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:45 UTC All use subject to https://about.jstor.org/terms

Appendices

Appendix 3a. Table showing the absolute and relative quantities, per region, of ITS, ESA, ESB, ESC, ESD and ARSW for the late Hellenistic-early Roman period (© Philip Bes/ICRATES Project).

Appendix 3b. Table showing the absolute and relative quantities, per region, of ESC, ESD, ARSW, LRD and LRC for the mid- to late Roman period (© Philip Bes/ICRATES Project).

185

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:45 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East

Phase 1a

Phase 1b

Phase 2a

Phase 2b

ESA

ESC

ESD

ESA

ESC

ESD

ITS

ESA

ESB

ESC

ESD

ITS

Abdera

 

 

 

 

 

 

1

 

2

 

 

1

Adulis

 

 

 

1

 

 

 

 

 

 

 

 

Aetolia

 

 

 

 

 

 

1

1

 

1

 

1

Aigeira

 

 

 

 

 

 

3

 

 

 

 

2

‘Ain Dara

1

 

 

1

 

 

 

 

 

 

 

 

Aizanoi

 

 

 

 

 

 

 

 

4

 

 

 

Alexandria

 

3

 

22

1

 

15

11

2

3

1

3

Alexandria Troas

 

 

 

 

 

 

 

 

 

 

 

 

Amathous

4

 

1

20

 

 

6

23

 

 

2

 

Amorion

 

 

 

 

 

 

 

 

1

1

 

 

Amphipolis

 

 

 

1

 

 

1

 

 

 

 

 

Anemorion

2

 

 

8

 

4

 

2

 

2

2

 

Antioch

45

 

 

29

49

1

 

1

7

Site

9

 

 

Antiocheia ad Pisidiam

 

 

 

 

 

 

 

 

 

 

 

 

Apamea

19

 

 

54

 

 

2

71

 

 

 

 

Argos

 

 

 

 

 

 

10

 

 

4

 

 

Arikamedu

 

 

 

 

 

 

18

1

 

 

 

1

Arsameia am Nymphaios

1

 

 

9

 

 

 

 

 

 

 

 

Ashdod

 

 

 

15

 

 

 

2

 

 

 

 

Ashkelon

1

 

 

1

 

 

1

2

 

 

 

 

Assos

 

11

 

6

16

 

3

5

5

150

 

 

Athens

 

 

 

17

1

 

13

17

2

4

 

4

Athis/Neokaesareia

 

 

 

 

 

 

 

 

 

 

 

 

Axum

 

 

 

 

 

 

 

 

 

 

 

 

Ayios Philon

3

 

 

5

 

 

 

5

 

 

 

 

Beirut

 

 

 

1

 

 

10

5

 

 

1

 

Berbati

 

 

 

 

 

 

 

 

 

 

 

 

Berenice

3

 

 

22

 

1

96

28

 

 

3

7

Berenike

 

 

 

 

 

 

5

 

1

 

 

 

Byblos

 

 

 

 

 

 

 

 

 

 

 

 

Caesarea Maritima

9

 

 

4

 

 

 

8

 

 

 

 

Corinth

1

 

 

8

 

 

72

17

7

2

 

11

Cyrene

 

 

 

5

 

 

4

1

 

 

 

 

Damascus

 

 

 

1

 

 

 

 

 

 

 

 

Delos

1

 

 

10

1

 

 

 

1

 

 

 

Delphi

 

 

 

 

 

 

 

 

 

 

 

 

Didyma

 

 

 

 

 

 

 

 

 

 

 

 

Diokaesareia

 

 

 

3

 

 

 

1

 

 

 

 

Doliche

1

 

 

5

 

 

 

5

 

 

 

 

Domuztepe

3

 

 

3

 

 

 

2

 

 

 

 

Dor

7

 

 

21

 

 

 

2

 

 

2

 

Dura Europos

5

 

 

10

 

 

 

8

 

 

 

 

Ed-Dur

 

 

 

2

 

 

1

3

 

 

 

 

Emporio

 

 

 

 

 

 

 

 

 

 

 

 

Ephesos

 

1

 

28

4

 

40

31

38

7

 

3

186

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:45 UTC All use subject to https://about.jstor.org/terms

Appendices

Phase 4

Phase 3 ITS

ESA

ESB

ESC

ESD

ITS

ESA

ESB

ESC

ESD

ARSW

 

2

 

 

 

 

11

63

66

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1

 

 

 

4

 

 

 

 

1

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

7

 

 

 

 

3

 

 

 

8

5

4

 

1

 

11

 

 

8

 

 

 

 

 

 

 

 

 

3

 

 

8

10

 

 

11

 

8

1

 

11

 

 

 

1

 

 

 

 

2

 

 

 

 

 

 

 

 

 

3

9

126

 

 

4

2

1

 

8

1

15

5

1

17

3

10

86

 

 

 

2

61

1

 

 

7

 

 

2

 

 

 

1

3

 

 

 

3

66

 

 

 

 

40

 

 

 

 

20

 

 

4

 

5

 

6

11

 

 

7

 

 

 

 

 

 

 

 

 

 

 

5

 

 

 

 

 

 

 

 

 

 

2

 

 

 

 

 

 

 

 

 

1

2

 

 

1

 

 

 

 

 

 

7

5

5

124

 

 

2

34

122

 

 

12

3

6

 

 

1

6

26

3

1

 

 

3

 

 

 

 

7

 

 

 

 

 

 

1

 

 

 

 

 

 

 

 

 

1

 

 

1

 

4

1

 

2

 

18

8

 

 

2

 

12

 

 

3

 

 

 

 

 

 

 

1

 

 

 

 

90

12

5

5

3

11

13

22

19

7

17

10

 

3

 

 

1

2

2

 

1

 

1

5

 

 

1

 

3

 

 

 

 

1

3

 

 

 

 

7

 

 

1

1

97

8

21

6

2

26

4

17

2

1

2

9

 

 

 

 

4

 

1

1

1

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

2

 

 

 

 

5

 

 

 

 

 

 

 

 

 

 

 

 

2

 

 

 

 

3

 

 

 

 

2

 

 

 

 

 

7

 

 

 

 

4

 

 

 

 

 

1

 

 

 

 

2

 

 

 

 

 

1

 

 

 

 

4

 

 

3

 

 

2

 

 

 

 

4

 

 

 

 

 

1

1

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

2

 

 

 

14

13

133

3

 

 

2

97

12

 

3

187

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:45 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East

Phase 1a

Phase 1b

Phase 2a

Phase 2b

ESA

ESC

ESD

ESA

ESC

ESD

ITS

ESA

ESB

ESC

ESD

ITS

Epiphaneia

19

 

 

296

 

 

7

135

 

 

5

 

Eretria

 

 

 

4

 

 

 

1

1

 

 

 

Gadara

1

 

 

9

 

 

 

1

 

 

 

 

Gerasa

 

 

 

 

 

 

 

 

 

 

 

 

Gindaros

29

 

 

66

 

 

 

34

 

1

2

 

Gortyna

2

 

 

8

 

1

23

7

1

1

1

4

Halikarnassos

 

 

 

3

 

 

 

 

 

 

 

 

Hammath Tiberias

 

 

 

 

 

 

 

1

 

 

 

 

Hippos-Sussita

7

 

 

18

 

 

 

13

 

 

1

 

Hyettos

 

 

 

1

 

 

 

 

 

 

 

 

Iasos

 

 

 

 

 

 

5

 

 

 

 

 

Isthmia

 

 

 

 

 

 

 

1

 

1

 

 

Jalame

3

 

1

1

 

 

 

 

 

 

 

 

Jebel Khalid

3

 

 

13

 

1

 

 

 

 

 

 

Jericho

 

 

 

 

 

 

 

2

 

 

 

 

Jerusalem

1

 

 

22

 

 

10

28

 

 

7

2

Kabirion

 

 

 

 

 

 

4

 

 

 

 

 

Kallirhoe

 

 

 

 

 

 

 

 

 

 

 

 

Kanatha

 

 

 

2

 

 

 

2

 

 

 

 

Karamildan

4

 

 

9

 

 

 

 

 

 

 

 

Karanòg

 

 

 

 

 

 

 

 

 

 

 

 

Kenchreai

1

 

 

7

 

 

8

17

2

1

1

 

Keos

 

 

 

 

 

 

 

1

 

 

 

 

Khor Rori

 

 

 

 

 

 

 

 

 

 

 

 

Kition-Bamboula

1

 

 

6

 

 

 

2

 

 

 

 

Knossos

1

 

 

30

 

3

58

29

6

2

7

4

Kommos

 

 

 

 

 

 

1

 

 

 

 

 

Kopetra

 

 

 

 

 

 

 

 

 

 

 

 

Koroneia

 

 

 

 

 

 

 

 

 

 

 

 

Site

Kourion

 

 

 

 

 

 

1

3

 

1

1

 

Kozluca

 

 

 

1

 

 

 

 

 

 

 

 

Küçük Burnaz

1

 

 

5

 

 

 

1

 

 

 

 

Kululu

6

 

 

6

 

 

 

 

 

 

 

 

Kydonia

 

 

 

 

 

 

2

 

 

 

 

 

Kythera

 

 

 

 

 

 

 

 

 

 

 

 

Labraunda

 

 

 

4

 

 

 

2

 

 

 

 

Leukos Limen

1

 

 

 

 

 

 

5

 

 

 

 

Lidar Höyük

1

 

 

10

 

 

 

6

 

 

 

 

Limassol

 

 

 

 

 

 

 

 

 

 

 

 

Mampsis

 

 

 

 

 

 

 

1

 

 

 

 

Marina el-Alamein

 

 

 

1

 

 

 

 

 

 

 

 

Maroni Petrera

 

 

 

 

 

 

 

 

 

 

 

 

Meroë

 

 

 

2

 

 

 

1

1

 

 

 

Methana

 

 

 

1

 

 

 

 

 

 

 

 

Methymna

 

 

 

2

 

 

 

 

 

 

 

 

188

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:45 UTC All use subject to https://about.jstor.org/terms

Appendices

Phase 4

Phase 3 ITS

ESA

ESB

ESC

ESD

ITS

ESA

ESB

ESC

ESD

ARSW

2

45

 

 

 

 

25

 

 

1

 

4

 

 

1

 

 

2

7

4

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1

 

 

 

 

5

 

 

 

 

 

25

 

 

 

 

33

1

 

 

 

24

2

1

2

 

2

6

29

13

1

2

 

 

 

 

 

 

 

 

 

 

 

 

1

 

 

 

 

 

 

 

 

 

 

3

 

 

 

 

1

 

 

 

 

 

 

 

 

 

1

 

1

 

 

 

2

 

5

 

 

 

 

2

 

 

 

3

 

 

 

 

 

 

3

 

 

 

 

 

 

 

 

 

 

 

 

 

1

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

7

22

1

1

4

 

20

1

 

8

 

3

 

 

 

 

 

 

 

 

 

 

 

1

 

 

 

 

 

 

 

 

 

 

3

 

 

 

 

23

 

 

 

 

 

 

 

 

 

 

1

 

 

 

 

 

1

 

 

 

 

 

1

 

 

 

3

5

2

 

3

 

6

3

2

1

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1

 

 

 

 

1

 

 

 

 

 

 

 

1

 

7

 

 

2

 

31

3

25

2

4

4

2

57

35

2

5

3

 

 

 

 

1

 

1

2

 

 

 

 

 

 

 

 

 

 

 

4

 

2

 

 

 

 

 

 

 

 

 

1

2

 

 

 

2

 

 

3

 

9

 

 

 

 

 

 

 

 

 

 

 

 

 

1

 

 

 

 

6

 

 

 

1

 

 

 

 

 

 

 

 

 

 

 

1

 

 

1

 

1

 

 

 

 

 

 

 

 

 

 

 

 

1

 

1

1

 

 

5

 

 

 

 

15

 

 

 

 

11

 

 

 

 

8

 

 

 

 

 

1

 

 

 

 

6

 

 

 

 

 

2

 

 

 

 

 

 

 

 

 

 

2

 

 

 

 

11

 

 

 

 

1

 

 

 

1

 

 

 

 

11

 

 

 

 

 

 

 

 

 

 

7

 

 

 

1

 

 

 

3

 

 

 

 

 

 

 

 

 

 

 

1

1

 

 

 

 

 

 

 

 

 

 

 

 

 

189

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:45 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East

Phase 1a

Phase 1b

Phase 2a

Phase 2b

ESA

ESC

ESD

ESA

ESC

ESD

ITS

ESA

ESB

ESC

ESD

ITS

Mutatio Heldua

1

 

 

2

 

 

 

1

 

 

 

 

Myos Hormos

 

 

 

2

 

1

19

19

2

 

 

2

Nemea

 

 

 

 

 

 

 

 

 

 

 

 

Nessana

1

 

 

157

 

 

 

14

 

 

13

 

Notion

 

 

 

 

 

 

3

 

 

 

 

 

Oboda

 

 

 

10

 

 

12

12

 

 

29

 

Olympia

 

 

 

 

 

 

 

 

1

 

 

 

Oumm ‘el-Amed

 

 

 

4

 

 

1

 

 

 

 

1

Palaipaphos

18

 

 

8

10

 

 

10

 

Site

4

 

1

Panayia Ematousa

 

 

1

4

 

 

 

7

 

 

3

 

Paphos

27

 

8

75

 

26

8

31

 

3

26

 

Patras

 

 

 

 

 

 

11

2

 

 

 

2

Pella

 

 

 

1

 

 

 

 

 

 

 

 

Pelusium

 

 

 

 

 

 

 

1

 

 

 

 

Pergamon

 

2

 

 

3

 

 

 

 

16

 

 

Perge

 

 

 

 

1

 

1

 

 

 

1

 

Petra

13

 

 

48

 

 

19

20

1

 

1

2

Phalasarna

1

 

 

 

 

 

 

 

 

 

 

 

Philadelphia

 

 

 

1

 

 

 

2

 

 

 

 

Phlius

 

 

 

 

 

 

 

 

 

 

 

 

Porphyreon

 

 

 

 

 

 

 

2

 

 

 

 

Porsuk

 

 

 

6

 

 

 

2

 

 

 

 

Priene

 

 

 

1

 

 

 

 

5

 

 

 

Pylos

 

 

 

 

 

 

 

3

 

 

 

 

Qana’

 

 

 

 

 

 

4

 

 

 

 

 

Qusair as-Saila

 

 

 

 

 

 

 

 

 

 

 

 

Resafa

 

 

 

 

 

 

 

 

 

 

 

 

Salamis/Constantia

 

 

 

3

 

 

1

1

 

 

 

 

Samaria-Sebaste

11

 

 

48

 

 

4

26

 

 

6

 

Samos

4

1

 

13

3

 

2

12

1

3

 

 

Sardis

 

 

 

1

 

 

 

 

 

 

 

 

Scythopolis

1

 

 

2

 

 

3

3

 

 

 

 

Seleukeia ad Tigrim

1

 

 

 

 

 

 

 

 

 

 

 

Siphnos

 

 

 

3

 

 

1

1

 

 

 

 

Smintheion

1

 

 

 

1

 

 

1

 

13

 

 

Sparta

 

 

 

 

 

 

 

 

 

2

 

 

Stobi

1

 

 

5

 

 

20

2

 

2

 

2

Sultantepe

2

 

 

2

 

 

 

4

 

 

 

 

Tall Šēh Hamad

 

 

 

 

 

 

 

 

 

 

 

 

Tanagra

 

 

 

6

 

 

6

 

 

 

 

 

Tarsos

5

 

 

105

 

 

8

22

1

 

4

 

Tel Anafa

22

1

 

176

2

2

1

45

1

2

4

 

Tel Mevorakh

 

 

 

1

 

 

 

1

 

 

3

 

Tell Arqa

1

 

 

9

 

 

2

4

 

 

 

 

Tell Atrib

 

 

 

6

 

 

 

4

 

 

 

 

190

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:45 UTC All use subject to https://about.jstor.org/terms

Appendices

Phase 4

Phase 3 ITS

ESA

ESB

ESC

ESD

ITS

ESA

ESB

ESC

ESD

ARSW

 

2

 

 

 

 

3

 

 

 

 

7

4

4

 

1

 

4

 

 

2

 

1

 

 

 

 

 

 

 

 

 

 

 

7

 

 

 

 

9

 

 

3

 

 

 

 

 

 

 

 

 

 

 

 

4

16

1

 

5

 

4

 

 

4

 

10

 

 

 

 

 

 

1

 

 

 

1

1

 

 

 

 

 

 

 

 

 

8

5

 

 

13

 

6

1

 

30

1

3

6

 

 

5

 

9

 

 

16

 

26

13

4

3

35

26

40

34

2

115

2

1

1

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1

 

 

 

1

 

 

 

 

3

 

2

 

1

7

 

 

 

 

2

 

 

1

6

 

 

1

 

3

1

3

1

7

14

9

 

 

1

1

65

1

 

1

1

 

 

 

 

 

 

 

 

 

 

 

1

3

 

 

 

 

5

 

 

 

 

1

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5

 

 

 

 

 

2

 

 

 

 

2

 

 

 

 

 

 

15

 

 

 

 

 

 

 

 

 

 

1

 

 

 

4

 

 

 

 

1

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

3

 

 

 

 

 

 

 

 

 

 

4

 

 

 

 

2

1

 

 

 

 

1

 

 

1

 

4

12

1

 

2

 

18

 

 

2

 

3

3

1

1

 

 

2

18

5

 

 

 

 

 

 

 

 

 

 

1

 

 

3

4

2

 

 

1

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1

1

 

 

 

 

1

3

 

 

 

1

 

1

12

 

 

1

6

11

 

 

 

 

 

 

 

 

 

3

1

 

 

18

 

1

3

 

1

 

38

 

 

 

 

4

 

 

 

 

 

 

 

 

 

 

1

 

 

 

 

 

 

 

 

 

3

1

 

1

 

 

 

6

11

 

1

14

99

1

 

6

1

36

6

 

4

 

 

23

 

 

4

 

1

1

 

1

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

191

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:45 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East

Phase 1a

Phase 1b

Phase 2a

Phase 2b

ESA

ESC

ESD

ESA

ESC

ESD

ITS

ESA

ESB

ESC

ESD

ITS

Tell Rifa’at

8

 

 

13

 

 

 

13

 

 

 

 

Tenos

 

1

 

2

1

 

 

 

 

1

1

 

Thasos

 

 

 

 

 

 

 

 

 

 

 

 

Thespiae

 

 

 

 

 

 

1

 

 

 

 

 

Timna’

 

 

 

1

 

 

2

 

 

 

 

 

Troia/Ilion

 

 

 

3

2

 

1

 

2

8

 

 

Umm el-Tlel

 

 

 

 

 

 

 

 

 

 

 

 

Uruk

2

 

 

 

 

 

 

 

 

 

 

 

Veloukovo

 

 

 

 

 

 

 

 

 

1

 

 

Zeugma

 

 

 

 

 

 

 

 

 

 

 

 

Site

Appendix 4a. Table showing the absolute quantities, per phase and per site, of ITS, ESA, ESB, ESC, ESD and ARSW for the late Hellenistic-early Roman period (© Philip Bes/ICRATES Project).

192

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:45 UTC All use subject to https://about.jstor.org/terms

Appendices

Phase 4

Phase 3 ITS

ESA

ESB

ESC

ESD

ITS

ESA

ESB

ESC

ESD

ARSW

 

11

 

 

 

 

9

 

 

 

 

 

 

 

 

 

 

 

2

 

2

 

 

 

 

1

 

 

1

3

49

 

1

4

 

 

 

 

 

 

1

 

 

 

1

 

 

 

 

 

 

 

 

 

 

2

 

1

12

 

1

 

6

14

 

 

 

1

 

 

 

 

7

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1

 

 

 

 

 

 

 

 

5

 

 

 

 

3

 

 

 

 

193

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:45 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East

Site

Phase 5a ARSW

Phase 5b ESC ARSW

Phase 6 ARSW LRD

LRC

Phase 7 ARSW LRD

LRC

Phase 8 ARSW LRD

Phase 9 LRD

LRC

ARSW

LRC

Abdera

 

105

 á

1

 

 

 

 

 

 

 

 

 

 

 

Adulis

 

 

 

1

 

 

 

 

 

 

 

 

 

 

 

Aetolia

 

 

 

 

 

 

 

 

 

 

 

Alahan

 

 

Alexandria

 

Alexandria Troas

  1

3

Ammata

 

Amorion Amphipolis

2

2

 

 

 

2

 

 

2

 

 

4

 

1

1

 

2

10

10

6

 

15

17

2

13

8

9

17

15

3

 

 

 

 

 

 

1

 

 

2

1

 

 

 

 

3

 

1

2

5

6

1

1

20

2

5

12

 

 

 

 

 

 

 

 

 

 

 

 

 

1

 

 

252

 

 

 

 

 

 

 

 

 

 

 

 

 

Anemorion

 

1

1

6

5

3

4

14

5

9

11

6

12

29

2

Antioch

3

 

16

31

1

2

31

6

29

31

3

20

17

3

7

Apollonia

 

 

 

 

 

 

2

 

 

7

1

3

16

2

1

Argos

 

 

 

7

1

 

3

1

2

32

 

9

10

 

13

Assos

 

12

4

6

 

19

4

 

16

4

 

41

 

 

4

Athens

 

 

5

28

 

 

18

1

12

8

 

6

3

 

 

Athis/Neokaesareia

 

 

 

1

 

 

6

 

8

7

 

4

2

 

 

Attika

 

 

 

 

 

 

 

 

1

 

 

 

 

 

 

Ayios Philon

 

 

 

6

2

2

2

5

3

 

1

2

2

5

3

Berbati

 

1

 

5

 

 

5

 

1

 

 

1

 

 

 

Berenice

9

5

31

17

2

6

21

3

14

16

 

8

13

2

4

Boeotia (CN3)

 

 

 

 

 

 

 

 

 

1

 

 

 

 

 

Caesarea Maritima

 

1

3

6

17

5

8

26

24

1

8

8

6

17

5

Constantinople

 

 

 

3

1

9

9

1

13

29

1

28

20

15

20

Corinth

 

6

3

18

 

4

9

1

17

9

 

5

15

 

10

Cyrene

 

 

7

 

 

 

 

 

 

 

 

 

 

 

 

Déhès

 

 

 

 

 

 

 

 

1

1

 

 

 

 

3

Didyma

 

 

 

 

 

 

3

 

3

 

 

5

1

1

8

Diokaesareia

 

 

1

 

 

 

1

 

1

 

 

 

1

 

1

Doliche

 

 

 

 

 

 

 

 

 

2

 

4

2

 

6

Domuztepe

 

 

 

4

 

 

1

 

3

 

 

4

1

 

2

Dura Europos

 

 

4

 

 

 

 

 

 

 

 

 

 

 

 

El-Haditha

 

 

 

1

 

 

1

 

 

 

 

 

 

 

 

Emporio

 

4

 

 

 

1

 

 

46

6

 

93

40

15

258

‘En Boqeq

 

 

 

 

 

 

 

6

30

22

51

204

6

203

38

Ephesos

 

3

5

3

 

23

6

 

20

2

 

31

5

 

6

Epiphaneia

 

 

10

17

1

1

15

8

4

12

 

6

1

 

10

Gadara

 

 

1

 

 

 

 

 

 

 

 

 

1

 

 

Gindaros

 

 

1

 

 

1

2

4

4

3

 

7

1

 

8

Gortyna

 

16

8

71

1

22

38

2

97

36

 

73

52

6

35

Halabiyya/Zenobia

 

 

 

 

 

 

1

 

 

2

 

 

1

 

 

Halieis

 

 

 

 

 

 

1

1

1

 

 

6

4

3

2

Hammat Gader

 

 

 

 

 

 

1

3

1

 

 

11

 

 

3

Hippos-Sussita

 

 

 

3

4

4

4

12

3

2

2

8

4

10

7

Hyettos

 

2

2

14

 

6

3

 

4

2

 

2

1

 

 

Jalame

 

 

1

27

24

4

5

52

28

2

33

11

 

17

 

Jerusalem

 

 

 

12

1

3

11

2

10

10

1

14

7

 

4

Kallirhoe

 

 

 

4

 

 

 

 

1

 

 

 

 

 

 

Kellia

 

 

 

30

 

3

8

18

2

5

7

 

1

11

 

Kenchreai

 

1

 

3

 

 

6

 

6

5

 

7

3

 

 

Keos

 

 

 

4

 

 

2

 

4

1

 

4

2

 

4

Kepia

 

2

 

 

 

 

 

 

 

 

 

 

 

 

 

Khirbet al-Karak

 

 

 

 

 

 

2

4

3

2

3

7

 

6

12

Khirbet ed-Deir

 

 

 

 

 

 

 

 

 

 

 

 

 

1

 

194

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:45 UTC All use subject to https://about.jstor.org/terms

Appendices

Site

Phase 5a ARSW

Phase 5b ESC ARSW

Phase 6 ARSW LRD

LRC

Phase 7 ARSW LRD

LRC

Phase 8 ARSW LRD

Phase 9 LRD

LRC

ARSW

LRC

Knossos

1

36

14

13

 

3

1

1

 

3

 

5

2

1

3

Kommos

 

5

 

 

 

 

 

 

 

 

 

 

 

 

 

Kopetra

 

 

 

1

1

1

3

7

5

3

26

12

15

38

15

Koroneia

 

1

3

8

 

 

10

 

2

6

 

 

 

 

 

Kourion

 

 

 

 

1

 

1

3

4

 

1

 

 

2

 

Küçük Burnaz

 

 

2

2

 

 

1

 

1

 

 

2

2

 

 

Kythera

 

 

1

3

 

 

 

 

3

2

 

7

6

 

8

Labraunda

 

 

 

3

 

 

2

 

5

 

 

1

 

 

 

Magen

 

 

 

1

3

 

 

3

 

 

 

1

1

1

 

Maroni Petrera

 

 

 

1

2

 

 

3

4

 

 

 

 

8

2

Methana

 

1

 

3

 

3

3

 

5

10

3

9

1

3

10

Methymna

 

 

 

 

 

 

 

 

 

 

 

1

 

 

 

Nabratein

 

 

 

 

 

 

 

 

 

3

 

1

 

 

 

Nessana

 

 

 

 

 

 

1

 

2

8

2

1

 

3

 

Olympia

 

 

1

1

 

 

 

 

 

 

 

 

 

 

 

Oropos

 

1

 

 

 

1

 

 

1

1

 

 

 

 

 

Palaipaphos

 

 

 

2

7

3

 

19

8

 

7

7

1

16

9

Panayia Ematousa

 

 

 

 

1

 

1

3

 

 

1

1

5

3

9

Paphos

1

1

6

3

6

1

4

45

15

2

33

21

8

47

2

Pella

 

 

 

 

 

 

 

 

 

 

 

 

2

 

2

Pelusium

 

 

 

 

1

 

1

2

 

 

 

 

 

 

 

Perge

 

 

 

3

3

 

2

12

 

7

8

 

7

2

 

Petra

 

 

16

29

 

 

9

 

1

 

 

2

3

 

1

Philadelphia

 

 

 

 

 

 

 

 

 

 

 

1

 

 

 

Phlius

 

 

 

1

 

 

 

 

 

 

 

 

 

 

 

Porphyreon

 

 

 

1

1

2

2

1

5

1

2

4

2

3

6

Porsuk

 

 

 

1

 

 

 

 

 

 

 

 

 

1

 

Priene

 

 

 

 

 

 

 

 

1

 

 

 

 

 

 

Pylos

 

 

1

5

 

 

2

 

6

1

 

2

 

 

5

Ramet Rahel Rehovot-in-theNegev

 

 

 

1

 

 

1

 

 

 

 

1

 

1

 

 

 

 

1

 

 

 

 

3

1

 

1

5

4

 

Resafa

 

 

 

1

 

 

5

4

22

15

 

255

39

3

180

Sagalassos

 

 

 

 

 

 

 

 

 

2

 

 

1

 

 

Salamis/Constantia

 

 

 

 

 

 

 

 

 

 

 

 

6

4

 

Samaria-Sebaste

 

2

2

1

 

 

4

2

2

1

 

3

 

 

2

Samos

 

 

3

2

 

 

 

 

 

 

 

3

 

 

 

Samothrake

 

2

 

 

 

 

 

 

 

 

 

 

 

 

 

Sardis

 

1

 

4

 

 

1

 

10

1

 

8

4

 

 

Siphnos

 

1

3

2

 

 

 

 

2

 

 

1

 

 

 

Smintheion

1

5

1

1

 

1

 

 

3

 

 

4

 

 

1

Smyrna

 

 

 

1

 

15

5

 

27

 

 

6

 

 

 

Sountha/Sa’neh

 

 

1

2

 

 

 

 

 

 

 

 

 

 

 

Southern Sinai

 

 

 

 

 

 

2

 

 

 

 

1

3

1

1

Southwest Coast

 

 

 

2

 

 

 

 

1

 

 

4

2

 

4

Sparta

 

2

1

6

 

 

7

 

1

 

 

 

 

 

 

Stobi

1

 

9

18

 

10

19

 

8

4

1

4

3

 

1

Tanagra

 

14

10

35

 

7

20

 

19

24

 

17

22

 

 

Tarsos

 

 

2

13

 

 

10

2

13

4

 

10

1

 

 

Tel Mevorakh

 

 

 

 

 

 

 

 

1

 

 

 

 

 

 

Tell Fakhariyah

 

 

 

2

 

 

1

 

3

 

 

4

 

 

 

Tell Keisan

 

 

 

 

 

 

2

2

5

3

11

34

2

6

16

Tell Rifa’at

 

 

 

 

 

2

2

2

10

 

 

12

 

 

10

Tenos

 

 

1

 

 

 

 

 

 

 

 

 

 

 

 

195

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:45 UTC All use subject to https://about.jstor.org/terms

The Chronological and Geographical Distribution of Terra Sigillata and Red Slip Ware in the Roman East

Site

Phase 5a ARSW

Phase 5b ESC ARSW

Phase 6 ARSW LRD

LRC

Phase 7 ARSW LRD

LRC

Phase 8 ARSW LRD

Phase 9 LRD

LRC

ARSW

LRC

Thasos

 

87

2

11

 

11

9

 

33

5

 

26

4

 

7

Thespiae

 

 

2

16

 

1

7

 

6

5

 

1

2

 

 

Tocra/Taucheira

 

 

 

1

 

1

6

1

5

11

1

11

26

1

7

Troia/Ilion

 

3

1

 

 

 

 

 

 

 

 

 

 

 

 

Umm el-Tlel

 

 

 

 

 

 

 

 

1

 

 

 

 

 

 

Upper Zohar

 

 

 

 

 

 

9

 

4

7

8

27

34

9

12

Veloukovo

 

 

2

 

 

 

 

 

 

 

 

 

 

 

 

Vrokastro

 

3

 

1

 

1

 

 

3

 

 

10

2

 

10

Xanthos

 

1

2

 

2

 

1

2

1

 

6

8

 

1

1

Zeugma

 

 

5

47

 

 

16

 

 

5

 

 

17

 

 

Appendix 4b. Table showing the absolute quantities, per phase and per site, of ESC, ESD, ARSW, LRD and LRC for the mid- to late Roman period (© Philip Bes/ICRATES Project).

196

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:45 UTC All use subject to https://about.jstor.org/terms

This content downloaded from 80.180.50.243 on Sun, 29 Mar 2020 11:09:45 UTC All use subject to https://about.jstor.org/terms