The Excavations of Khirbet er-Rasm, Israel: The changing faces of the countryside 9781407307428, 9781407337357

Table of contents :
Front Cover
Title Page
Copyright
Dedication
Table of Contents
LIST OF FIGURES
LIST OF TABLES
LIST OF CONTRIBUTORS
0.1 Introduction
0.2 Geographical and Environmental Background
PART I: ARCHITECTURE AND STRATIGRAPHY
PART II: CERAMICS
PART III: OTHER REPORTS
PART IV: SYNTHESIS
PART V: CONCLUSIONS
ABBREVIATIONS
BIBLIOGRAPHY

Citation preview

BAR S2187 2011

The Excavations of Khirbet er-Rasm, Israel The changing faces of the countryside

FAUST & ERLICH

Avraham Faust Adi Erlich

THE EXCAVATIONS OF KHIRBET ER-RASM, ISRAEL

B A R

BAR International Series 2187 2011

The Excavations of Khirbet er-Rasm, Israel The changing faces of the countryside

Avraham Faust Adi Erlich with contributions by Oren Ackerman, Einat Armon-Ambar, Guy Bar-Oz, Daniella E. Bar-Yosef Mayer, Rachel Barkay, Elisabetta Boaretto, Deborah Cassuto, Anat Cohen-Weinberger, Yael Gorin-Rosen, Nili Liphschitz, Ofer Marder, Ravit Nenner-Soriano, Rinat Peshin, Jessie A. Pincus, Noa Raban-Gerstel, Débora Sandhaus, Avi Sasson, Izhak Shai, Inbal Shoam, Ehud Weiss and Yair Zoran

BAR International Series 2187 2011

Published in 2016 by BAR Publishing, Oxford BAR International Series 2187 The Excavations of Khirbet er-Rasm, Israel © The authors individually and the Publisher 2011 The authors' moral rights under the 1988 UK Copyright, Designs and Patents Act are hereby expressly asserted. All rights reserved. No part of this work may be copied, reproduced, stored, sold, distributed, scanned, saved in any form of digital format or transmitted in any form digitally, without the written permission of the Publisher.

ISBN 9781407307428 paperback ISBN 9781407337357 e-format DOI https://doi.org/10.30861/9781407307428 A catalogue record for this book is available from the British Library BAR Publishing is the trading name of British Archaeological Reports (Oxford) Ltd. British Archaeological Reports was first incorporated in 1974 to publish the BAR Series, International and British. In 1992 Hadrian Books Ltd became part of the BAR group. This volume was originally published by Archaeopress in conjunction with British Archaeological Reports (Oxford) Ltd / Hadrian Books Ltd, the Series principal publisher, in 2011. This present volume is published by BAR Publishing, 2016.

BAR PUBLISHING BAR titles are available from:

E MAIL P HONE F AX

BAR Publishing 122 Banbury Rd, Oxford, OX2 7BP, UK [email protected] +44 (0)1865 310431 +44 (0)1865 316916 www.barpublishing.com

The Khirbet er-Rasm expedition was greatly assisted by a generous donation in loving memory of Elaine Levin Bard by her children and family

To Iris Faust and Amitai Erlich

Table of Contents List of Figures

iii

List of Tables

viii

List of Contributors

ix

0.1 Introduction

1

0.1.1 Human Activity at Khirbet er-Rasm: A Chronological Table 0.1.2 Acknowledgments

0.2 Geographical and Environmental Background 0.2.1 Environmental Background: Conditions of the Area in General (Oren Ackerman)

Part I: Architecture and Stratigraphy

3 3 5 5 7

1.1 The Main Building 1.1.1 The Main Building: Architectural Phasing 1.1.2 The Main Building: Stratigraphy Other debris: 1.1.3 The Main Building: Architecture 1.2 Area C: The Stepped Structure – Architecture and Finds 1.3 Caves and Subterranean Complexes (Avi Sasson, Yair Zoran, Adi Erlich and Avraham Faust) 1.3.1 Caves D and G: Caves and a Hiding Burrow 1.3.2 Cave F: The Columbarium 1.3.3. Additional Subterranean Complexes 1.4 Appendices 1.4.1 List of Loci 1.4.2 Locus / Baskets List 1.4.3 List of Walls

9 9 22 23 38 52 66 66 76 83 92 92 96 98

Part II: Ceramics

101

2.1 The Iron Age and Persian Period Ceramic Finds (Itzhaq Shai) 2.2. The Pottery from Khirbet er-Rasm: The Typology of the Pottery from the Hellenistic, Roman and Later Periods (Débora Sandhaus) 2.2.1 The Hellenistic Pottery 2.2.2 The Pottery from Roman, Byzantine and Later Periods 2.3 Oil Lamps from Khirbet er-Rasm (Einat Armon-Ambar) 2.4 Dating the Hellenistic Ceramic Assemblage of Khirbet er-Rasm (Débora Sandhaus, Adi Erlich and Avraham Faust) 2.5 Petrographic Results (Anat Cohen-Weinberger) 2.6 Appendix: pottery database (at the back of the volume)

103 105 106 126 131 134 145

Part III: Other Reports

149

3.1 The Flint Collection (Ofer Marder) 3.2 The Coins (Rachel Barkay) 3.3 Stone Vessels (Rinat Peshin) 3.4 Small Finds 3.4.1 The Metal Artifacts (Ravit Nenner-Soriano) 3.4.2 Glass (Yael Gorin-Rosen) 3.4.3 Beads (Deborah Cassuto)

151 152 156 163 163 163 163

i

3.5 Faunal Remains (Inbal Shoam, Noa Raban-Gerstel and Guy Bar-Oz) 3.6 Plant Remains (Ehud Weiss) 3.6.1: Number of identified seeds of each taxon represented at Kh. er-Rasm (at the back of the volume). 3.6.2: Distribution and density of plant remains retrieved from Kh. er-Rasm (at the back of the volume). 3.7 Dendroarchaeological Investigations (Nili Liphschitz) 3.8 Shells (Daniella E. Bar-Yosef Mayer) 3.8.1 appendix: Shells uncovered at Kh. er-Rasm (at the back of the volume) 3.9 Radiocarbon Dating (Elisabetta Boaretto) 3.10 Geomorphological/Sedimentological Structure (Oren Ackerman) 3.11 Ground Penetrating Radar and Ground-Truth Results (Jessie Pincus Ben-Avraham) 3.12 Khirbet er-Rasm in the Ottoman and British Mandate Periods (Avi Sasson)

164 166

Part IV: Synthesis

191

4.0 Introduction 4.1 Nature of Settlements at Khirbet er-Rasm Prior to the Later Part of the Second Century BCE 4.1.1 Pre-Hellenistic Finds 4.1.2 The Hellenistic Period 4.1.3 Additional Finds from the Early Hellenistic Period: Subterranean Complexes 4.1.4 The Nature of the Site during the First Stages of the Hellenistic Period 4.1.5 The Foundation Date of the Main Building 4.1.6 Khirbet er-Rasm in the Iron Age Reconsidered 4.1.7 Summary: The Pre-Late Second Century History of the Main Building 4.2 Khirbet er-Rasm in the Late Second Century BCE 4.2.1 Dating the Destruction of the Site 4.2.2. Abandonment, Destruction and Collapse: Historical Events and Formation Processes at Khirbet er-Rasm 4.2.3 The Use of Space at Khirbet er-Rasm: The Function of the Main Building during the Late Second Century BCE 4.2.4 Khirbet er-Rasm during the Late Second Century BCE in Relation to Its Surroundings 4.2.5 Economy and Subsistence at Khirbet er-Rasm during the Second Century BCE 4.2.6 Summary 4.3 Khirbet er-Rasm after the Second Century BCE 4.3.1 Reuse and Later Occupation in the Early Roman Period 4.3.2 The Stepped Stone Structure (Area C): Function and Date Area C: A Summary 4.3.3 Agricultural Activity, Firing Zone, and a Park: The Site to the Present

193 193 193 193 196 197 202 205 207 208 208

Part V: Conclusions

245

5.1 The Hasmonean Policy toward the Gentile Population: The View from Khirbet er-Rasm 5.1.1 Khirbet er-Rasm as an “Idumaean”, Non-Jewish Site 5.1.2 Khirbet er-Rasm as Representative of the Situation in the Non-Urban Sector 5.1.3 Rural Settlement in Judea in the Second Century BCE 5.1.4 Discussion and Conclusions 5.2 A Brief History of the Site from the First Settlement to the Creation of the British Park 5.3 Endnote: The Importance of the Excavations at Khirbet er-Rasm

247 247 251 254 255 256 257

Abbreviations

259

Bibliography

260 ii

167 167 167 168 169 173 178 185

209 216 227 230 231 232 232 233 240 242

LIST OF FIGURES 0.1 Location map of Khirbet er-Rasm (prepared by Silvia Krapiwko) 1.1 Plan of main building 1.2 Phases 1 and 1a 1.3 The entrance to the main building, facing north 1.4 W.21 (Square J52), facing west 1.5 Phases 2a and 2b 1.6 W.41 (Square E49), facing east 1.7 Phases 2c and 3 1.8 W.36 (Square E54), kirton installation in room 1, facing northwest 1.9 W.43 (Square E54), tabun in room 1, facing east 1.10 W.6 (Square G53), eastern wall of room 2, facing south 1.11 Phase 4 1.12 View of the main building with the ramp leading to the entrance, facing north-northwest 1.13 Ramp leading to the entrance of the main building, facing east 1.14 W.4, row of monoliths, facing north 1.15 W.15 (Square J52), kirton wall, facing south 1.16 Collapsed cave in Square I49, facing north 1.17 Collapsed cave in courtyard, facing north 1.18 The function of the building during phase 2C 1.19 West section of Square I53 1.20 North section of Square I52, showing upper and lower occupational debris 1.21 Top of wall fall in Square J53 facing south 1.22 Top of wall fall in Square I48 facing south 1.23 Part of east section of Square I48 during excavation 1.24 South and east sections of Square I48 after excavation, facing southeast 1.25 South section of Square G53 1.26 Square G53 facing east at the beginning of excavation 1.27 South section of Square G53 1.28 Square E54 facing north at the beginning of excavation 1.29 Square E54 after excavation, facing north 1.30 Squares D51, E51 and F51 facing south 1.31 Squares G48 and H48 at the beginning of excavation, facing west 1.32 Top of wall fall in Square G48 facing east 1.33 Ceramic vessels in situ in Square G48 1.34 Top of wall fall in Square D49 facing south 1.35 West section of E49 1.36 North section of E49 1.37 Top of wall fall in Square E49 facing east 1.38 North section of Square E49 1.39 General view of Square E49 after excavation, facing west 1.40 Top of wall fall in Square F49 facing west 1.41 W.8 (Square E54), facing north 1.42 Plan of main building with room numbers 1.43 General view of the western part of the main building, facing north-northwest 1.44 General view of the western part of the main building, facing north 1.45 General view of the eastern part of the main building, facing north

iii

x 9 10 11 12 13 14 15 16 16 17 18 19 20 20 21 21 22 23 24 25 26 27 27 28 28 29 29 30 30 31 31 32 32 33 33 34 34 35 35 36 36 37 38 39 39

1.46 General view of the eastern part of the main building, facing south 1.47 General view of the western part of the main building, facing north 1.48 General view of the northern part of the main building, facing west 1.49 General view of the northwest part of the main building, facing west 1.50 Rooms 1 and 2 facing north 1.51 W5 (Square F53) facing west 1.52 W5 (Square F53) facing north 1.53 Rooms 2 and 3, facing northwest 1.54 Rooms 3, 4 and 5 facing northwest 1.55 Rooms 6, 7, 8 and 9 facing northwest 1.56 Room 6 facing east 1.57 Room 7 facing east 1.58 Room 9 facing north 1.59 W.15 and W.17 in room 9 facing north 1.60 Room 14 facing west 1.61 Room 15 facing north 1.62 Room 17 facing south 1.63 Cell 18 facing south 1.64 Courtyard 20 facing north 1.65 Plan of the main building and Area C 1.66 The main building and Area C, facing east 1.67 The structure at Area C as viewed from the main building, facing south-southwest 1.68 Plan, section and elevation of Area C 1.69 Area C facing west-southwest 1.70 Area C facing west 1.71 Area C facing north 1.72 Area C facing southwest 1.73 Area C, north wall of structure facing south 1.74 Area C facing east 1.75 Area C facing east 1.76 West wall of structure at Area C facing southeast 1.77 Section in the structure at Area C, facing east-southeast 1.78 Section in the structure at Area C, facing east 1.79 The structure at Area C before excavation facing north 1.80 South wall of structure at Area C facing northwest 1.81 South wall of structure at Area C and collapse during excavation, facing north 1.82 South wall of structure at Area C after excavation, facing north 1.83 The structure at Area C facing west 1.84 The structure at Area C facing west 1.85 The structure at Area C facing west 1.86 Worked stone near the structure in Area C 1.87 Worked stone near the structure in Area C 1.88 Worked stone near the structure in Area C 1.89 Plan of the site with location of caves 1.90 Aerial view of the site facing north, with location of some of the caves 1.91 Plan of Caves G and D and the burrow connecting them 1.92 Section of Caves G and D and the burrow connecting them 1.93 General view of Cave G facing southwest

iv

40 41 42 43 43 44 44 45 45 46 46 47 47 48 49 50 51 51 52 53 54 54 55 56 57 57 58 58 59 59 60 60 60 61 61 62 62 63 64 64 65 65 65 67 68 69 70 70

1.94 Entrance to the main hall of Cave G 1.95 Quarrying marks on ceiling of Cave G 1.96 Quarrying marks made with a shovel in Cave G 1.97 Arched quarrying marks in Cave G 1.98 Drawing of arched quarrying marks in Cave G 1.99 Drilled holes in the eastern wall in Cave G 1.100 Drawing of drilled holes in the eastern wall in Cave G 1.101 Grooves and drilled holes in the eastern wall of Cave G 1.102 Drawing and section of grooves and drilled holes in the eastern wall of Cave G 1.103 Graffiti on the façade wall of Cave G 1.104 Drawing of graffiti on the façade wall of Cave G 1.105 The entrance to Cave D facing northwest 1.106 The entrance to the hiding burrow from Cave G 1.107 The entrance to the hiding burrow from Cave G 1.108 The entrance to the hiding burrow from Cave D 1.109 Interior of hiding burrow 1.110 Plan and section of Cave F 1.111 Cave F facing north 1.112 Built wall at the entrance of Cave F, facing south 1.113 Pillars in Cave F 1.114 Western pillar in Cave F 1.115 Kirton block in Cave F 1.116 South east pillar in Cave F 1.117 Arcosolium-like niche in the western wall of Cave F 1.118 Arcosolium-like niche in the western wall of Cave F 1.119 Small niche in the western wall of Cave F 1.120 Columbarium niches in Cave F 1.121 Quarrying marks made with a thick chisel on the eastern wall of Cave F 1.122 Drawing of quarrying marks made with a thick chisel on the eastern wall of Cave F 1.123 Quarrying marks made with a thick chisel on the eastern wall of Cave F 1.124 Drawing of quarrying marks made with a thick chisel on the eastern wall of Cave F 1.125 Fissure on the eastern wall of Cave F 1.126 Graffito 1 on the western wall of Cave F 1.127 Drawing of graffito 1 on western wall of Cave F 1.128 Graffito 2 on western wall of Cave F 1.129 Clay plates on the floor of Cave F 1.130 Axis of Cave H underneath the main building 1.131 Western entrance to Cave H (west of the main building) 1.132 Collapse of Cave H in courtyard 20, facing west 1.133 Collapse of Cave H east of the main building (east of Square D51) 1.134 Entrance to Cave I facing south 1.135 Plan and section of Cave I 1.136 General view of Cave I 1.137 A niche in the southern wall of Cave I 1.138 Small burrow in the southwestern wall of Cave I 1.139 Opening of cistern J 1.140 Plan and sections of cistern J 1.141 Cistern J from inside

v

70 70 70 71 71 72 72 72 73 73 73 74 74 75 75 76 77 78 78 78 78 78 79 79 79 79 79 79 80 81 81 82 82 82 83 83 84 84 85 85 86 87 88 88 89 89 90 90

1.142 Troughs near the entrance to cistern J 1.143 Plan and section of troughs near the entrance to cistern J 2.1: Plate of Iron Age and Persian Period pottery 2.2: Plate of Utility Vessels – Storage Jars 2.3: Plate of Utility Vessels – Storage Jugs and Mortaria 2.4: Plate of Kitchen Ware 2.5: Plate of Table Ware – Pouring and Serving vessels 2.6: Plate of Personal Table Ware – Bowls and Plates 2.7: Plate of Personal Vessels 2.8: Plate of Diagnostic Sherds from Later Periods 2.9: Plate of Lamps 2.10: Plate of Assemblages – Storage Jars, Pouring and Serving Vessels 2.11: Plate of Assemblages – Kitchen Ware 2.12: Plate of Assemblages – Bowls and Plates 2.13: Plate of Assemblages – Personal Vessels and Lamps 2.14 Complete vessels from the site 3.1 Flint artifacts 3.2 Coins 3.3 Stone vessels 3.4 Metal artifacts 3.5 Beads 3.6 Probability distribution of the radiocarbon dates 3.7 Probability distribution of the average for the samples RTT 5227 and 5229 3.8 Plan of the site with location of Areas Z1 and Z2 3.9 Section in Z1 3.10 Section in Z2 3.11 Eastern section in Z1 facing south 3.12 Eastern section in Z1 facing west 3.13 Western section in Z1 facing south 3.14 Western section in Z1 facing north 3.15 Upper probe at Z2 facing west 3.16 Map of the GPR study areas 3.17 GPR survey study area 1 (Area E before excavation), facing west 3.18 Overview of the features seen in the GPR data from study area 1 3.19 Area E (GPR study area 1 after excavation, facing west) 3.20 Area E (GPR study area 1 after excavation, facing west) 3.21 Cone geometry and the “footprint” radius at different depths. GPR waves propagate into the sub-surface in a cone shape (after Conyers and Goodman 1997: 36; courtesy of AltaMira Press) 3.22 Cross-section diagram of a buried surface showing the scattering and focusing of signal data at different antenna locations (after Conyers and Goodman 1997: 54; Courtesy of AltaMira Press) 3.23 GPR area study 1 – Square I48 3.24 Map of the area 1945 (1:20,000) (2007 © All Rights Reserved by the Survey of Israel; Survey of Israel Persmission) 3.25 Map of the area 1950 (1:20,000) (2007 © All Rights Reserved by the Survey of Israel; Survey of Israel Persmission) 3.26 Map of the area 1959 (1:20,000) (2007 © All Rights Reserved by the Survey of Israel; Survey of Israel Persmission)

vi

90 91 104 108 110 113 118 121 125 128 132 137 138 139 140 141 151 152 158 162 164 170 171 172 174 174 175 176 176 176 176 180 181 181 182 182

183

184 185 186 187 187

3.27 Area of Kh. er-Rasm aerial photo from 1956 (2007 © All Rights Reserved by the Survey of Israel; Survey of Israel Persmission) 3.28 View from Kh. er-Rasm toward Moshav Agur 4.1 Phases 0, 1 and 1a of the main building 4.2 Phases 2a and 2b of the main building 4.3 Phases 2c and 3 of the main building 4.4 Oil press weight in Square H49 4.5 Oil press in complex 44 from Maresha with weights in situ (after Kloner and Sagiv 2003:61; courtesy of Prof. Amos Kloner and the Israel Antiquities Authority) 4.6 Plan of structure at Kh. Abu et-Twein (after Mazar 1981: 231; courtesy of Prof. Amihai Mazar and the Israel Exploration Society) 4.7 Plan of structure at Deir Baghl (after Kochavi 1972: 50; courtesy of the Israel Antiquities Authority) 4.8 Plan of structure at Aderet (after Yogev 1982: 82; courtesy of the Israel Antiquities Authority) 4.9 Plan of structure at Tirat Yehuda (after Yeivin and Edelstein 1970: 57; courtesy of the Israel Antiquities Authority) 4.10 Plan of structure at Mazor (after Amit and Zilberbod 1998: fig. 117; courtesy of the Israel Antiquities Authority) 4.11 Isometric reconstruction of the main building at Kh. er-Rasm 4.12 Lamp in situ 4.13 Storage jar in situ 4.14 Pottery in situ 4.15 Blockage of entrance to room 4 4.16 Blockage of passage between rooms 13 and 14 4.17 Blocked house from Ethiopia 4.18 Ash layer in Area E (Squares E49, F49) 4.19 Storage jar in Square H49 4.20 Map of distribution of complete and nearly complete vessels 4.21 Plan of Horvat Eqed (after Gichon 1993: 416; courtesy of the Israel Exploration Society, and Prof. Moshe Fisher) 4.22 Plan of fortress at Beth-Zur (after Funk 1993: 261; courtesy of the Israel Exploration Society) 4.23 Aerial view of Kh. er-Rasm, looking east toward the Elah Valley 4.24 Map of ancient roads during the Hellenistic period 4.25 Aerial view of Kh. er-Rasm, looking east toward the Elah Valley 4.26 Sites with Hellenistic remains in the vicinity of Kh. er-Rasm (based on the data provided by Dagan's Shephelah survey) 4.27 Modern-day olive trees in Kh. er-Rasm 4.28 Stone vessels from room 8 4.29 Plan and section of tower from Kh. er-Ras (courtesy of Prof. Shimon Dar) 4.30 Nefesh structure from Suweida in Syria (after Brünnow and Domaszewski 2004 [1909]:97–100 4.31 Necropolis of Mampis with pyramidal nefesh (after Negev 1993: 892; courtesy of the Israel Exploration Society) 4.32 Nefesh structure at Kh. Midras 4.33 Graffito of nefesh from Kh. El-‘Ein (after Zissu 2005: 32; courtesy of the Israel Antiquities Authority) 4.34 Isometric reconstruction of the nefesh at Kh. er-Rasm 4.35 Plough marks on W.41 5.1 Map of Idumea during the Hellenistic period 5.2 Map of the Land of Israel with sites mentioned in discussion

vii

189 189 198 199 200 201 202 203 204 205 206 206 207 209 209 209 210 210 213 214 215 218 222 223 223 224 226 229 230 232 233 234 236 236 238 241 242 248 251

List of Tables Table 1.1 Indicative Ceramic Found in Topsoil Loci Mixed with Collapse on Top of the Structure 65 Table 1.2 Ceramics from the Sections Surrounding the Structure 66 Table 2.1: Open Lamps (Saucer Lamps/Pinched Lamps) 135 Table 2.2: Closed Wheel-made Lamps 135 Table 2.3: Closed Mould-made Lamp 136 Table 2.4: Unidentifiable Lamp Fragments 136 Table 2.5: The Dating of the Complete Vessels of the Hellenistic Period 142 Table 2.6: The Pottery from Room 14, L.823 (Ash layer; “clean,” below the wall fall) 143 Table 2.7: The Pottery from Room 14, L.821 (below the wall fall; occupational debris/ash layer) 143 Table 2.8: The Pottery from Room 14, L.813 and L.814 (Kirton Wall Fall) 144 Table 2.9: The Pottery from Room 15, L.824 (ash layer/floor) 144 Table 2.10: The Pottery from Room 15, L.818 (below wall fall/occupational debris mixed in ash layer) 144 Table 2.11: The Pottery from Room 15, L. 810 (Kirton Wall Fall) 145 Table 2.12: Petrographic Results 147 Table 3.1. Waste Material Frequencies 151 Table 3.2: Stone Objects 157 Table 3.3: The Beads 164 Table 3.4: Distribution of Faunal Remains According to Area, Locus and Basket 165 Table 3.5: Number of Identified Specimen (NISP) and Minimum Number of Individuals (MNI) of Each Taxon Represented at Kh. er-Rasm Cave (Area D) 165 Table 3.6: Number of Identified Specimen (NISP) and Minimum Number of Individuals (MNI) of Each Taxon Represented at Kh. er-Rasm (Areas A, B and E) 166 Table 3.7: Location of tree species at Kh. er-Rasm 168 Table 3.8: Sample Details and Radiocarbon Results 169 Table 3.9: Location and context of the sediments analyzed with FTIR 170 Table 3.10: Layers in Square F49, North Side 173 Table 3.11: Section in Terrace Z1W, Upper Probe 177 Table 3.12: Section in Terrace Z1E, Lower Probe 177 Table 3.13: Section in Terrace Z2U, Upper Probe 177 Table 3.14: Section in Terrace Z2L, Lower Probe 177 Table 3.15: The Demographic Development of the Village of Agur 188

viii

List of Contributors Oren Ackerman, The Institute of Archaeology, The Martin (Szusz) department of Land of Israel Studies and Archaeology, Bar Ilan University, Ramat Gan, 52900, Israel Einat Armon-Ambar, Hecht Museum, University of Haifa, Haifa, 31905, Israel. Guy Bar-Oz, Laboratory of Archaeozoology, Zinman Institute of Archaeology, University of Haifa, Haifa, 31905, Israel. Daniella E. Bar-Yosef Mayer, Zinman Institute of Archaeology, University of Haifa, Haifa, 31905, Israel Rachel Barkay, The Bank of Israel, 91007, Jerusalem, Israel. Elisabetta Boaretto, Radiocarbon and Cosmogenic Isotopes Laboratory, Kimmel Center for Archaeological Science, Weizmann Institute of Science, Rehovot, 76100, and the Martin (Szusz) department of Land of Israel Studies and Archaeology, Bar Ilan University, Ramat Gan, 52900, Israel. Deborah Cassuto, The Institute of Archaeology, The Martin (Szusz) department of Land of Israel Studies and Archaeology, Bar Ilan University, Ramat Gan, 52900, Israel. Anat Cohen-Weinberger, The Israel Antiquities Authority, P.O.Box 586, Jerusalem, 91004, Israel. Yael Gorin-Rosen, The Israel Antiquities Authority, P.O.Box 586, Jerusalem, 91004, Israel. Nili Liphschitz, Institute of Archaeology – The Botanical Laboratories, Tel Aviv University, Tel Aviv, 69978, Israel. Ofer Marder, The Israel Antiquities Authority, P.O.Box 586, Jerusalem, 91004, Israel. Ravit Nenner-Soriano, The Institute of Archaeology, The Hebrew University of Jerusalem, Jerusalem, 91905, Israel Rinat Peshin, The Institute of Archaeology, The Martin (Szusz) department of Land of Israel Studies and Archaeology, Bar Ilan University, Ramat Gan, 52900, Israel Jessie A. Pincus, Mnemotrix Systems, Inc. P.O.B. 3030 Ra‛anana, 43662, Israel Noa Raban-Gerstel, Laboratory of Archaeozoology, Zinman Institute of Archaeology, University of Haifa, Haifa, 31905, Israel Débora Sandhaus, The Institute of Archaeology, The Hebrew University of Jerusalem, Jerusalem, 91905, Israel and The Israel Antiquities Authority, P.O.Box 586, Jerusalem, 91004, Israel. Avi Sasson, Ashkelon Academic College, Ashkelon, Isarel. Izhak Shai, the Martin (Szusz) department of Land of Israel Studies and Archaeology, Bar Ilan University, Ramat Gan, 52900, Israel Inbal Shoam, Laboratory of Archaeozoology, Zinman Institute of Archaeology, University of Haifa, Haifa, 31905, Israel Ehud Weiss, Archaeobotanical laboratory, Kimmel Center for Archaeological Science, Weizmann Institute of Science, Rehovot, 76100, and the Institute of Archaeology, The Martin (Szusz) Department of Land of Israel Studies and Archaeology, Bar-Ilan University, Ramat Gan, 52900, Israel Yair Zoran, Nehusha, 99833, Israel

ix

0.1 Location map of Khirbet er-Rasm (prepared by Silvia Krapiwko)

x

Avraham Faust and Adi Erlich

0.1 Introduction

including a short chapter on the Iron Age and Persian period ceramic finds (by Itzhaq Shai), a long discussion of the Hellenistic and later pottery forms (by Débora Sandhaus), a chapter on the oil lamps unearthed at the site (by Einat Ambar-Armon), a more integrative discussion of Hellenistic period assemblages (by Débora Sandhaus, Adi Erlich and Avraham Faust), and a report of the petrographical analysis (by Anat Cohen-Weinberger).

Kh. er-Rasm (R.P. 14351219) is a small site in the upper Shephelah, about 1 km south-southwest of Tel ‘Azekah. The remains include mainly a concentration of ruins on top of the hill, where many walls were visible before the beginning of the excavations, including a row of still standing monoliths. This concentration is surrounded with remains of additional walls, a heap of stones, terraces and caves of various sorts. The site is small and rural in nature, and is not identified with any known historical sites (as indicated also by its name, see Sasson, Chapter 3.12), and this was in part the reason for its exploration (below).

The third part will include all the additional reports, i.e., flints (by Ofer Marder), coins (by Rachel Barkay), stone vessels (by Rinat Peshin), metal objects (by Ravit Nenner-Soriano), glass (by Yael Gorin-Rosen), beads (by Deborah Cassuto), bones (by Inbal Shoam, Noa RabanGerstel and Guy Bar-Oz), botanical finds (by Ehud Weiss), charcoals/wood (by Nili Liphschitz), shells (by Daniella E. Bar-Yosef Mayer), carbon 14 (by Elisabetta Boaretto), geomorphology (by Oren Ackerman), the result of the ground penetrating radar (GPR) work at the site (by Jessie Pincus Ben-Avraham), as well as a brief discussion of the function of the site in the Ottoman period and under the British Mandate (by Avi Sasson).

The site was excavated and surveyed in the years 1997–2003. Excavations concentrated on the main area of the ruins (where three areas of excavations, A, B and E, were opened), henceforth the main building. Additional excavation areas were opened on top of structural remains to the south of the main building (Area C), which turned out to be a pyramid-like stone structure, and in one of the caves, which included a small hiding burrow (Area D). The rest of the area, and especially the caves, were intensively mapped and surveyed (though very little pottery was collected during this mapping operation).

The fourth part will include a detailed discussion of the history of the site, based on the above. The section will be divided between discussion of pre-second century, late second century, and post-second century finds. This part will generally progress in chronological order, but will also include the integrative analysis, regarding, for example, the dating of the various phases, the function of the site during the different stages, the economy of each settlement phase, etc., as well as discussion of the site formation processes in relation to abandonment, destruction and collapse, as reflected by the finds at the site.

Kh. er-Rasm was first settled during the Chalcolithic period, but remains from this period are meagre. The site was then resettled during the late Iron Age I and / or early Iron Age II, but these remains are also very poor, and do not include any architecture (perhaps one wall). More significant remains were dated to the late Iron Age II, and some finds are attributed to the Persian period, but the main period of occupation at the site dates from the early Hellenistic period up to the late second century BCE. The vast majority of the finds at Kh. er-Rasm are dated to the late second century BCE, as this is the time when the site was destroyed, and this is the period for which we have the most data. Some reoccupation took place in the Early Roman period, and from then on the site was abandoned and was used by farmers and herders. During the early years of the State of Israel the site was used as a firing zone, and later on was turned into part of the British Park, where the site is located today.

The fifth part of the monograph will include a historical discussion of Hasmonean policy in the late second century in light of the findings at Kh. er-Rasm, followed by a summary of the history of the site. Since we found the analysis to be integrative, we have decided to combine the bibliography and publish only one list of references at the end of the book, and not, as is sometimes done, at the end of each chapter.

Background: History of Research and Methods of Excavations

Structure of the Report The present report represents the results of the excavations and the analysis of the finds.

In 1996, as an outgrowth of his MA thesis on The Rural Settlement in the Land of Israel during the Period of the Monarchy, Avraham Faust began looking for a small Iron Age II rural site that could be excavated, in order to shed light on this neglected settlement sector. After consultation with teachers and colleagues, including Prof. Shlomo Bunimovitz, Dr. Yehuda Dagan, Prof. Zeev Safrai and Prof. Amos Kloner, the site that was chosen was Kh. er-Rasm – a small site about one km south-southwest of Tel ‘Azekah. The survey conducted by Dagan, who discovered the site, produced relatively large quantities of Iron Age pottery, and a row of monoliths that was still standing (as it still

Following this introduction, the first part of the report will summarize the results of the excavations in the various areas and the outcomes of the survey of the caves (jointly with Avi Sasson and Yair Zoran), with detailed discussion of the architecture and stratigraphy (although without any further elaboration, i.e., no discussion of parallels, dating, etc., since this is only possible after the additional finds are discussed). The second part will include all the ceramic analysis,

1

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

is today), all leading us to view the site as appropriate for investigating the Iron Age II rural settlement (although there were finds from other periods as well, including the Hellenistic, Roman and Byzantine periods; Dagan 1992b:131, site 100).

participation of about 50 students from BIU and a few volunteers. Excavations continued in Areas A, B and C, and a new area (Area D) was opened in one of the caves, including a hiding burrow. The final season of excavation was carried out during 2003 (G27/2003). This season lasted three weeks (with about 30 students each week), and was the largest season as it served as BIU’s field school for this year. During this season excavations continued in Areas A, B and C, and a new area was opened in the northern part of the main building – Area E. We estimate that this season accounted for over 50% of the entire excavations, and hence its importance.

The actual excavations at Kh. er-Rasm materialized in 1997, as part of a joint project of the Martin (Szusz) Department of the Land of Israel Studies and Archaeology at Bar-Ilan University (BIU) and the Jewish National Fund (JNF). The larger project, which included several excavations, was initiated by Prof. Zeev Safrai of BIU and Eli Shenhav of the JNF, and was intended to allow youth who come to the JNF youth centers to work in an archaeological dig. Three excavation projects were planned, one in the south, one in the center and one in the north. Kh. er-Rasm was suggested by Avraham Faust as an appropriate site for the central district of the JNF, and the suggestion was accepted. Prior to the excavations we began an architectural survey of the walls, many of which were easily observable without excavations.

Throughout the years, the excavations were accompanied by a survey of the entire site, and all the caves and additional features were located and mapped. We should note, however, that the survey was primarily architectural. Pottery was collected in the excavations area, and only rarely from the surface. Hence the survey finds, as discussed in some parts of this monograph, are very limited in number. There were several changes in the excavation’s sampling procedure over the years, and these must be acknowledged. Since during the first season of excavations progress was very slow, and to a large extent we excavated topsoil, sieving was done only randomly (usually 1 out of 10 buckets). In the main seasons, however, when we had reached occupational debris, we attempted to reach total sieving in those layers. In the final season of 2003, we attempted 100% sieving, also within the heavy wall-falls.

The first short season of excavation (about one week in June and an extra day in September) took place during 1997 (G57/1997). During this season we opened only a few squares in Area A (on the western part of the main building). Since we were working mainly with youth, progress was very slow, and only in half a square (working with a few students) did we penetrate below the upper level of what we later understood was a massive wall fall. Here we found insitu Hellenistic pottery. At the same time Adi Erlich became a co-director of the dig. Previously she took part in other projects in the Judean Shephelah, flanking Kh. er-Rasm from the north (Tel Beth-Shemesh) and south (Maresha), therefore she was highly interested in excavating another, still smaller site in that region. Although Erlich agreed to join the dig before the discovery of the Hellenistic pottery, the latter turned out to be the major occupational phase in the main building, which increased her interest in the Kh. er-Rasm project.

Similar changes took place in other aspects of the work as well. For example, during most seasons we kept very few soil samples for later analysis (flotation and chemistry), in 2003 we kept at least two buckets from every locus for flotation (all of which were examined) and soil analysis. While the unequal distribution of the sampling procedure over the years might pose a problem for spatial analysis, the fact that the 2003 season was probably as large as all other seasons combined (or even larger) made up for this deficiency, and allowed us to use the findings for spatial analysis.

The following seasons were also short, and progress was very slow since most of the workers were youth, and it was imperative not to penetrate too deep without proper control. The 1998 season (G155/1998), took place during the second week of October, and the 1999 season (G131/1999) took place in September (again, for about a week).

Needless to say, throughout all seasons all the finds were collected and kept, including every piece of bone, shell, charcoal, seed, non-local stones (including flints), etc., even if later proven too small for analysis (e.g., tiny bone fragments) or indistinct (e.g., most of the flint fragments).

This gradually changed. The 2000 season (G50/2000), while also short (one week in August) was carried out with a relatively larger number of students (about 30) from BIU, and a few volunteers. Excavations at Area A expanded, and we opened two additional excavation areas (Area B, in the southeast corner of the building, and Area C, south of the main building, in what turned out to be a pyramid-like building).

It turned out that although some human activity took place at the site during the Iron Age, and it is likely that the central building was originally erected at the time, the site is dated mainly to the Hellenistic period. While this should caution us against over-reliance on the data from surveys, it does not impact the importance of the dig. Planned excavations of a small site in the countryside can produce a wealth of information, and can compensate for the urban bias of the tel minded Near Eastern archaeology. Most of the site at Kh. er-Rasm is well preserved, and we hope that the slow

The 2001 season (G96/2001) was longer and larger than previous ones (probably the same scope as all the previous seasons together). The season lasted two weeks, with the

2

Avraham Faust and Adi Erlich

0.1.2 Acknowledgments

and careful excavations allowed us to extract the most of the available information. We strongly believe that the data included in this report are of great importance to the study of the Hellenistic period in southern Israel.

Over the years we have benefited greatly from the help of many young people in the context of JNF National Service and the JNF Trailblazers, as well as students from Bar-Ilan University and archaeologists, who have participated in the dig. We would like to thank them all.

Moreover, given the usual treatment of small rural sites in the recent scholarship, when they are simply “counted,” with the size estimated very roughly and then “multiplied” by a certain coefficient in order to “calculate” the number of inhabitants, we find the results of the excavations at Kh. er-Rasm to be very revealing, as they show that small sites in the countryside change their character over the years, and hence caution us against simply treating all such sites as similar.

Special thanks are due to the students and the professionals who worked with us over the years and were part of the Kh. er-Rasm expedition: Naomi Akiva, Einat Armon, Tehila Atkins, Mechael Azband, Eyal Baruch, Rishona Fine, Michal Gamarsani, Uzi Leibner, Michal Meishar, Anat Michaeli, Rachel Sagir, Itzhaq Shai, Rinat Shaki, Rotem Shelef, Aran Yardeny and Yair Zoran served as area supervisors and area supervisors’ assistants. Help in registration and documentation was given by Tamar Almog, Yifat Ast, Deborah Cassuto Eti Sa‛ada, Yael Suweid and Mi Yeong-Im. Administration was handled by Uri Rice. Surveying and measurements were conducted by Dan Behar, Jay Rosenberg, Roni Saban, Yehuda Shapira and Yair Zoran. Help with the metal detector was provided by Moshe and Yuval Lopan and Rami Chen. Pottery restoration was carried out by Dina Castel. Yulia Rodman drew the pottery. Ravit Nenner-Soriano prepared the plates. Some of the people who analyzed the finds (below) also participated in the field work, including Avi Sasson and Oren Ackerman. Much of the lab work was done by Debi Cassuto, and later by Pirchia Eyal. Some of the plans were prepared or modified by Michal Klein,

The excavation of small sites can provide a wealth of information about various aspects of life, including “grand” history (see Chapter 5.1), and most important – it can teach us the history of the people who are otherwise left out. Given the overemphasis on large sites and impressive buildings, we feel that excavations of smaller sites have a greater potential to contribute to our knowledge, as each excavation provides significant additional data about this relatively unknown settlement sector, whereas the excavation of larger sites simply adds a little to our already large database. We hope that the present report will support this claim.

0.1.1 Human Activity at Khirbet er-Rasm: A Chronological Table Period Chalcolithic Iron Age I\Iron Age II Late Iron Age II Persian period Early Hellenistic period

Finds Flints (and pottery in Dagan’s survey) – very limited amounts Pottery (and perhaps a wall) – limited amount Pottery and architecture – limited amount Pottery (and probably usage in the building) – limited amount Pottery, architecture and hewing of caves

Architectural Phase (main building)

Comments

---------

Limited settlement

Phase 0

Limited settlement

Phases 1, 1a

Founding of the main building Limited activity

Phases 2a, 2b

Second half of 2nd century BCE

Pottery and architecture

Phase 2c

End of 2nd century BCE

Pottery and architecture

Phase 3

Early Roman period

Pottery and architecture

Phase 4

Late Roman and onward

Scattered finds

----------

3

Reusing the main building (an estate?) After the collapse of the cave system. Reuse of the building The blocking of doorways and abandonment of the building Limited reoccupation, and construction of the pyramidal structure Occasional usage by shepherds and farmers

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

who also prepared most of the maps. Graphic assistance was provided by Silvia Krapiwko and Einat Ehrlich. The isometric reconstructions were done by Rachel Bordowicz, and the English was edited by Miriam Schlusseberg and Aviva Levine.

this report and commenting on it. Special thanks are due to Dr Shimon Cooper of the Sociology and Anthropology department at Bar-Ilan University. We would also like to thank Eli Shenhav, the JNF archaeologist, for his immense support in initiating the project and during the first season of excavations. The JNF staff was very helpful during all the stages of the excavations, and we would like to mention Meir Malka, Dani Mador, Yigal Sitri, Meir Cohen and Iris Bernstein. Special thanks are due to Eli Ben-Shitrit, the warden of the British Park, for his assistance.

The diverse finds were analyzed by various experts, whose reports are included in this monograph. This include Débora Sandhaus (Hellenistic pottery and later pottery), Einat Armon-Ambar (oil lamps), Itzhaq Shai (Iron Age Pottery), Anat Cohen-Weinberger (petrography), Ofer Marder (flints), Rachel Barkay (coins), Rinat Peshin (stone vessels), Inbal Shoam, Noa Raban-Gerstel and Guy Bar-Oz (faunal remains), Ehud Weiss (botanical remains), Nili Liphschitz (wood), Daniella E. Bar-Yosef Mayer (shells), Elisabetta Boaretto (Carbon 14), Oren Ackerman (geomorphology), Ravit Nenner-Soriano (metal objects), Yael Gorin-Rosen (glass) and Deborah Cassuto (beads). Ground Penetration Radar work was carried out by Mnemotrix Systems, Inc. (and the report was compiled by Jessie Pincus BenAvraham). Avi Sasson and Yair Zoran documented the caves and Avi Sasson also contributed a chapter on the role of Kh. er-Rasm during the Ottoman period.

We would also like to thank Profs. Amihai Mazar, Moshe Fisher and Shimon Dar, as well as to the Israel Exploration Society and the Israel Antiquities Authority for allowing us to use illustrations. Financial support was given by the Institute of Archaeology and the Martin (Szusz) Department of Land of Israel Studies and Archaeology at Bar-Ilan University. Equipment was provided during the first seasons by the JNF. Additional financial support was given by the Krauthammer Chair in archaeology, the Moskovitz Chair for the Study of Historical Land of Israel, and the Rivlin Institute, all at Bar-Ilan University.

We would like to express our gratitude to all the above, as well as to Prof. Zeev Safrai for his help in initiating the project and throughout the work, and to Prof. Amos Kloner, Prof. Joshua Schwartz, Prof. Andrea Berlin, Dr Boaz Zissu, and Prof. Shlomo Bunimovitz for their help and advice, and to Dr Yehuda Dagan for bringing the site to our attention and for his advice and information. We would also like to thank Prof. Aren Maeir for reading an earlier draft of

The Kh. er-Rasm expedition was greatly assisted by a generous donation in loving memory of Elaine Levin Bard by her children and family. This contribution enabled much of the data processing and analysis, and we are grateful to the Bard Family for their contribution. Avraham Faust and Adi Erlich

4

Avraham Faust and Adi Erlich

0.2 Geographical and Environmental Background

(a.m.s.l.) borders the lower Shephelah in the west. A slope of approximately 60 m in height separates these two regions.

Kh. er-Rasm is located in the upper Shephelah, about one km south-southwest of Tel ‘Azekah, and about half a km west of the western part of the Elah Valley. This is a central location, near some of the region’s main sites and especially main routes, as the Elah Valley had served as a main junction throughout history (see also Chapter 4.2). It should be noted that the valley itself contains large tracts of arable lands.

The geological stratigraphy of the Shephelah is primarily composed of white chalk with chert lenses covered by a thick calcrete crust, known locally as nari (see also Chapter 2.5). The chalk is of the middle Eocene Age, being the Shephelah group Maresha and Adulam members of the Zor‘ah formation.

The site is located on a salient hill, and although Tel ‘Azekah and Kh. Shiqlon are higher in terms of absolute elevation, Kh. er-Rasm is still much higher than its immediate surroundings, with excellent observation over most of the region, including to the direct east (hindered today by high trees on the eastern part of the hill), west, north and south.

Relics of Neocene and Pleistocene sediments cover the Shephelah group by the following formations: Ziqlag, composed of bioclastic limestone and coralline and algal reefs; Beit Nir, composed of conglomerate altered to nari; Pleshet (in the lower Shephelah), composed of Calcareous pebbly sandstone; Ahuzam and Nahshon, composed of conglomerate of river terraces (Buchbinder 1969; Bar et al. 2006).

0.2.1 Environmental Background: Conditions of the Area in General

The pedological structure of the area is primarily composed of two soils types: Brown rendzina soil, found on the nari crust, and pale rendzina soil, found on soft chalk (Dan et al. 1972; 1976; 2007; Singer 2007).

Oren Ackermann The site is located in the upper Shephelah, at an elevation of 328 m above mean sea level (a.m.s.l). The Shephelah (lowlands) is part of the Israel mountain ridge stretching from north-northeast to south-southwest. The ridge is composed of several geomorphological sub-parallel strips. According to Bar et al. (2006) these strips are as follows (from east to west):

The climate is Mediterranean sub-humid, characterised by a hot, dry summer and a cool, rainy winter. The mean temperature in January is 11°C and the mean temperature in August is 26°C. The rainy season generally lasts from October to May, and the mean annual rainfall is 350 to 400 mm (Department of Surveys 1985; Gonen and Tahal 2004).

1. The Mountain Plateau, with summits ranging from 750 to 1020 m a.m.s.l.; 2. The Western Mountain Front, characterized by steep slopes ranging from 400 to 800 m a.m.s.l.; 3. The Higher Shephelah, with summits ranging from 320 to 480 m a.m.s.l.; 4. The Lower Shephelah, with summits ranging from 180 to 320 m a.m.s.l..

The vegetation is composed of a variety of Mediterranean formations: dwarf shrubs, shrubs and a large variety of herbaceous species. The most abundant species are: Quercus calliprinos (Kermes Oak); Pistacia lentiscus L. (Lentisk); Phillyrea latifolia (Mock privet; Jasmine box); Ceratonia siliqua (Carob); Rhamnus lycioides palaestinus (Palestine buckthorn); Sarcopoterium spinosum (Thorny burnet); Thymelaea hirsute (Gnidium); and Hyparrhenia hirta (Thatching grass) (Enviroplan 2004).

The Coastal Plain (with elevation of less than 150 m

5

Part I:

Architecture and Stratigraphy

Avraham Faust and Adi Erlich

1.1 The Main Building

The following is a stratigraphical-architectural discussion of the entire building, phase by phase. This discussion will not include any type of evidence other than architectural. The data presented here will be integrated with other types of data in section 4 below.

1.1.1 The Main Building: Architectural Phasing What we call here the main building (or the main structure) is the main architectural feature at Kh. er-Rasm. It is a large building, located on top of the hill at about 330 m above sea level, and it encompasses three excavation areas: A, B, and E (Fig. 1.1).

Overall View Phase 1 (Fig. 1.2): The first phase of the building was its

1.1 Plan of main building

9

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

1.2 Phases 1 and 1a

erection, as an almost square building measuring some 31 x 31 m, with a courtyard in the middle and four wings in the north, west, south and east. In this initial phase we include all the outer walls which gave it its contour, as well as the minimal number of inner walls which must have existed in order for it to function as a structure, of whatever nature (the evidence is, of course, to a large extent circumstantial).

At this stage the building was very symmetrical, and included at least all four outer walls (W.1 [western wall], W.2 [northern wall], W.27 [eastern wall], W.12, W.7, W.29 [all three are different parts of the southern wall]), the main parallel inner walls (W.3, W.4 [west wing], W.28 [south wing], W.20, W.31, W.42 [the last three are part of the inner wall bordering the courtyard from the north], W.51, W.10,

10

Avraham Faust and Adi Erlich

and probably also W.5 [eastern wing]), and at least some of the connecting walls (W.8, W.9, in the southeastern corner). It is likely that most of the connecting walls, at least those built of hard limestone (nari), were also built during this initial phase (below), but since this is not necessary for the building to function, and because all these walls ‘approach’ the main walls of the structure, we prefer to err on the side of caution, and to discuss them as a possible separate architectural phase.

It is very possible that W.21 (Fig. 1.4), which lies below W.1, is earlier, and belongs to a different building phase (phase 0 [?], see also Chapter 4.1), but because this wall lies almost precisely below W.1, it could also have been part of the foundation, built for construction purposes. We tend to view this wall as a remnant of an earlier occupation phase, but this cannot be proved. Phase 1a (Fig. 1.2): A number of connecting walls were probably constructed when the entire building was erected (phase 1, above), but since they could have also been built later, we prefer to regard them as a possible later phase. These include W.33, W.34, W.44 (northern wing), W.19, W.23, W.22 (western wing), W.50, W.35, and W.14. W.6 (southern wing), might have also been connected to this phase, but it is incomplete and we tend to date it to an even later phase (below). W.41 (northern wing) might have also belonged to this phase, at least its original / lower phase (for this unique feature, see below).

The entrance to the building was through a gap in the eastern part of the southern system of walls (and subsequently rooms). This gap led to the courtyard in the center, and was flanked by two rooms (one of them was probably completely ‘open’ toward the entrance), which functioned as a small gateway (Fig. 1.3). In addition, W.37, the ‘access ramp’ which lead to the entrance (see also discussion in Chapter 3.10) should also, on circumstantial grounds, be dated to this period. There was no point in constructing this wall at any time other than that when the building was entered through this entrance. Technically, it is possible that it was added at a later stage, but it appears that the most substantial building phase was the first, and hence it is more likely that this ‘ramp’ was built at this stage.

Although the walls discussed here (attributed to phase 1a) technically approach the main, structural, walls of the building (phase 1), and could therefore be later, we believe that it is more likely that they also belong to the initial phase of the building. We have no evidence for the removal of walls, and it is likely that these were the original inner walls

1.3 The entrance to the main building, facing north

11

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

1.4 W.21 (Square J52), facing west

of the structure which created the inner rooms. Therefore, despite attributing the walls to phase 1a, it is more likely that those walls stood there from the time the building was erected in phase 1.

It is possible, of course, that the upper courses of some or all of them were built of soft limestone (kirton), but this assumption is based only on the large number of kirton blocks found in the falls, and the fact that no such stone was found in-situ. Moreover, if kirton was used in connection with these walls, there is no way of knowing if this was the original usage, or if they were added only at a later stage.

Phase 2a (Fig. 1.5): In this phase we include small walls that appear to reflect changes and modifications. Those include W.39, W.32, W.40 (creating a small cell, bordering the northern wing from the south), and perhaps also W.26,1 which is a small separation wall.

The only exception to the walls mentioned so far is W.41. The lower course is constructed of nari on top of which a well built course of kirton blocks was added (Fig. 1.6). It is possible that this is representative of how the structure was built, and that many walls were built in this fashion: foundation and lower courses of hard nari stone, and upper part with kirton blocks. Another option is that the walls were initially built of nari, and that at a later stage they were rebuilt using kirton for the upper courses. From an architectural perspective, we prefer the second option. After all, we found walls that were made only of kirton, and it is likely that had a combination of stones been the preferred option of the builders, there would have been no need to build walls of kirton only. However, notice should be taken of the situation in Maresha, in which all types of walls were built contemporaneously (Kloner, Erlich and Whetstone 2000). Also, it should be stressed that when the building ceased to exist, it did not function for the purposes for which it was built (below); this means that there were several phases of usage, and not only one, and changes in building technique over-time are therefore possible. It

In addition, there are two walls adjoining the building from the outside: W.30 and W.38. There is no way of knowing when and why these walls were built, but it is likely that it was not during the initial phase of the building, otherwise these walls were the continuation of the original wall (W.1), rather than approaching it. There is no way of knowing which of the various walls attributed here to this phase (2a) are earlier and which are later, but we believe all of them postdate the initial phase of construction (phase 1 and probably also phase 1a). All the walls discussed so far including the lower / original part of W.41, but with the exception of its upper / later part, see below, were built only of hard limestone (nari). 1 Although it is possible that W.25 was built at the same time as W.26, we prefer to differentiate between them, and date the former to phase 2b, on the basis of the building materials (and since it approaches W.26). As we shall see below, despite this division, we tend to view both phases as contemporaneous.

12

Avraham Faust and Adi Erlich

1.5 Phases 2a and 2b

is possible of course to reconstruct additional stages, but apparently during the first phase the walls were built only of nari. Later, walls were added and rebuilt, using kirton. At this stage, the entirely new walls were built only of kirton, while in the case of older walls that were rebuilt, kirton was used on top of the nari courses that remained. We therefore tend to view the lower part W.41 as belonging to phase

2a (or maybe even phase 1, see above), while the higher part of W.41 is ascribed by us to phase 2b, to be presently discussed. We will return to this issue, of nari and kirton usage, below. Phase 2b (Fig. 1.5): In this phase we include W.16, W.15 (kirton walls, western wing), and W.25 (small wall

13

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

1.6 W.41 (Square E49), facing east

in H48, creating cells), as well as the building of kirton superstructure on the nari walls. While only one such wall (W.41) was uncovered, the number of kirton blocks unearthed in the debris was enormous, and it is more than likely that most of it was built during this phase.

the continued usage of nari (technically, phase 2a), for building purposes. The nari walls attributed to this phase with certainty were all small walls (creating cells), probably without a kirton superstructure. It should also be stressed that all the features that were added during phase 2b (and 2a of course), were placed within the large main building, and there is no reason to doubt that during this phase the main building, while modified, continued to function as one large square building.

A cave complex that was observed below the main building most probably also dates to this phase. While evidence is only circumstantial, it is likely that the caves were the source of the kirton blocks, and hence the caves are dated to the time of the kirton construction (see extensive discussion in Chapters 1.3 and 4.1).

Phase 2c (Fig. 1.7): A change that should be attributed to this phase is the blocking of the entrance at the edge of W.15 (the blocking is W.49). W.15 wall, built during phase 2b, originally had an opening, which was later blocked by kirton blocks (W.49). While most blockings in the structure were attributed to phase 3 (below), the fact that kirton was used for this blocking (and not nari, as in the other blockings) might hint that it was done during this phase, and prior to phase 3 (below). In addition, the fact that a kirton installation (W.17) is built with reference to this blocking indicates that this installation could not have been part of an earlier phase (i.e., phase 2b, before the construction of the blocking), nor could it have been of phase 3 (since during phase 3 the building was abandoned, and no installations would have been built; see below). Hence the kirton

Additional kirton walls, W.17 and W.36 (kirton installations, western wing and southeastern room, respectively), while possibly dating to this phase, were most probably constructed only during the next phase, as was the ‘modification’ of W.15 (the blocking – W.49). It is also possible that W.43 (tabun, southeastern corner) belongs to this phase as well, but again, it is more than likely that it too, belongs to phase 2c. It is probable that phase 2b and phase 2a are actually one phase. The difference is only technical, due to the appearance of a new building material. But the introduction of the kirton blocks (technically, phase 2b) did not prevent

14

Avraham Faust and Adi Erlich

1.7 Phases 2c and 3

installation must have followed the blocking (probably immediately), and therefore is dated to phase 2c.

Fig. 1.9), were most probably constructed during this phase. This is indicated not only by the similarity to the above (the building of kirton installations, as W.17), but one should also note that those features completely changed the character of this room and of the building’s gateway. The same is true of W.6 (Fig. 1.10). This wall, which closes room 2 west of the entrance, does not suit the original use

It is likely that some additional kirton features were added at this stage. W.36 (Fig. 1.8; kirton installation in the southeastern room – room 1), as well as W.43 (the tabun, which was found in the same room and at the same level;

15

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

1.8 W.36 (Square E54), kirton installation in room 1, facing northwest

1.9 W.43 (Square E54), tabun in room 1, facing east

16

Avraham Faust and Adi Erlich

result of the collapse of the cave system. That the collapse came after phase 2b – the major phase in which kirton was used – is obvious since the caves probably supplied the kirton needed for the building of the walls. The fact that kirton blocks are used only in phases 2b and 2c indicates that they were probably not far apart chronologically. It is likely that they were almost contemporaneous, and it is only the collapse of the caves which caused the changes in the use of the building (see extended discussion in section 4). The new building plan can be seen in Fig. 1.18. Phase 3 (Fig. 1.7): This is the final stage of the building, just prior to its destruction. Walls built in this phase include W.24 (blocking of doorway between W.4 and W.50), W.46 (blocking in W.34), W.47 (blocking in W.23), and W.48 (blocking in W.3). Those blockings will be discussed in Chapter 4.2 (see also Figs. 4.15, 4.16). All those features were apparently built just before the destruction, since once they were built, the building could not have functioned any longer. They mark the last phase of the main building, just prior to its destruction. Phase 4 (Fig. 1.11): Based on both the stratigraphical analysis and the small finds (pottery and other artifacts), we know that a small part of the building was reused at a later stage, after its destruction. Still, this use was very local in nature, and we did not identify its exact boundaries architecturally.

Discussion 1.10 W.6 (Square G53), eastern wall of room 2, facing south

Originally, a large and symmetrical building was erected on the top of the hill. This was a well built structure composed of large nari field-stones. At least the main contour of the structure was built at the time in one operation. We tend to think that the (nari) inner walls, connecting the main walls, were also built at this time, but this cannot be proven. After these walls were built (phase 1 and phase 1a) whether at the same time or not, however, we can identify the main features of the building (Fig. 1.2):

of the building either, and was probably built during this phase. The wall is quite flimsy, and includes stones (a monolith) that seem to be in secondary use, as well as some kirton stones. It appears that up to this phase this room was open toward the entrance. We tend to attribute W.25 and W.26 to earlier phases, but it is not impossible that these small cells should be attributed to phase 2c.

It was a rectangular building, measuring some 31 x 31 m, with rooms surrounding a courtyard (13.7 x 20 m) in the center. The entrance to the building was through an opening in the eastern part of the southern wall. A well built ramp leads to this entrance from two directions, forcing the visitors to turn either right or left, but preventing a direct approach from the south (Figs. 1.12, 1.13).

All the above clearly demonstrates that at some point, the building changed its character. The main entrance ceased to function as such, installations were added, etc. The available evidence we possess also suggests that before phase 3 (the abandonment and destruction, below), the entire building changed its function, and was used in a completely different manner from the way in which the original square and symmetrical building had been used (see Chapters 4.1 and 4.2). The changes include not only using the gateway rooms for installations, but also the abandonment of the area above the caves, probably due to their collapse.

The western and eastern parts were very similar at this stage, but since the western part was better preserved, the following description is based mainly on the finds from this area. Generally speaking, the rooms on the eastern and western sides of the structure were quite long, the distance from their outer to inner walls was about 6–7 m (inner dimensions, western and eastern wings respectively; when the walls are included, the width was some 8–9 m, western and eastern wings respectively). Between the inner and outer walls of those rooms, a long row of monoliths was

We therefore suggest that while phase 2b (probably the same as 2a) still used the building in its original plan, though changes took place, and kirton walls were added, during phase 2c major changes took place, most likely as a

17

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

1.11 Phase 4

unearthed (Fig. 1.14). Five of these in the western wing are still standing, and two others were found lying on the ground. In the eastern wing, only one monolith was found standing, and three others were found lying on the ground. These monoliths appear to have been a constructional feature, probably used to support the roof. In most places the monoliths did not serve as a long north-south wall, and small connecting (west-east) walls approached them.

In the south and north wings no monoliths were found, and the distance from the outer and inner walls here was only about 5.5–6 m. At a later stage, other changes were made. A number of walls were built using nari stones outside the building and in the inner courtyard, slightly changing the overall plan (phase 2a).

18

Avraham Faust and Adi Erlich

1.12 View of the main building with the ramp leading to the entrance, facing north-northwest

In addition, a few kirton walls were built inside the building (Fig. 1.15), and at least one wall was rebuilt using kirton for the upper courses (phase 2b). Due to the huge number of kirton stones found at the wall falls, mainly in the northern and western wings, we tend to believe that at this stage many of the walls were rebuilt using kirton blocks above the nari. A complex of caves below the structure probably

belongs to this phase, and the kirton blocks were probably hewn there. While we have differentiated between phases 2a and 2b on the basis of the use of kirton, we tend to believe that both belong to the same phase; because the walls attributed to phase 2a were all either built outside the main structure

19

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

1.13 Ramp leading to the entrance of the main building, facing east

or simply served as small separation walls, the builders preferred to use nari, whereas for walls inside the building they usually used kirton. Kirton (i.e., with no nari) was used when they built the walls from scratch, and for upper courses when reusing existing walls. The features built at the time include a number of small walls and installations, but it is more than likely that all the older walls in the northern and western wings were rebuilt using kirton above the nari courses, probably with the exception of the outer walls. In any event, the above changes modified the use of the main building, but it was still used in its original form, i.e., a square and symmetrical building. Later, however, we find many internal changes (phase 2c). It appears that the cave complex collapsed (see, e.g., Figs. 1.16, 1.17), and this led to a complete change in the use of the building. The areas above the axis of the caves were abandoned, and the building completely changed its character. The gateway rooms were now turned into part of the courtyard, and installations were built there and in other places. Habitation was concentrated in the western and northern wings, as is evident from the finds (see more in Chapters 4.1, 4.2). If we are correct, then the building plan was completely different at this point (Fig. 1.18). After the changes, the building continued to be used for a certain period of time, until the inhabitants blocked many

1.14 W.4, row of monoliths, facing north

20

Avraham Faust and Adi Erlich

1.15 W.15 (Square J52), kirton wall, facing south 1.16 Collapsed cave in Square I49, facing north

21

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

1.17 Collapsed cave in courtyard, facing north

1.1.2 The Main Building: Stratigraphy

of the entrances just before it was destroyed. The blocking of the doorways (phase 3) marks the end of the original use of the building.

The stratigraphy of the main building differs slightly from one area to another. Below we have attempted to simplify the discussion whenever possible. Basically, one can identify several schematic stratigraphical phases in the excavated layers. Not all layers were found in all areas of course, but the overall schematic types of layers (in the main building) include (generally speaking, from the top down):

Still, after a certain interval, there was a minor phase of reuse in a small part of the building (phase 4). The new inhabitants did not build any new walls, but simply used one room of the main building. This final phase was identified during the excavation, but without any specific architectural phasing (though it was stratigraphically distinct), and will therefore be discussed in the next section (see also Chapter 4.3).

1. Topsoil. 2. Topsoil mixed with stones (stoning mound). 3. Upper part of wall fall and topsoil mixed (due to the proximity of the upper part of the fall to the surface, in many cases there was a layer that was part of both). 4. Upper wall fall.2 5. Upper occupational debris.3 6. Wall fall, accumulation (usually very thick).

Summary In summary, the building was built as a square. Later it was modified, mainly using a different building material (kirton). Following the collapse of a cave complex, the building changed its use: only parts of it were now used, and its plan appears to be totally different (this reconstruction is supported by the analysis of the findings inside the building). The structure functioned according to this plan until the building was destroyed. A minor phase of reuse followed the destruction of the building.

Note that this phase was found only in one small area. It was identified during the excavation, and confirmed later during the analysis of the pottery and the stone vessels. 3 This phase, too, was found in only one small area. It was identified during the excavation, and confirmed later during the analysis of the pottery and the stone vessels 2

22

Avraham Faust and Adi Erlich

1.18 The function of the building during phase 2C

7. Wall fall and occupational debris. 8. Occupational debris. 9. Fills or accumulations below occupational debris (might be mixed). 10. Virgin soil. 11. Foundation rock.

13. Fill. In some places where there was disturbance, and it is not clear how the accumulated material assumed its final form. It is important to stress that types 3 and 4 (i.e., the upper wall fall and the upper occupational debris), were found only in one square in Area A, and in most parts of the building what one finds below layer 2 is layer 5. Layers type 12 and 13 are also very limited in distribution – type 12 is in the yard and outside the building and type 13 is found mainly in areas of disturbances.

Other debris: 12. Natural fill, for example in the courtyard or outside the main building.

23

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

1.19 West section of Square I53

It should also be noted that in most cases not all nine other types of layers were found: Sometimes bedrock, or other layers, were not reached. In other cases the excavated loci could easily be defined, for example as wall fall (type 6) and occupational debris (type 8), and no loci were therefore attributed to schematic type 7.

the eastern wings and part of the southern wings on the one hand, and the western and northern wings on the other (and the courtyard is, again, another story). We shall begin with a brief description of the stratigraphy by areas and squares, and will continue with a brief overview (for loci list and loci distribution map, see also Chapter 1.4).

It is worth noting that in a few parts of the site preservation was poor, and the bedrock is very high. Most types are, of course, missing from those areas.

Area A I53 (Fig. 1.19) Topsoil: L.169. Wall fall: L.170 (perhaps still mixed with some topsoil), L.184. Occupational debris (and a kirton [?] wall fall): L.176. Occupational debris / accumulations below occupational debris: L.186.

We should also note that the floors were of unbeaten earth, and were therefore not always directly identified as such – only the presence of artifacts and / or architectural features (e.g., blockings of doorways) indicated that this was the floor level. This also complicated the differentiation among the various types of loci, and increased the possibility of some (partial) mixing among the types (as can be seen below).

H53 Topsoil: L.188 = L.189.

Moreover, the fact that many seasons of excavations were short, and that we always opened new loci at the beginning of each season, also increased the number of loci, and is the main reason for the large number of ‘equal’ loci, as well as for ‘mixed’ loci like ‘topsoil mixed wall fall’, etc

I52 (Fig. 1.20) Topsoil: L.106, L.113 (in the courtyard), L.104, L.141. Topsoil (mixed within a stoning ‘mound’): L.117. Topsoil mixed with a wall fall: L.140, L.172. Upper occupational debris: L.175.

There are also differences in the nature of the finds between

24

Avraham Faust and Adi Erlich

1.20 North section of Square I52, showing upper and lower occupational debris

J51

Natural fill (in the courtyard): L.152. Wall fall: L.145. Lower occupational debris and kirton wall fall: L.147, L.185.

Topsoil: L.102, L.105 (outside the structure). Wall fall (maybe mixed with topsoil): L.112 (inside structure), L.111 (outside the building). Wall fall: L.122, L.123, L.125, L.126.

J53 (Fig. 1.21)

J50

Topsoil: L.108, L.118, L.136, and L.173 (outside the structure). Wall fall: L.150. Natural fill (outside structure): L.174. Occupational debris and kirton wall fall: L.157, L.159. Virgin soil: L.167.

Topsoil: L.100 (outside the structure), L.101. Wall fall mixed with topsoil: L.119.

I47 Topsoil: L.179, L.181.

J52

I49

Topsoil: L.103, L.107 (outside the main building). Wall fall: L.114 (mixed with topsoil). Wall fall (probably mixed with occupational debris, at least the lower loci, marked here as a =): L.120 (=L.130), L.121 (=L.129), L.109 (=L.116), and L.110 (=L.115). Occupational debris (and kirton wall fall): L.134, L.135, L.139, L.128 (=L.127). Virgin soil: L.144, L.131 (=L.133).

Topsoil: L.155. Fill / wall fall: L.164 (=L.165), L.168.

I50 Topsoil: L.156. Fill: L.163.

I51 Topsoil: L.183.

25

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

1.21 Top of wall fall in Square J53 facing south

H49

G54

Topsoil: L.143. Wall fall mixed with topsoil: L.148. Occupational debris, mixed with fill: L.149, L.153, L.154, L.161.

Topsoil: L.187.

E54 (Figs. 1.28, 1.29). Topsoil: L.302, L.316. Wall fall (mixed with occupational debris?): L.321, L.326, L.325 (inside installation), L.331 (tabun).

H54 Topsoil: L.180.

F53

I48 (practically, part of Area E) (Figs. 1.22– 1.24).

Topsoil: L.306. Below topsoil (mixed): L.311, L.312, L.313, L.314.

Topsoil: L.124. Wall fall: L.132, L.142, L.151, L.160. Fills of installations in southern part: L.137, L.138, L.162, L.166.

D51 (Fig. 1.30) Topsoil: L.320. Below topsoil (mixed): L.332.

Area B

F51 (Fig. 1.30)

Notably, the preservation in Area B was, in most cases, not as good as in Areas A and E.

Topsoil: L.322.

E51 (Fig. 1.30)

G53 (Figs. 1.25–1.27)

Topsoil: L.330.

Topsoil: L.301 = L.303 = L.304 = L.308. Below topsoil (wall fall?): L.305, L.307, L.308, L.309. Wall fall (fill?): L.315, L.317, L.318, L.319. Wall fall mixed with occupational debris (?): L.323, L.324, L.327, L.328. Virgin soil: L.329.

F54 Topsoil: L.310.

D50 Topsoil: L.333.

26

Avraham Faust and Adi Erlich

1.22 Top of wall fall in Square I48 facing south

Area E H48 Topsoil: L.803. Wall fall (kirton): L.813, L.814. Below wall fall / occupational debris / ash layer: L.821. Ash layer (clean, below the wall fall): L.823. G48 (Figs. 1.31–1.33) Topsoil: L.802. Wall fall (kirton): L.810. Below wall fall / occupational debris / ash layer: L.818. Ash layer (clean, below wall fall): L.824. D49 (Fig. 1.34) Topsoil: L.820. E49 (Figs. 1.35–1.39) Topsoil: L.804. Wall fall: L.815 (northern part), L.816 (southwest part), L.819 (southeast part). Occupational debris (ash layer): L.822, L.825 (northern part). Fill below floor: L.826. F49 (Fig. 1.40) 1.23 Part of east section of Square I48 during excavation

27

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

1.24 South and east sections of Square I48 after excavation, facing southeast

1.25 South section of Square G53

28

Avraham Faust and Adi Erlich

1.26 Square G53 facing east at the beginning of excavation

1.27 South section of Square G53

29

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

1.28 Square E54 facing north at the beginning of excavation

1.29 Square E54 after excavation, facing north

30

Avraham Faust and Adi Erlich

1.30 Squares D51, E51 and F51 facing south

1.31 Squares G48 and H48 at the beginning of excavation, facing west

31

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

1.32 Top of wall fall in Square G48 facing east

1.33 Ceramic vessels in situ in Square G48

32

Avraham Faust and Adi Erlich

1.34 Top of wall fall in Square D49 facing south

1.35 West section of E49

33

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

1.36 North section of E49

1.37 Top of wall fall in Square E49 facing east

34

Avraham Faust and Adi Erlich

1.38 North section of Square E49

Topsoil: L.801. Wall fall (kirton): L.805 (in cell), L.811 (south of the cell), L.806 (east of cell), L.807 (north of cell). Fill (in cell): L.808. Occupational debris (ash layer): L.817 (north of cell). G50 (above a cave in the courtyard) (Fig. 1.17) Topsoil (and fill): L.809.

Discussion There are some differences between the various wings of the structure, in terms of stratigraphy. Below we will first describe the stratigraphy of the northern wing, then that of the western (and southwestern) wing, and finally that of the eastern (and southeastern) wing. This will be followed by an integrative discussion of the stratigraphy of the entire structure. Northern Wing (Area E and square I48 of Area A): Basically, the stratigraphy of this area is quite simple. A thin layer of topsoil is followed by a very thick accumulation of wall fall, composed mainly of kirton blocks. At the bottom of the wall fall we find a layer of ash, with vessels (see Figs. 1.33, 1.38–1.39). 1.39 General view of Square E49 after excavation, facing west

35

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

1.40 Top of wall fall in Square F49 facing west

1.41 W.8 (Square E54), facing north

36

Avraham Faust and Adi Erlich

1.42 Plan of main building with room numbers

37

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

Western Wing (mainly Area A): The stratigraphy in this area is very similar to that of Area E. A layer of topsoil covers the area, although at some places the stones seem to have been visible before excavation. It appears, however, that these stones were not part of the original wall fall (below), but part of a ‘stoning mound’ (a place to which stones were removed when cleaning fields), formed at a later stage. The topsoil is followed by a wall fall. In this case, the fall includes a larger portion of nari field stone (in contrast to the northern wing). Below the wall fall we find many sherds and vessels. Ash was found in some places, but not as a uniform layer covering the entire area. Below this level we find virgin soil.

The different stratigraphy of the various areas is an outcome of the architectural history of the site (above, see also section 4). 1.1.3 The Main Building: Architecture The following chapter will describe the main building (for general views, see Figs. 1.42–1.49), room by room, or space by space. We will refer to only two phases: the last phase of the building, including all the installations and partition walls, and what appears to have been the original plan (see Chapter 4.1). We shall begin our description of the building in the southeastern corner, near the entrance to the building. The entrance to the building is flanked by two rooms (Fig. 1.50).

In one room we identified a phase (phase 4) of reoccupation (Fig. 1.20).

Room 1: The room to the east of the entrance (room 1) is bounded by W.8 (north), W.27 (east), W.29 (south) and W.9 (west). Its dimensions are some 3 x 3.5 m (inner size only; the width of the wall is approximately one m). The entrance to the room was not unearthed, but since the southern part of the room’s western wall (W.9) was outside the squares and was not excavated, it is likely that this is where the entrance was located. This is especially likely since most of the doorways in the main building were located near the corner of the walls, rather than in the middle. The walls themselves were not preserved to the same degree. While the northern wall (W.8) was preserved to a height of some 1.7 m (see above, Fig. 1.41), preservation is much lower in the south (the southern wall [W.29] is preserved

Eastern Wing (mainly Area B): Most of the area is quite disturbed, and the topsoil almost reaches bedrock (Fig. 1.30). However, the southeastern part, near the gate, is well preserved (Fig. 1.41). Still, even here the stratigraphy is somewhat different. Following the topsoil (which was at times indistinguishable from the wall fall, maybe due to the use of the area as a stoning mound for the nearby fields), one finds a wall fall. The fall is composed almost solely of nari field stones. Below the wall fall, on the bottom of the fall, we find the floor level, with a few installations (and kirton bricks), but with very few additional finds (no vessels, and only a limited amount of pottery sherds). Below this level, we reached bedrock.

1.43 General view of the western part of the main building, facing north-northwest

38

Avraham Faust and Adi Erlich

1.44 General view of the western part of the main building, facing north

1.45 General view of the eastern part of the main building, facing north

39

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

1.46 General view of the eastern part of the main building, facing south

40

Avraham Faust and Adi Erlich

1.47 General view of the western part of the main building, facing north

to a height of only one or two courses). In the northeastern corner of the room a square installation was found (Fig. 1.8), bordered in the south and east by kirton walls, and probably dating to the last phase of the building (phase 2c). Adjacent to the eastern wall (W.27), near its southern corner, a complete tabun was unearthed (Fig. 1.9). It is also probably dated to the last phase of the building.

and allowed access from the east and west, and blocked any direct approach to the entrance (see Figs. 1.12, 1.13). Room 2 (Figs. 1.50, 1.53): This room is located west of the entrance, parallel to room 1, and is bounded by W.14 (west) W.28 (north), W.7 (south) and W.6 (west). The room was not excavated in its entirety (only its northern half was excavated). It appears that its dimensions were 2.5 x 3.5 m (inner dimensions), but this is not certain. Moreover, the wall which separated it from the entrance (W.6; see Fig. 1.10) seems to have been added only in the later phase (2c). During the earlier phase(s) the room seems to have been open toward this direction (i.e., toward the entryway).

The entrance to the building was between room 1 and room 2. It was created simply by means of a void in the system of rooms which existed in the southern part of the building. The entryway passed between rooms 1 and 2 (approximately 4.7 m apart), and led to the courtyard. W.5 seems to have stood in the middle of the entrance (Figs. 1.51, 1.52), and might have interfered,4 but we cannot attribute it to a different phase and so the wall and the entrance must have coexisted. The approach to the entrance from the outside was not direct. It was possible to reach the entrance only from the east or west, and then by turning north to enter the building. For this purpose a ramp, W.37, was constructed (it was not a terrace wall, cf. Chapter 3.10),

Space 3 (Figs. 1.53, 1.54): This area (approximately 6 x 4 m – inner dimensions) is bounded by W.14 (east), W.7 (south), W.35 (west) and the continuation of W.28 (north). It is located between Area A and Area B and was not excavated. It probably consisted of one or two rooms. The possible entrance was from the courtyard (space 20) and was probably located in the northwest part of the area. Room 4 (Fig. 1.54): This is a large room (approximately 4.4 x 5 m, inner dimensions) west of space 3. It is bounded by W.35 (in the east), W.12 (south), W.50 (west) and W.23

Note that for technical reasons (accessibility to the main building and soil removal), the square was covered again with soil, and is not visible in some of the pictures. 4

41

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

1.48 General view of the northern part of the main building, facing west

42

Avraham Faust and Adi Erlich

1.49 General view of the northwest part of the main building, facing west

1.50 Rooms 1 and 2 facing north

43

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

1.51 W5 (Square F53) facing west

1.52 W5 (Square F53) facing north

44

Avraham Faust and Adi Erlich

1.53 Rooms 2 and 3, facing northwest

1.54 Rooms 3, 4 and 5 facing northwest

45

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

1.55 Rooms 6, 7, 8 and 9 facing northwest

(north). The entrance to the room was from the courtyard, and was located in the northeastern corner of the room (in the eastern edge of W.23). Later, before the building went out of use, the entrance was blocked (W.47). Room 5 (see Fig. 1.54): This room (inner dimensions 3 x 4.4 m) is located in the southeastern corner of the building. It is bounded by W.50 (east), W.12 (south), W.1 (west) and W.19 (north; this wall is the continuation of W.23). The entrance to the room was through room 6 (and 7, see below), with the doorway located in the western part of the northern wall (W.19). Room 6 (Figs. 1.55, 1.56): This is a small room (inner dimensions 1.4 x 2.8 m), which was created after the erection of W.15 (in the north). It is bounded on the west by W.1, on the south by W.19, and on the east by W.4 (the row of monoliths). The entrance was through a gap in the monolith wall. Just before the building was destroyed this gap was blocked by W.24. This small corridor, which leads to room 5 (through room 7, from the courtyard) was created only at a late phase, when W.15 (kirton wall) was built. Interestingly, it is possible that this wall had two phases. Originally, it had an opening in its western part (phase 2b). The opening was later blocked (phase 2c) using kirton bricks; W.49. Prior to that, room 6 was part of larger space which also included room 9 and probably space 10. The connection with space 11 and rooms 8 and 7 is less clear,

1.56 Room 6 facing east

46

Avraham Faust and Adi Erlich

but it is possible that during the initial phase they were all part of one big space. Room 7 (Figs. 1.55, 1.57): This small room (1.2 x 3.8 m) is practically a corridor. It is bounded on the west by W.4 (or by the blocking W.24, see above), on the south by W.23, on the north by W.22, and it was open to the courtyard in the east. This opening, or doorway, was later blocked by W.48 (a blocking in W.3). This room served only as a corridor, during all the phases. Initially it led from the courtyard (space 20) to a larger space of which room 6 was only a small part. Later, it led to room 6, which was practically its continuation, and from there to room 5. Room 8 (Fig. 1.55): The dimensions of the room are unknown, as only its southern part was excavated. It was bounded by W.3 on the east, W.22 on the south and W.4 on the west. The area north of it is labeled space 11, and the relationship between them is unclear. Room 9 (Figs. 1.55, 1.58): This appears to be a relatively large room (inner dimensions, 5 x 2.4 m). It is bounded by W.1 (west), W.15 (south), W.4 (east) and W.16 (north). Since both, the northern and southern walls, are regarded as belonging to the later phase (2b), it is clear that this particular room existed only then. It is possible that at the time it was part of a larger room which included space 9, or part of it, and maybe room 8 as well. It is not clear where the entrance to this room was located, but there was an 1.57 Room 7 facing east

1.58 Room 9 facing north

47

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

1.59 W.15 and W.17 in room 9 facing north

to its south means that we cannot say much about its plan. There is a connection with a subterranean space in this area, but it seems to be a result of a collapse and not an original entrance.

early doorway in W.15. Since it is clear, however, that this doorway was blocked (W.49) before the last phase of the building (i.e., prior to phase 3), as an installation (W.17) was built ‘on’ the blocking, this large room must have had another entrance.

Room 13 (Fig. 1.24): This room stands in the northwestern corner of the building. Its inner dimensions are app. 3.5 x 3.3 m, and it is bounded by W.2 (north), W.1 (west), W.34 (east) and W.20 (south). Preservation of the walls is very good, and some are standing to a maximum height of 1.2 m. Later the room was divided by an east-west wall (W.26), and another flimsy wall (W.25) which divided the smaller, southern, room into two halves (east and west), thereby creating a large northern room (13a) and two smaller cells in the south (13b in the east and 13c in the west). The smaller sub-rooms might have served as installations. It is also worth mentioning that there was some quarrying in this area, but its nature is unclear. Many parts of a tabun were found here, but nothing in-situ (probably fell from the roof). The entrance to the room was from the northern part of the eastern wall (W.34) by the way of room 14, and through it to the courtyard. Before the building fell out of use, the doorway was blocked by W.46 (phase 3).

In the southwestern corner of the room, a semi-circular installation was unearthed, built of narrow standing kirton slabs (Fig. 1.59). It was probably used for storage (it belongs to phase 2c). During the earlier phase, this space was probably part of a larger room. Room 10: To the north of room 9 is another room (from the second phase). Almost nothing is known about its dimensions. It could have been connected to space 12, but since the area was only partially excavated and because this is the least preserved part of the western wing of the building, not much can be said. In the earlier phase this room did not exist, and apparently was part of a larger space. Space 11: A large space, or room(s), bounded by W.3 on the east and W.4 on the west. It is connected with room 8 in the south and maybe with spaces 12 in the north, and 9 and 10 on the west. The area was not excavated, so not much can be said about its plan.

Room 14 (Fig. 1.60): This is a small room (inner dimensions app. 3.7 x 2.8 m), bounded by W.2 (north), W.44 (east), W.20 (south) and W.34 (east). Preservation is excellent and the walls are standing to a maximum height of 1.2 m. The entrance to the room was through its southeastern part, in the eastern part of W.20. Another doorway connected this room with room 13. This latter doorway was blocked before

Space 12 (Fig. 1.16): The area to the north of room 10 and space 11. It is poorly preserved, and the lack of excavations

48

Avraham Faust and Adi Erlich

1.60 Room 14 facing west

the building stopped being in use, though not the entrance to the room itself, which appears not to have been blocked.

not unearthed, although its location is quite certain as W.31 and W.20 are built in a straight line, which the southern wall of room 15 is supposed to complete. This assumption is reinforced by the fact that even further east, the very same line continues by W.42. We cannot, however, know where the entrance to the room was, though it is likely that it was either on the eastern or western side of the southern wall.

Room 15 (Fig. 1.61): This is a fairly large room (inner dimensions app. 4.3 x 3.5 m; note that this size is reconstructed, below), bounded by W.2 (north), W.33 (east), W.44 (west) and the continuation of W.31 (south). The preservation here, as in the entire area, is excellent, with the walls standing to a height of 1.25 m. The southern wall was outside the excavated squares. Therefore, the wall itself was

Room 16 (Fig. 1.48): While not all the space was excavated, the area that was excavated makes the plan quite clear. This

49

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

1.61 Room 15 facing north

is a very large room (inner dimensions app. 9.7 x 3.5 m), bounded on the north by the continuation of W.2, on the east by W.41, on the west by W.33, and on the south by W.31 and W.42. Preservation is very good. The large area that was excavated does not allow us to reconstruct any additional wall that could have divided this room any further. The entrance to this room, which led from the courtyard, was atypical and was located in the middle of its southern wall.

western part of the southern wall (the upper course of the wall was not visible there, perhaps because this is where the entrance was located). Cell 18 (Fig. 1.63): This is a small cell (1 x 2.3 m) built in the courtyard adjacent to W.31, which was created by the construction of three walls (W.32, W.39 and W.40). This was a later addition to the main building. It contains mainly garbage.

We would like to note that while the northern wall of this room is quite obvious (W.2), the area north of the wall does not slope down sharply (as, for example, north of rooms 13 and 14), and one can speculate that perhaps there was an additional room or structure in this area. Still, we do not have any real evidence to support such speculation, and here we just wish to note this possibility. Naturally, we cannot comment on the possible stratigraphy of such a theoretical room / structure.

Space 19 (Fig. 1.45): This figure is used to designate the entire eastern wing of the building (south of room 17 and north of room 1). It was only partially excavated, and when it was, the remains were very poor. Generally speaking, it was bounded on the north by W.51 (southern wall of room 17), by W.8 (northern wall of room 1) on the south, by W.5 on the west and by the continuation of W.27 on the east. In the original phase it was probably subdivided (on the northsouth axis) by a row of monoliths (W.10). The entire space measures app. 5.4 x 20 m (inner dimensions). In any event, this area is the least known and least preserved part of the building (probably due to the collapse of the cave system).

Room 17 (Fig. 1.62). This room, which stands at the northeastern corner of the building, was bounded by W.41 (west), the continuation of W.2 (north) the continuation of W.27 (east) and W.51 (south). It was barely excavated; only the topsoil was cleaned, and the massive wall fall exposed, hinting at the existence of walls, but on the basis of the outer walls its inner dimensions can reasonably be reconstructed as app. 3.4 x 4.2 m. It is not clear where the entrance was, but it is likely to have been located at the

Space 20 (Fig. 1.64): This is the central courtyard of the building. With the exception of cell 18, it was practically empty (at least as far as we know). In its northwestern part an opening to the subterranean complex was observed, but it is not clear whether this was an original entrance or

50

Avraham Faust and Adi Erlich

1.62 Room 17 facing south

1.63 Cell 18 facing south

51

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

1.64 Courtyard 20 facing north

not. The complex itself was not excavated, due both to the danger of collapse and to the presence of cave ticks.

The relationship between the main building and Area C is apparent both in the access between the two and in their common orientation (Figs. 1.65, 1.66).

The courtyard is bounded by W.42 / W.31 on the north (note that we are not referring to cell 18), W.3 on the west, W.28 and its continuation on the south, and probably W.5 on the west. The situation in the east, however, is unclear, especially not during the last phase, when this part of the building did not fully function. Taking into consideration the above boundaries, the overall dimensions of the courtyard are 13.7 x 20 m.

Area C was excavated during the last three seasons of excavation, in the years 2000, 2001 and 2003. Today, it is surrounded by olive trees and other plants (Fig. 1.67). Prior to the dig, a heap of stones was visible in the area, which was thought to be some kind of a building composed of walls, accompanying the main building from the south. Eventually, it turned out to be a solid stepped structure rather than an ordinary building.

1.2 Area C: The Stepped Structure – Architecture and Finds

The top of the structure was cleaned and its walls were excavated (mainly on three sides: north, west and south). The squares of Area C do not fit the grid of the main building, but are numbered within the area. They cover the building and its immediate surroundings. Squares nos. C1, C2, C3, form the northern strip, from west to east. The central strip was dug in squares C4, C5, and partially also C6, from west to east. The southern strip includes squares C7, C8, C9, of which only the middle one, square no. C8, was excavated (Fig. 1.68).

The discussion in this chapter will be somewhat different from that of the main building, and the finds within this area will be addressed here, in order to simplify the report.5 Area C is situated on the southern slope of the mound, about 25 m south / southwest of the main building. Wall W.37, which runs south of the entrance of the main building, leads toward the structure of Area C, and appears to allow passage between the two. Wall W.37 (Area Z1) and terrace wall W.45 (Area Z2) frame a terrace sloping moderately to the south, on which the structure of Area C was constructed.

The Top of the Structure The surface of the structure was cleaned of topsoil, plants, roots, and fallen stones of varying size, which most likely originated from the structure itself. The cleaning of the

Since most of the report deals with the main building, it is not possible to discuss the finds unearthed in it in this section, as there are separate chapters that are devoted to it later. 5

52

Avraham Faust and Adi Erlich

1.65 Plan of the main building and Area C

53

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

1.66 The main building and Area C, facing east

1.67 The structure at Area C as viewed from the main building, facing south-southwest

54

Avraham Faust and Adi Erlich

1.68 Plan, section and elevation of Area C

55

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

1.69 Area C facing west-southwest

The Outline of the Structure and Wall Faces

northern part of the structure was done in loci L.511, L.512 (square C1, including also the area outside the northwestern corner), L.515 (square C2) and L.513 (square C3). The central area of the structure was cleaned in loci L.504 (square C4), L.501, L.514 (square C5, the centre of the structure), and L.502, L.509 (Square C6). The southern part of the structure was revealed in loci L.508 (squares C4 and C7), L.503, L.507, L.524 (square C8), and L.509 (square C9). Loci L.501 and L.515 supplied one complete cooking pot (see below).

In order to outline the building, four sections were excavated along its outer walls: one on its northern wall, a second one on the northwestern corner, a third on the western wall and the fourth at the southern wall.

Northern and Northwestern Sections An olive tree northeast of the structure (Fig. 1.72) limited our investigation of the northern face of the building in square C2.6 In the northern half of the square, north of the building, we exposed a small area measuring approximately 4 x 2 m. The loci (L.516, L.517, L.518. L.519, L.520, L.521, L.522) lay on top of the uneven bedrock, and contained mostly topsoil and few fallen stones. This section revealed the northern wall of the structure, composed of two courses of irregular stones, which were founded on bedrock or on a foundation of rubble (Fig. 1.73). Above these two courses, the building ascends in terraced courses, part of which, as described above, were found in collapse on top of the structure. In the northwestern corner the surface of

The upper surface of the structure suffered collapses, but its contour in straight stepped courses is still clearly visible, especially in square C1 at the top level (Figs. 1.69, 1.70, 1.71). Judging by the fallen stones on top of the building, as well as around it (see below), it originally had more courses and stones. The flattened highest level shown in Figs. 1.69–1.71 might be the original top (otherwise, it could be one of the middle courses which survived after the collapse, and became flat as a result of its exposure to both natural and anthropogenic erosion over the years, but we find this interpretation very unlikely).

The site is located within a JNF park (The British Park), and no trees were cut down during the excavations (trees were only trimmed by the JNF inspectors). 6

56

Avraham Faust and Adi Erlich

1.70 Area C facing west

1.71 Area C facing north

57

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

1.72 Area C facing southwest

the building was excavated together with a small section outside it (loci L.511, L.512) revealing two lines of stones approaching its base (Fig. 1.68).

Western Section In the west, it appeared that a section was previously cut in the building, possibly by a tractor poaching stones (Figs. 1.74–1.75). This section, measuring 1.5 m in width and cutting 2 m deep into the structure, overlaps part of square C4. We cleaned this section in loci L.505, L.506, and exposed the core of the building (Fig. 1.76). The section was filled with soil and large irregular stones (1 x 0.5 m) which fell into the section. A complete cooking pot has been restored from these loci, together with loci L.501 and L.507 (see below). The core of the structure was made of up to 10 courses of irregular stones laid one on top of the other, and mixed with soil (Figs. 1.77, 1.78). The topsoil of L.510 was dug in square C4 to the west of the section and along the southern part of the western wall. It revealed the western wall of the structure, as well as the lower course of stones which was not damaged by the tractor, and allowed us to draw the whole western outline. The western wall of the structure changes according to topography: the northern part of the west wall is composed of two courses, while its southern part, with its base sloping downwards, consists of three courses. Above this western wall, the structure rises in steps upwards and withdraws to its centre. Even after suffering from collapses, the structure revealed from the west shows its clear, solid and pyramidlike character.

1.73 Area C, north wall of structure facing south

58

Avraham Faust and Adi Erlich

1.74 Area C facing east

1.75 Area C facing east

59

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

1.76 West wall of structure at Area C facing southeast

1.77 Section in the structure at Area C, facing east-southeast

1.78 Section in the structure at Area C, facing east

60

Avraham Faust and Adi Erlich

1.79 The structure at Area C before excavation facing north

1.80 South wall of structure at Area C facing northwest

61

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

1.81 South wall of structure at Area C and collapse during excavation, facing north

1.82 South wall of structure at Area C after excavation, facing north

62

Avraham Faust and Adi Erlich

Southern Façade

site. However, the two complete Early Roman cooking pots of the same type, found on top of the structure, are notable. They are of the same type as the one that was found complete in Area A of the main building (see Chapter 2.2). For some reason, these two pots were placed on the structure and left there. There are a few more Early Roman sherds in the area, relatively more than usual in the site.

In the south of the structure, in square C8, the relatively steep slope allowed the accumulation of wall fall at the façade of the structure (Fig. 1.79). The topsoil was dug in locus L.523, and the massive collapse of large stones from the structure was excavated in locus L.526 (along the southern wall) and L.525 (south part of the square, Fig. 1.80). The excavation revealed an upright wall of three courses, on top of which the stepped structure ascends (Figs. 1.81, 1.82). Like the other façades, it was constructed on top of bedrock and some rubble.

Dressed Stones near the Structure To the north of the structure, two dressed stones were found on the surface. One of them seems to be a cover of a step (Figs. 1.86, 1.87). The other stone is more roughly cut (Fig. 1.88). Its large size and flat shape are unusual in the structure. These two stones might be remains of the original dressed stones covering the whole structure. No remains of any other covering or coating were found on or around the building.

Eastern Façade The eastern wall of the structure was only partially cleaned in L.527 (Figs. 1.83–1.85). It seems to be similar to the other walls, with a short upright wall framing the structure, and additional courses on top of it withdrawing eastward, creating the pyramid-like structure.

Summary

The Finds (Tables 1.1, 1.2)

The cleaning of the structure, and the excavation along its facing walls, revealed a rectangular solid structure, with short facing walls on the north and south and long facing walls on the east and west. The northern wall is about 7.30 m long. The western wall measures about 9.20 m, the southern wall about 7.70 m and the eastern wall about 8.80 m. All measurements are approximate, as the eastern wall was not really excavated, the walls are irregular, and in certain places seem to have been slightly moved or inclined during the years.

Since the debris of soil covering the area is rather thin above bedrock or above the solid structure, all loci should be treated as topsoil. Still, in some cases we could distinguish surface soil from collapses or virgin soil. In the case of pottery, no locus could be defined as sealed or clean. The ceramic finds from Area C are typical of topsoil, with samples from almost all periods represented at the

1.83 The structure at Area C facing west

63

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

1.84 The structure at Area C facing west

1.85 The structure at Area C facing west

64

Avraham Faust and Adi Erlich

Table 1.1 Indicative Ceramic Found in Topsoil Loci Mixed with Collapse on Top of the Structure Locus

Square

Location of Topsoil on Structure

Types*** (quantity)

Date

L.511* L.515+ L.501 L.501

C1 C2 C5 C5

Northwest of structure North and centre of structure Centre of structure

CP9

late 1st century BCE–early 2nd century CE

CP9 (complete)

late 1st century BCE–early 2nd century CE

SJ1C BL1B CP9 (2) JT3 (2)

Late 2nd century BCE Second half of 4th–2nd centuries BCE 2nd–1st centuries (at Tel Dor, it appears as early as the 4th century) late 1st century BCE–early 2nd century CE Late 2nd century BCE to 1st century CE

SJ9

Medieval

R-KRX1

Iron Age I

BL1B

Second half of 4th–2nd centuries BCE

BL2

L.509

C6–C9

L.508

C4

L.507

C8

Middle and southern east South west of structure Middle-south of structure

L.507+L.505+ CP9 (Complete) late 1st century BCE–early 2nd century CE L.506**+L.501 * L.511 expands also outside the structure. ** L.505 and L.506 are not on the top of the structure, but in the section cut in its western wall. *** The ‘types’ and ‘date’ rubrics follow the discussions in section 2 of this report.

1.87 Worked stone near the structure in Area C

1.86 Worked stone near the structure in Area C

1.88 Worked stone near the structure in Area C

65

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

Table 1.2 Ceramics from the Sections Surrounding the Structure Locus

Square

Location and Character

L.516

C2

Topsoil north of structure

L.517

C2

Topsoil north of structure

L.518 L.521

C2 C2

Topsoil north of structure Soil on bedrock north of structure

L.510

C4

Topsoil west of section

L.523

C8

Topsoil south of structure

Types (quantity) SJ1B SJ1E SJ1C BL1B Shai’s type 4 Shai’s type 4 Shai’s type 4 BL1B JT2 SJ1B BL2 JT3 CP9 (2)

Date 3rd–2nd centuries BCE Hellenistic Late 2nd century BCE Second half of 4th–2nd centuries BCE Iron Age II (mainly IIA) Iron Age II (mainly IIA) Iron Age II (mainly IIA) Second half of 4th–2nd centuries BCE 2nd century BCE Hellenistic 2nd–1st centuries BCE (at Tel Dor, it appears as early as the 4th century) Late 2nd century BCE to 1st century CE Late 1st century BCE–early 2nd century CE

The northern wall, which is built on the upper spot of the slope, consists of two courses, while the southern wall, which is constructed on the lower spot of the slope, has three courses in order to level the layers of building. The eastern wall levels with the other walls. The western wall has two courses in its northern part and three courses in its southern part. The section cut in the west side of the structure revealed its solid nature, consisting of layers of stones laid one on top of the other. The inner stones are irregular and filled with earth, while the outer (facing) stones are more finely cut.

situated all over the site, except for the eastern slope of the mound. Some of the caves are currently inhabited by ticks carrying cave fever. The southern cave is columbarium F, to the north of which lies the cave excavated as Area D. Further north there is a large cave numbered G, connected to Area D by a short burrow. Caves D, F, and G are located on a north-east axis to the east of the main building. A large subterranean complex, H, lies under the northern part of the main building, crossing it on an east-west axis. Two more unexcavated subterranean features, I and J, are in the northern slope of the mound (Figs. 1.89, 1.90).

Although many stones fell off the structure, the original shape of the structure is still visible. It was a stepped pyramid-like structure, standing on top of upright walls on four sides. It was probably covered with dressed stones, which may have been robbed, two of which are still found near the structure. The upper layer, which remains until the present, is flattened and seems to be finished, what might suggest its being the original upper level. Otherwise, the pyramid could have ended in a pointed top, which collapsed or was robbed later. From the flat top of the structure, there is a panoramic view to the south and west. Had it not been surrounded by olive trees, the structure could also have been visible from the distance. The building will be discussed in detail in Chapter 4.3.2.

During the survey of the caves we had three goals, in addition to the ‘basic’ recording of the data: First, we wish to study the techniques and technologies which were used by the stonecutters. Second, we would like to examine the use and development of the caves. And third, we find it important to record all the remains in order to place them on the map for tourists and travelers, as part of the British Park. Until now, these caves have been missing from the historical-topographical maps, including those of the British Mandate. Some of them were surveyed in the Israel Antiquities Authority (IAA) survey, to be published in the future by Yehuda Dagan. The survey and partial dig in the caves have resulted in only few finds, therefore the dating of all installations is circumstantial only.

1.3 Caves and Subterranean Complexes

1.3.1 Caves D and G: Caves and a Hiding Burrow

Cave G (the Cactus Cave) (Figs. 1.91–1.94)

Avi Sasson, Yair Zoran, Adi Erlich and Avraham Faust7

Cave G is located several dozen meters west of the western wall of the main building. In the space between the two lies the central room of this cave, whose ceiling collapsed in ancient times; today it is covered with cactus bushes. The length of this collapsed room is approximately 15 m and its width is 13 m. In the remaining large space, a few stepped levels of quarry are traceable, the lowest of which is the preserved western room. Today, the entrance to the caves is

During the excavations at the site, several caves and cisterns were surveyed, with one of them also partially excavated (Area D). Some of these spaces consist of a single cave, while others are subterranean complexes. The caves are The research of the caves was conducted with the assistance of the Research Fund of the Ashkelon Academic College. We would like to thank them. Most of the caves were drawn by Yair Zoran. 7

66

Avraham Faust and Adi Erlich

1.89 Plan of the site with location of caves

67

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

1.90 Aerial view of the site facing north, with location of some of the caves

oriented eastward, toward the main building, and this also seems to have been the case in antiquity.

ceiling of the cave. In the lower part of the western wall there is an opening to a burrow which was covered by a heap of rubble and earth, which we removed while digging the burrow and Cave D (see below).

The ceiling of the cave is composed of a layer of nari 1.5–2 m in width, which is prominent at the site and in its vicinity. The inner ceiling was hewn in an almost even level, with only slight variations. The collapsed space leads to a second space, still roofed, which seems to have been the central room of the cave. The entrance to this room is some 5.5 m in width and about 2 m in height. The overall length of this room is 13 m from the entrance to the opening of the hiding burrow (below). The width of the room narrows from the outside inward. The eastern part is 10 m wide, the central part about 6 m, and the inner western side of the cave is only 3 m wide.

Methods of Quarrying in Cave G While studying Kh. er-Rasm and its installations, we also examined the methods of quarrying. Although our study focuses only on one cave in the Judean Shephelah, we know that it reflects technologies and tools which were in use all over the Land of Israel in antiquity. As described above, there are different levels of quarrying in the cave, which are a result of different methods of cutting the stone, or of different phases of quarrying (Safrai and Sasson 2001:7).

The Quarrying of the Ceiling

In the northeastern corner of the cave the ceiling is cut at a right angle, forming a square room in this part of the cave. The floor of the cave is filled with sediment fill, ash and stones. Along the northern and western walls of the cave, in the western part of the cave and on the present floor level, there are remains of a wall built of field stones in dry masonry, measuring 1 meter in width and 0.5 meter in height. Another section of wall, which is preserved to a height of 1.2 m, lies near the entrance in the northeastern part of the cave. These walls seem to be traces of shepherds’ activity in modern times, as are the ashy floor and sooty

Description and research of hewing cave ceilings has not been carried out as yet, except for the study done by Sasson (2002) in the Zedekiah Cave. The comparatively leveled ceiling of the cave testifies to the quarrying of the ceiling along a horizontal line in order to cut stone blocks. There are quarrying marks in the ceiling in the shape of elongated grooves every 1.5 to 6 cm. These grooves are about 5.5–7 cm long, 0.3–0.5 cm thick, and 0.5 cm deep. They testify to the use of a pickaxe or shovel, which leave

68

Avraham Faust and Adi Erlich

1.91 Plan of Caves G and D and the burrow connecting them

69

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

1.92 Section of Caves G and D and the burrow connecting them

1.93 General view of Cave G facing southwest

1.95 Quarrying marks on ceiling of Cave G

1.94 Entrance to the main hall of Cave G

1.96 Quarrying marks made with a shovel in Cave G

70

Avraham Faust and Adi Erlich

such marks (Ben Arieh 1970; Safrai and Sasson 2001:40). Judging by the grooves, the quarrying orientation which is indicated by the deep part of the groove is from south to north, generally speaking from the entrance inward. In certain places, especially in the inner part of the cave, there are also some grooves that cross the general south-north lines. They indicate stones which were extracted from all four sides (Figs. 1.95, 1.96).

The Quarrying of the Cave’s Walls In the inner part of the cave (from a distance of 4.5 m from the current entrance and deeper) one can identify arched quarrying marks that were made with a shovel on the upper part of the walls. This technique resulted in wave-like levels on the rock surface, each wave measuring 5–7 cm in width and 2 cm in depth, in round-diagonal lines. These waves indicate that the work was done from the top downward, from the ceiling and down to about a meter from it (Figs. 1.97, 1.98).

1.97 Arched quarrying marks in Cave G

A second level of quarrying was traced one meter from the cave’s ceiling and downward. There are signs of one tool, a relatively thick chisel (Safrai and Sasson 2001:23). The grooves are 2–3 cm long, 1 cm wide and 0.5 cm deep. On a lower level on the eastern wall, there are signs made by a quarrying peg and chisel, arranged in a line of 6 drilled holes 3 cm diameter in depth. These holes are located 1.2 m above the cave floor. In our opinion, such holes, which are found elsewhere in the cave, testify to the use of a long quarrying peg in order to extract the stones (Safrai and Sasson 2001:39, fig. 35) (see Figs. 1.99, 1.100). On the walls in the central part of the cave there are several rows of rock-cut steps, probably indicating quarrying levels. The height of each quarrying level is about 0.4 m, what points to relatively small cut stones. These quarrying lines are accompanied by marks made by tools similar to those mentioned above. The rounded and diagonally drilled holes were made with a quarrying peg, and were drilled from the top down at an angle of 60 degrees (Figs. 1.101, 1.02). On the outer façade of the cave, there were trace marks of a pointed chisel which left holes about 0.5 cm wide. The grooves are 2–5 cm long and 0.5 cm deep. The gaps between the grooves are large, 3–7 cm. The quarrying was done from the top down under the layer of nari. These signs represent the peeling of the chalk layer just below the nari with a chisel or a pole. There are natural cracks in the rock along the walls, which were probably identified by the stone cutters and used as quarrying channels for extracting stones.

1.98 Drawing of arched quarrying marks in Cave G

Nine triangular niches are cut in the walls, four in the eastern and five in the western wall. These niches, which were probably used for oil lamps, measure 17–22 cm in width, 13–25 cm in height and 10–13 cm in depth.

71

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

1.99 Drilled holes in the eastern wall in Cave G

1.101 Grooves and drilled holes in the eastern wall of Cave G

Graffiti On the eastern façade wall, to the right of the entrance, there are graffiti similar to those of the columbarium (below). One of them is almost completely preserved. It shows a rectangular frame divided into a grid of squares. There seem to have been 7 x 5 squares in each graffito originally. Unlike other graffiti to be described below, there are no drilled holes in these. At this point of the research we are not sure about the meaning or function of these graffiti, but judging by their location and the patina covering them, we believe they were incised during the quarrying of the cave (Figs. 1.103, 1.104). The original purpose of Cave G is uncertain. According to its shape, this is not a typical ‘bell-shaped cave’, which was common in the Judean Shephelah and used as a cistern, among other things. We assume that this cave served originally as a quarry, like many other caves in the region.

Cave D (Fig. 1.105; see also Fig. 1.91) The southern cave (Area D), is a collapsed cave with two entrances formed by the collapse. The collapsing of the central space makes it difficult to document and study this subterranean system. The eastern slope leads to a rectangular room with straight walls. This room measures 12 x 4 m, but originally, before the collapse, it was larger. A supporting rectangular pier was cut in the eastern part of the room, behind which there is a smaller room, measuring 3 x 4.5 m. In the southeastern corner of the small room as well as the northeastern corner of the large room there are some finely-cut round niches, which are typical of olive presses in the vicinity of Maresha. The particular shape of this cave, together with the round niches, point to the possible original function of this space as an olive press. The collapse of the central hall does not allow us to determine the purpose of these rooms with certainty.

1.100 Drawing of drilled holes in the eastern wall in Cave G

At the eastern end of the northern wall of the large 72

Avraham Faust and Adi Erlich

1.102 Drawing and section of grooves and drilled holes in the eastern wall of Cave G

1.103 Graffiti on the façade wall of Cave G

1.104 Drawing of graffiti on the façade wall of Cave G

73

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

1.105 The entrance to Cave D facing northwest

1.106 The entrance to the hiding burrow from Cave G

74

Avraham Faust and Adi Erlich

1.107 The entrance to the hiding burrow from Cave G

1.108 The entrance to the hiding burrow from Cave D

75

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

1.109 Interior of hiding burrow

rectangular room is the southern entrance to the burrow, connecting Caves D and G (below). The western half of this room was excavated. It contained topsoil (L.702), an occupation level which probably served shepherds in recent times or earlier and contained nari and kirton stones (L.703), and packed earth on the kirton floor which contained sherds and bones (L.705, L.706). L.701 is the cleaning of a niche on the western side of the collapsed cave, as it was suspected of being another burrow but turned out to be a small niche, perhaps of an olive press.

entrance, which is relatively unhidden, can be locked from the inside with a flat block. The western entrance is more hidden, and is cut slightly above the floor of the rectangular room. The tunnel was cleaned (L.700), but no finds have been revealed. The burrow was probably looted in the past. The burrow described above is typical of the hideout systems discovered all over the Shephelah, dated to the Great revolt or the Bar Kokhba revolt (Kloner and Tepper 1987; Kloner and Zissu 2003). It clearly postdates the two subterranean complexes it connects – Caves D and G, which are probably Hellenistic.

The Hideout Burrow (Figs. 1.106-1.109)

1.3.2 Cave F: The Columbarium (Fig. 1.110)

The burrow is cut between Cave G in the north-east and Cave D in the southwest, and connects the two (Figs. 1.106–1.109; see also Figs. 1.91, 1.92). The entrance to the burrow from Cave G was covered with a rubble heap. The stone and earth were removed as L.704. The locus contained a Gaza Ware sherd, and a kirton flat side of an object, perhaps an ossuary. The burrow connecting the two caves is approximately on a north-south axis. It is about 4.5 m long and 0.5 to 0.8 high. The southern half is lower than the northern half, with a step connecting the two parts. This is probably due to the cutting of the tunnel from both Cave D and Cave G, resulting in a difference of height at the meeting point of the two sections, or it may have been done on purpose for security reasons. Next to the eastern entrance there is a curving in the cave’s outlines. There are small niches for lamps along the walls. The eastern

The columbarium is situated southwest of Area C, hidden between bushes and a large tree. The cave was discovered by Roi Porat during a survey made in 1998. The entrance to the cave today is from the south. Part of the cave’s nari ceiling collapsed, resulting in a sloping ramp leading to the cave, probably on the location of the original entrance. The original entrance could have been through a spiral staircase (Fig. 1.111). On the left side (south) of the entrance there is a wall, which seems to be a later addition (Fig. 1.112). The cave is cut in the soft kirton rock, leaving a nari layer of 1.5 to 2 m above as roofing. The cave is round in shape, measuring 15 m on the north-

76

Avraham Faust and Adi Erlich

1.110 Plan and section of Cave F

77

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

1.111 Cave F facing north

1.114 Western pillar in Cave F

1.112 Built wall at the entrance of Cave F, facing south

1.115 Kirton block in Cave F

1.113 Pillars in Cave F

78

Avraham Faust and Adi Erlich

1.116 South east pillar in Cave F

1.119 Small niche in the western wall of Cave F

1.117 Arcosolium-like niche in the western wall of Cave F

1.120 Columbarium niches in Cave F

1.118 Arcosolium-like niche in the western wall of Cave F

1.121 Quarrying marks made with a thick chisel on the eastern wall of Cave F

79

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

1.122 Drawing of quarrying marks made with a thick chisel on the eastern wall of Cave F

south axis and 14 m on the east-west axis. The height of the cave is 3 m, but it is not the original height as the floor is filled with collapses and erosion to a probable height of at least one meter. The cave ceiling is supported by three large piers and another engaged pilaster. They are asymmetrical and are oval in section (Fig. 1.113). The one at the west, near the entrance, was cut at a later stage, perhaps for masonry stones, resulting in a narrow neck at its upper part (Fig. 1.114). A large kirton block lying next to the northern wall of the cave is perhaps the missing piece of the entrance pier (Fig. 1.115). Kirton blocks 0.5 m long were cut on the eastern side of the south-east pillar (Fig. 1.116). The piers seem to be a planned feature in the cave. They were left in place when the cave was quarried, as attested by the shovel marks on the eastern side of the ceiling, approaching the piers and ending there. The large size of the cave and the piers supporting it might suggest that the cave originally served as a quarry.

one cm deep and 5–7 cm long (similar to those in Cave G). It seems that this tool was used after peeling the first layer of rock. This tool left diagonal marks, unlike the work of the thick chisel, which left its vertical marks on the upper part of the walls. Generally speaking, whenever the cutters had to peel off layers of rock they used thick flat blades, while in places where fine peeling was required, such as columbarium niches, the cutters used shovels and chisels with pointed blades. On the western side of the cave, left of the ramp leading into it, there is an arcosolium-shaped large niche (Figs. 1.117, 1.118). The arcosolium was cut with a shovel and chisel with a narrow pointed blade. This arcosolium is late to the first stage of quarrying, as attested by the light colored patina covering the niche, unlike the dark patina framing the niche. The arcosolium is 2.8 m long, 0.9 m deep and 1 m high. On its southern side there is a small niche for an oil lamp, while on its northern side a small basin, perhaps for collecting water, was hewn, measuring 25 cm in diameter and 15 cm deep. The floor of the arcosolium was leveled with a layer of crushed kirton. Another irregular niche is cut just north of the arcosolium. Further north on the western wall there is a niche which was cut while using a natural crack in the wall. Two blocks of stone were cut out of it, measuring 0.9 and 1.2 m in length (Fig. 1.119). Another two smaller niches were cut on the same wall. Some parts of the walls were also peeled off, exposing the fresh white kirton rock.

The ceiling was cut on a comparatively even level, as a result of quarrying along a horizontal layer of the rock. The stone cutters followed the natural cracks between the layers of rock, and used them for quarrying and cutting detachment channels, in what we refer to as ‘upper quarrying’ or ‘ceiling quarrying’. The ceiling is partially collapsed and covered with black soot where preserved. Consequently, the quarrying marks in the ceiling were not always preserved. On the lower part of the cave wall, the work of a chisel with a pointed blade could be traced, which left grooves

On the northern wall of the cave there are traces of pigeon

80

Avraham Faust and Adi Erlich

niches, most of which were peeled off with the wall itself. On the east and south walls the pigeon niches are complete, and arranged in four rows (Fig. 1.120). There are 45 preserved niches on the southern and eastern wall, and 20 more niches on the northern pilaster. The erosion of the rock prevents us from determining the tools which were used for the quarrying of these walls. On the eastern wall there are still visible marks attesting to the use of a thick chisel, with a blade 2.5 cm wide and 5–10 cm long, for rough quarrying (especially near the entrance) (Figs. 1.121, 1.122). Perhaps these are signs of a long crowbar. These marks predate the cutting of the columbarium niches. Similar marks 2 cm thick which were made with a thick blade chisel are visible on the southern part of the eastern columbarium wall (Figs. 1.123, 1.124). The relationship between the thick quarrying marks and the niches indicates that the columbarium was cut after the quarrying of the original cave. In addition to the columbarium niches which are typical of the cave, there

1.123 Quarrying marks made with a thick chisel on the eastern wall of Cave F

1.124 Drawing of quarrying marks made with a thick chisel on the eastern wall of Cave F

81

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

1.126 Graffito 1 on the western wall of Cave F

1.125 Fissure on the eastern wall of Cave F

1.127 Drawing of graffito 1 on western wall of Cave F

82

Avraham Faust and Adi Erlich

1.129 Clay plates on the floor of Cave F

Next to the southeast pillar there are remains of clay plates 5 cm thick (Fig. 1.129). They are composed of mud and straw flattened and dried in the sun. Such plates were common during the Ottoman and Mandate periods, when they were used for building small walls for sealing small niches and storage vessels, like small storage bins or barrels for storing cereals, neither of which were found in the cave. The clay plates testify to the active use of the cave in modern times either for residence or for shepherds and their flocks (cf. Avitsur 1975; Hirschfeld 1995:135–137, figs. 90–91; Sasson 2003:24, fig. 7). These layers of clay also testify to the presence of shepherds in the cave, as some of these plates are laid closely on the cave floor.

1.128 Graffito 2 on western wall of Cave F

are also niches similar to those in Cave G which are used for oil lamps, 12 cm wide, 15 cm high and 10 cm deep. To the right of the engaged pilaster on the eastern wall there is a natural crack descending from the ceiling to the bottom of the cave (Fig. 1.125). This crack was treated by the users of the cave: the two meters of the bottom of the crack were widened up to 10 cm and deepened up to 5 cm, and then covered with plaster. This is the only crack which was treated that way, perhaps indicating the use of the cave as a cistern at a certain stage. As we shall see below, such treatment is typical of cisterns in the Judean Shephelah and it is also to be found in Cave J.

The cave shows different stages of usage, typical of caves and subterranean complexes in the Judean Shephelah. It was probably cut as a quarry for kirton bricks, probably to build the inner walls of the main building, during the Hellenistic period (see Chapter 4.1). Thereafter the cave was used as a cistern, as indicated by the plastered crack. The next use of the cave, probably still in the Hellenistic period, was as a dovecote. That this usage of the cave as a dovecote is only secondary, is attested by the partial use of the walls for pigeon niches. The last use of the cave was in modern times, as part of the surrounding of the village Ajur, when the cave was used by shepherds and fellahs. As the cave has not been excavated, the dating of all stages is circumstantial and based on the survey only (and see Chapters 4.3, 5.2).

Above the niche cut in the western wall of the cave, two graffiti are depicted on the wall. The first one (Figs. 1.126, 1.127) is engraved 1.6 m above the present floor level, and is 19 cm high and 17 cm wide. It depicts a rectangle divided into five vertical sections. Each rectangular section inhabits two round holes one on top of the other, the upper one framed by another small rectangular. The central section is wider and its frame exceeds the general frame of the whole composition. The second graffito (Fig. 1.128) depicts a few geometrical frames one on top of the other, perhaps meant to portray a schematic human figure with two short legs, similar to human figures engraved on walls of caves from Maresha (Bliss and Macalister 1902:242, fig. 90). Some other unclear graffiti are engraved on the wall. The meaning of these graffiti is still vague. They could refer to the process of quarrying the cave, or perhaps to a cultic use, as the graffiti are depicted above the arcosolium.

1.3.3. Additional Subterranean Complexes

Subterranean Complex H (Fig. 1.130) Subterranean complex H lies under the northern part of the main building, on a west-northwest to east-southeast axis. The system was not excavated due to safety problems: the ceiling has collapsed or is in danger of collapse in some of the rooms, and furthermore, the cave is infested with

83

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

1.130 Axis of Cave H underneath the main building

1.131 Western entrance to Cave H (west of the main building)

84

Avraham Faust and Adi Erlich

1.132 Collapse of Cave H in courtyard 20, facing west

1.133 Collapse of Cave H east of the main building (east of Square D51)

85

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

1.134 Entrance to Cave I facing south

relapsing cave fever ticks. We could partly identify the rooms of the subterranean complex below the western half of the building; we could only identify an opening, resulting from collapse, in the eastern part, just outside the eastern wall of the building, east of square D51. The original entrance to this system has not been found for certain, and all the entrances identified seem to be a result of collapse. One entrance is situated to the west of rooms 12 and 13 of the main building (Fig. 1.131). It leads to subterranean halls which are partially collapsed and filled with sediment or deliberate fills. Another opening which is a result of a collapse was found in square I49, and another entrance, which was cleaned of rubble and stones, is situated in the northwest corner of the courtyard (space 20), in square G50 (L.809). A built wall to the west of this entrance, as well as its location in the courtyard of the building, may indicate that it was the original entrance which collapsed (Fig. 1.132). Outside the building, to the east of square D51, there is a large depression, which is probably another collapse of the cave (Fig. 1.133).

(Kloner 2003:20). Similarly, subterranean H relates to the main building in its layout, axis and entrances, which either flank the building or are located in its courtyard, and apparently served it.

If indeed the original entrance to the cave has been identified in the courtyard, this allows us to clarify the relationship between the building and the subterranean complex. In Hellenistic Maresha, the caves are related to the houses or streets above and they were hewn in order to serve them

Cave I is situated about 20 m north of the northeast corner of the main building, near the present JNF ‘wells path’ (Fig. 1.134). The cave is filled with sediment, erosion and collapses, which do not allow for a thorough survey. The cave is hewn as a large round hall measuring 7 m in

The cave, however, seems to postdate the original building. The subterranean caves were probably dug below the original nari-built building at the time of phase 2b, and the rock material quarried out of subterranean system H (and other caves around the site) supplied the kirton stones for the walls of this phase (see Chapter 4.1). We do not know what purposes this system served. The olive press weight, found out of context in room 12 of the building, not far away from one of the collapses, may suggest the usage of one of the rooms of this cave, but it is likely that the subterranean system under the main building, which resembles other systems in the Shephelah, probably served various purposes.

Cave I

86

Avraham Faust and Adi Erlich

1.135 Plan and section of Cave I

87

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

1.136 General view of Cave I

1.137 A niche in the southern wall of Cave I

88

Avraham Faust and Adi Erlich

1.138 Small burrow in the southwestern wall of Cave I

1.139 Opening of cistern J

89

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

1.141 Cistern J from inside

1.140 Plan and sections of cistern J

1.142 Troughs near the entrance to cistern J

diameter. A dromos-like entrance on a north-south axis descends below the layer of nari (Figs. 1.135, 1.136). The maximum height of the cave in its inner southern part is 3 m. On the southern and eastern walls two large niches were cut, probably for storage purposes. The niche on the southern wall is 1.5 m long and it has remains of a parapet 20 cm high. Its floor is covered with crushed kirton. Next to this cell there are a few small triangular niches for oil lamps. Below this cell the cave was broadened, but the fill of soil and stones does not allow a further examination (Fig. 1.137). On the eastern wall there is another installation which probably served as a stall, measuring 0.6 x 1.5 m,

and its walls are 10 cm thick. There are some other niches for daily use in the cave walls. On the southwestern side of the cave there is a small burrow which was probably dug by animals, but could nevertheless be an opening to a hideout burrow (Fig. 1.138). The cave shows the use of the normal quarrying tools that are also indicated in other caves at the site, such as shovels, chisels, long bar-type poles, etc. The pole has left quarrying marks of 4 cm wide. There is no evidence such as plaster and other typical installations for the use of the cave as a cistern or reservoir. Therefore, this cave seems to have been hewn

90

Avraham Faust and Adi Erlich

1.143 Plan and section of troughs near the entrance to cistern J

91

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

originally as a quarry, and later on it was used by shepherds and fellahs in the vicinity. A high stone parapet which changes the original landscape was built by the JNF during the development of the British Park.

the water from organic infestations. Sediments of erosion, soil and stones cover the cistern floor. On its walls there are fern plants. We did not detect any recess for collecting the sediment, as is typical of cisterns from all periods. However, south of the cistern opening, on the natural rock surface, there are two small depressions where the water flowed into the cistern. Perhaps these depressions served for collecting sediment. Northeast of the cistern’s entrance there was a large boulder measuring 1.4 x 1.5 x 0.7 m, with two basins cut on its surface, serving as water troughs (Figs. 1.142, 1.143). The smaller trough is 10 cm deep and holds about two liters, while the larger trough is 25 cm deep and holds about five liters. The large trough is drained by a channel 30 cm long. This is uncharacteristic for troughs, and it is possible that this basin served other purposes, perhaps for producing liquids or juice from fruits. Recently the cistern was covered with a metal net by JNF workers.

Cave J (the Cistern) The cistern is situated 50 m northwest of the main building, in the upper part of a rocky slope, bordering the JNF wood (Fig. 1.139). The quarrying of the cistern began with a rectangular depression in the nari layer, directed north-south, and measuring 1 x 1.4 m, and is 30 cm deep. Thereafter the base of this recess was hewn further in most of its surface and deepened by 2.5 m, as if to be used as a cistern. The opening of the cistern eventually measured 0.7 x 0.9 m (Figs. 1.140, 1.141). Natural fissures in the rock surface were widened and plastered in order to keep the water from seeping into the rock. On the cave walls there are some small niches used either for oil lamps or for spawning cells for fish, which were useful for cleansing

1.4 Appendices 1.4.1 List of Loci Locus

Area

Square

Upper Level

Lower Level

Description

Room

100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128

A A A A A A A A A A A A A A A A A A A A A A A A A(E) A A A A

J50 J50 J51 J52 I52 J51 I52 J52 J53 J52 J52 J51 J51 I52 J52 J52 J52 I52 J53 J50 J52 J52 J51 J51 I48 J51 J51 J52 J52

328.84 328.84 329.02 329.02 329.68 328.84 328.89 328.91 329.20 329.68 328.68 328.09 328.65 328.76 328.67 328.34 328.24 329.38 329.01 328.44 328.54 328.53 328.43 328.43 328.59 328.91 328.38 328.11 328.13

327.84 328.26 328.71 328.35 329.05 327.91 328.45 328.37 328.61 328.34 328.34 327.90 328.43 328.33 328.50 328.11 328.09 328.99 328.61 328.10 328.45 328.29 328.35 328.25 328.18 328.25 327.87 327.88 327.97

Topsoil outside the structure Topsoil Topsoil Topsoil Topsoil Topsoil outside the structure Topsoil Topsoil outside the structure Topsoil Wall fall in room Wall fall inside installation Wall fall outside structure Wall fall inside structure Topsoil in courtyard Topsoil mixed with wall fall Wall fall inside installation Wall fall in room Topsoil mixed with heap of stones Topsoil Topsoil mixed with wall fall Wall fall south of wall W.19 Wall fall in corridor Wall fall south of W.16 Wall fall north of W.16 Topsoil Wall fall north of W.16 Wall fall south of W.16 Occupational debris inside installation Occupational debris outside installation

--------10 9, 10 5, 6, 9 7, 8 --------20 --------5 9 9 --------9, 10 20 5, 6 9 9 7, 8 5 10 5 6 9 10 13 10 9 9 9

92

Avraham Faust and Adi Erlich

Upper Level

Lower Level

J52 J52 J52 I48 J52 J52 J52 J53 I48 I48 J52 I52 I52 I48 H49 J52 I52 Canceled

328.53 328.64 327.88 328.46 328.02 328.39 327.87 328.94 328.50 328.05 328.31 328.99 329.09 328.87 328.60 327.94 328.96

328.29 328.29

A

I52

328.70

328.44

A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A

H49 H49 J53 I48 I52 H49 H49 I49 I50 J53 J53 J53 I48 H49 I48 I50 I49 I49 I48 J53 I49 I53 I53 Canceled I52 J53 J53 I52 I53 I53 152 I47 H54 I47 I47 I51 I53

328.41 328.59 328.84 328.71 329.35 328.60 328.11 328.64 328.64 328.33

328.17 327.72 328.22 327.49 328.37 327.92 327.94 328.17 328.27 328.01

329.08 327.06 328.13 327.66 328.43 328.23 328.35 327.53 328.07 328.02 329.39 329.46

328.07 327.00 327.53 327.08 327.65 327.89 327.88 327.40 327.54 327.53 329.05 328.55

Lower occupational debris and kirton wall fall north of W.22 Topsoil mixed with wall fall Fill and occupational debris Wall fall Wall fall Natural fill in courtyard Fill and occupational debris Fill and occupational debris Topsoil Topsoil Occupational debris and wall fall Mixed – baulk cleaning Occupational debris and wall fall Wall fall Fill and occupational debris Fill in eastern installation Fill Fill / wall fall Fill / wall fall Fill in western installation Virgin soil Fill / wall fall Topsoil Topsoil mixed with wall fall

329.77 328.98 328.65 328.75 328.82 329.89 329.44 327.68 329.64 327.05 328.51 329.92 328.88

328.60 328.53 327.72 328.43 328.40 328.17 328.10 325.96 328.36 325.96 325.96 328.71 328.41

Topsoil mixed with wall fall Topsoil out side the structure south of W.12 Natural fill outside the structure south of W.12 Upper occupational debris north of W.22 Wall fall and occupational debris Clean up of bulks and walls Clean up of bulks and walls Topsoil outside structure Topsoil Topsoil outside structure Clean up of bulks and walls Topsoil Wall fall

Locus

Area

Square

129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146

A A A A A A A A A A A A A A A A A A

147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184

327.71 327.84 327.82 327.70 328.52 327.49 327.49 328.20 328.95 328.94 327.61 328.19 327.80 328.70

Description

Room

Wall fall in corridor Wall fall south of wall W.19 Virgin soil inside installation Wall fall Virgin soil inside installation Occupational debris and wall fall in corridor Occupational debris and wall fall in corridor Topsoil Fill in eastern installation Fill in western installation Occupational debris and kirton wall fall south of W.19 Topsoil mixed with wall fall north of W.22 Topsoil Wall fall Topsoil Virgin soil between W.15 and W.19 Wall fall north of W.22

6 5 9 13 9 6 6 5 13 13 5 7 7, 8 13 12 6 7

93

7 12 12 5 13 20 12 12 12 10, 11 5 5 13 12 13 10, 11 12 12 13 5 12 4 4 8 ----------------8 4 4 8 --------------------------------11 4

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

Locus

Area

Square

185 186 187 188 189 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519

A A A A A B B B B B B B B B B B B B B B B B B B B B B B B B B B B B B B B B C C C C C C C C C C C C C C C C C C C

I52 I53 G54 H53 H53 G53 E54 G53 G53 G53 F53 G53 G53 G53 F54 F53 F53 F53 F53 G53 E54 G53 G53 G53 D51 E54 F51 G53 G53 E54 E54 G53 G53 G53 E51 E54 D51 D50 C5 C6 C8 C4 C4 C4 C8 C7, 4 C6, 9 C4 C1 C1 C3 C5 C2 C2 C2 C2 C2

Upper Level

Lower Level

328.51 328.62 329.53 329.62 328.88 329.65 330.20 329.38 329.45 329.09 329.58 329.96 329.65 329.92

328.07 328.17 328.31 328.76 328.78 329.28 328.48 328.98 329.19 328.95 328.75 329.01 328.95 328.75

328.99 328.85 328.60 328.68 328.99 329.71 328.72 328.81 328.63 329.00 329.05 328.72 328.58 328.49 329.33 329.40 328.80 328.46 328.40 329.32 328.74 329.06 328.88 327.95 325.77 325.77 327.40 326.84 326.07 326.90 326.58 327.07 326.55 326.80 327.43 327.67 327.23 327.85 326.98 326.61 326.52 326.44

328.64 328.62 328.34 328.53 328.54 328.69 328.54 328.72 328.50 328.77 328.41 327.95 328.42 328.30 328.70 328.82 328.40 328.13 328.08 328.17 328.47 328.70 328.45 327.46 324.47 325.47 327.00 326.02 325.69 325.95 325.62 325.50 325.62 326.25 326.50 326.80 326.90 327.16 326.45 326.44 326.44 326.28

Description

Room

Lower occupational debris and wall fall north of W.22 Wall fall and occupational debris Topsoil outside structure Topsoil Topsoil in niche of W.28 Topsoil Topsoil Topsoil Topsoil Wall fall? Topsoil Wall fall? Topsoil, removal of earth with ‘bobcat’ Wall fall? Topsoil - removal of debris east of W.9 Mixed, bellow topsoil Mixed, bellow topsoil Mixed, bellow topsoil Mixed, bellow topsoil Wall fall with fill Topsoil Wall fall with fill Wall fall with fill Wall fall with fill Topsoil Wall fall Topsoil Wall fall mixed with occupational debris Wall fall mixed with occupational debris Wall fall inside installation Wall fall Wall fall mixed with occupational debris Wall fall mixed with occupational debris Virgin soil Topsoil Tabun Mixed below topsoil Topsoil Topsoil, cleaning the structure Topsoil, cleaning the structure Topsoil, cleaning the structure Topsoil, cleaning the structure Topsoil, cleaning the eastern section Topsoil, cleaning the eastern section Topsoil, cleaning the structure Topsoil, cleaning the structure Topsoil, cleaning the structure Topsoil, east of the eastern section Topsoil on the northeastern corner of structure Topsoil on the northeastern corner of structure Topsoil, cleaning the structure Topsoil, cleaning the structure Topsoil, cleaning the structure Topsoil, north of the structure Topsoil, north of the structure Topsoil, north of the structure Topsoil, north of the structure

8 4 --------3 20 2 1 2 2 2 --------2 2 2 ----------------------------------------2 1 2 2 2 --------1 19 2 2 1 1 2 2 2 19 1 --------19 ---------------------------------------------------------------------------------------------------------------------------------------------------------

94

Avraham Faust and Adi Erlich

Locus

Area

Square

520 521 522 523 524 525 526 527 601 602 603 604 605 606 607 700 701 702 703 704 705 706 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 998 999

C C C C C C C C Z1 Z1 Z1 Z1 Z1 Z2 Z2 D D D D D D D E E E E E E E E E E E E E E E E E E E E E E E E E E E

C2 C2 C2 C8 C8 C8 C8 C6 G55 G55 G55 H55 H55

F49 G48 H48 E49 F49 F49 F49 F49 G50 G48 F49 Cancelled H48 H48 E49 E49 F49 G48 E49 D49 H48 E49 H48 G48 E49 E49

Upper Level

Lower Level

327.56 326.39 326.50 326.79 326.84 325.70 326.06 326.81 329.10 328.50 328.13 328.29 327.86 325.32 324.93

326.31 326.09 326.03 326.06 325.92 324.76 325.08 325.93 328.45 328.41 327.75 327.65 327.32 286.22 323.70

320.20 319.88

319.88 319.31

319.31 319.03 328.91 328.91 328.74 328.99 328.61 328.60 328.55 328.40

319.03 318.97 328.21 328.22 328.27 328.50 328.30 327.91 327.75 327.85

328.51 328.38

327.66 327.86

328.39 328.22 328.66 328.80 327.97 327.89 328.73 328.98 327.81 328.08 327.49 327.60 327.94 327.57

327.56 327.56 327.98 327.82 327.53 327.48 327.79 328.51 327.36 327.74 327.21 327.33 327.60 327.57

Description

Room

Topsoil, north of the structure Virgin soil, north of the structure Virgin soil, north of the structure Topsoil, south of structure Topsoil, cleaning the structure Collapsed stones on the south side of square Collapsed stones along south wall of structure Topsoil along east wall of structure Topsoil Topsoil of eastern section, north side Topsoil of eastern section, south side Fill of western section, north side Fill of western section, south side Fill north of terrace wall Fill south of terrace wall Cleaning of hideout tunnel Fill of niche on west side of cave Topsoil of cave Upper occupational level Cave G, rubble heap at the entrance to burrow Fill in cave Fill in cave Topsoil Topsoil Topsoil Topsoil Wall fall in cell Wall fall east of cell Wall fall north of cell Fill in cell Topsoil, removal of boulders from entrance to Cave H Wall fall Wall fall south of cell

--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------16, 18 15 14 16, 19, 20 18 20 16 18 20 15 20 --------14 14 16 20 16 15 19 17 14 16 14 15 16 16 -----------------

Wall fall Wall fall Wall fall at northern part of square Wall fall at southwest part of square Occupational debris with ash layer north of cell Occupational debris with ash layer Wall fall, southeast part of square Topsoil Occupational debris Occupational debris with ash layer Ash layer Ash layer Occupational debris with ash layer Fill below floor, coin by metal detector Surface allover Area E Surface throughout the site

95

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

1.4.2 Locus / Baskets List Area A Locus

Baskets

Locus

Baskets

100 101 102 103

1000, 1006, 1010, 1017 1001, 1014, 1018, 1025 1002, 1015, 1019, 1029 1003, 1008, 1012, 1023

145 146 147 148

104

1004, 1009, 1013, 1022, 1026

149

105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124

1005, 1007, 1011, 1027 1016, 1022 1021, 1028 1024 1030 1031 1032 1033 1034, 1042, 1050, 1059 1035 1036, 1044, 1056 1037, 1045, 1055 1038 1039, 1046, 1051, 1058 1040, 1041, 1052 1049 1043, 1057 1048, 1054 1047, 1053 1059, 1066, 1070

150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169

125

1060

170

126 127 128 129 130 131 132

1061, 1067, 1071, 1078, 1082, 1087 1062, 1068 1063 1064, 1069, 1073, 1074 1065, 1084, 1085, 1086, 1088 1072 1075, 1077, 1081

171 172 173 174 175 176 177

133

1076, 1079

178

134 135 136 137 138 139 140 141 142 143 144

1080, 1083 1089, 1095 1090, 1099, 1104, 1108 1108, 1112, 1123, 1132 1100, 1116, 1131, 1145 1093, 1096, 1105, 1124, 1125, 1127, 1130 1094 1098 1103, 1122 1101, 1105, 1109 1107, 1113

179 180 181 182 183 184 185 186 187 188 189

1097, 1106 Cancelled 1110, 1114, 1121, 1129 1111, 1115 1117, 1134, 1135A, 1135B, 1138, 1141, 1143, 1144, 1149 1118, 1128, 1139, 1142 1133 1126 1136, 1140 1137 1150, 1153, 1157 1151, 1154, 1158 1152, 1163, 1168, 1173, 1179 1155, 1193 1156, 1162 1160, 1171, 1174, 1180, 1188, 1196, 1199 1159, 1166, 1167, 1170, 1190 1161, 1177, 1181 1164, 1169, 1175, 1183 1165, 1172 1176, 1182, 1186 1178, 1184, 1195 1185, 1189 1187, 1192, 1194 1191, 1197, 1198 1200, 1201, 1204, 1205, 1208, 1212, 1213, 1214 Cancelled 1202, 1207, 1209, 1211, 1216, 1217 1203, 1206, 1210, 1215 1218, 1222, 1225, 1233 1219, 1232, 1240, 1245, 1253 1237, 1243, 1255 1220, 1221, 1226, 1229, 1236, 1242, 1274 1223, 1224, 1227, 1228, 12231, 1235, 1239, 1241, 1277 1238, 1251 1244, 1260 1247, 1252 1248 1249, 1263, 1265, 1269 1230, 1234, 1246, 1250 1256, 1359, 1262, 1272, 1276 1257, 1258, 1261, 1264, 1268, 1270, 1273 1266, 1267 1271, 1275 No baskets

96

Avraham Faust and Adi Erlich

Area B Locus

Baskets

Locus

Baskets

301 302

3001, 3003, 3005, 3007 3002, 3004, 3006, 3009

318 319

303

3008, 3010, 3012

320

304

3011, 3013

321

305 306 307 308 309 310 311 312

3014 3015, 3016, 3019, 3023 3017 3018 3020, 3021, 3022, 3028 3024 3026 3025, 3027

322 323 324 325 326 327 328 329

313

3030, 3035, 3039

330

314 315 316 317

3029 3031, 3037, 3038, 3041 3032, 3033, 3034, 3036, 3040, 3043 3042

331 332 333

No baskets 3045 3044, 3047, 3055, 3060, 3065, 3069, 3071, 3075, 3084, 3090, 3096, 3105 3046, 3048, 3054, 3059, 3061, 3062, 3077, 3078, 3085, 3094, 3100, 3104, 3107, 3112, 3117 3049, 3067, 3070, 3074, 3080, 3081, 3092 3050, 3053, 3058, 3064, 3068 3072, 3079, 3089 3051, 3056, 3063, 3086, 3088, 3099 3052 3057, 3066 3073, 3076 3082 3083, 3091, 3095, 3102, 3106, 3110, 3111, 3113, 3115, 3119 3087, 3098 3093, 3097, 33103, 3122 3108, 3109, 3114, 3118

Area C Locus

Baskets

Locus

Baskets

501 502 503 504 505 506 507 508 509 510 511 512 513 514

5001, 5002, 5003 5004 5005 5006 5007, 5008, 5009 5010 5011, 5029 5012, 5015, 5028, 5030, 5032 5013, 5014, 5018, 5019, 5024, 5034 5017, 5021 5016, 5020, 5023 5022, 5026, 5031 5025 5033

515 516 517 518 519 520 521 522 523 524 525 526 527

5036, 5038, 5039, 5041, 5044 5034, 5035, 5037, 5040, 5042, 5045 5043, 5046 5047, 5050 5048, 5051 5049 5052, 5054 5055 5056, 5057, 5058, 5059, 5053 No finds 5062 5063 5061

Area D Locus

Baskets

Locus

Baskets

700 701 702 703

7000 7001 No finds 7003

704 705 706

7002 7004 7005

Area Z Locus

Baskets

Locus

Baskets

601 602

No finds 6001

605 606

603

6004

607

6006, 6009, 6013 6016 6017, 6018, 6019, 6020, 6021, 6022, 6023, 6024

604

6005, 6008, 6012, 6014, 6015

97

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

Area E Locus

Baskets

Locus

Baskets

801 802

8001, 8004, 8007 8005, 8008, 8012, 8016, 8024 8003, 8006, 8009, 8011, 8015, 8025, 8034, 8035 8010, 8030, 8041, 8049, 8058 8013 8014, 8017, 8021, 8033, 8037, 8045 8018, 8022, 8031, 8038, 8046, 8056, 8059, 8262, 8066 8019, 8023, 8032, 8039, 8047, 8057, 8068, 8079 8020, 8026, 8029, 8043 8027, 8036, 8044, 8055, 8060, 8065 8028, 8040, 8048 Cancelled 8050, 8054, 8064, 8071

814 815

8053, 8063, 8070, 8076 8061, 8067, 8074, 8080

816

8069, 8075, 8081, 8087, 8095

817 818 819

8073, 8078 8072, 8077, 8085, 8090, 8093 8082, 8088, 8096

820

8083, 8089, 8097

821

8084, 8092, 8098

822 823 824 825 826

8086, 8094 8101 8100 8099 8102

803 804 805 806 807 808 809 810 811 812 813

1.4.3 List of Walls (PH = maximum preserved height) Wall

Area

Squares

Phase

Description

Notes

W.1

A

J50, J51, J52, J53, K53, I48, I49

1

Nari; 100 cm wide; two rows, up to 3 courses; 120 cm PH

Western wall of complex

W.2

A, E

I47, G48, H48

1

Nari; 100 cm wide; 2 rows; 4–5 courses; 140 cm PH

Northern wall of building; Northern wing of large stones in the western part becomes one row of larger stones

W.3

A

I51, I52

1

W.4

A

W.5

B, E

W.6

I49, I50, I51, J52 E49, E51, F53

1 (likely)

A

G53, G54

1a / 2c

W.7

A

G54, H54

1

W.8

B

E53

1

W.9

B

W.10 W.11

A

W.12

A

I53, J53, K53

1

W.14

B

G53

1a

W.15

A

J52

2b

W.16

A

J51

2b

E53, F53, F54 E51, E52

1

1 1

W.13

Nari; 50 cm wide; 1 row; 2–3 courses; 130 cm PH Nari monoliths; 60 cm wide; 1 row; 135 cm high Nari and some kirton; 60–100 cm wide; 1 row; 1–2 courses; 60 cm PH 1 monolith, nari and kirton; 100–120 cm; 2 rows; 2 courses; 50 cm PH at kirton blocks Nari; two rows; only outer side partially cleaned Nari; 100 cm wide; 1 row; 4 courses; 170 cm PH Nari; 90–100 cm wide; 1–2 rows; 3–4 courses; 160 cm PH unexcavated Cancelled Nari; 1.05 m wide, 2 rows; 3 courses; 80 cm PH Cancelled Nari; 130 cm wide; 2 rows; 3 courses; 145 cm PH Kirton; 50 cm wide; 1–2 rows; 2–4 courses; 70 cm PH Kirton; 20 cm wide; 1 row; 1–2 courses; 35 cm PH

98

Abutted by W.22–W.23 room 7 Row of monolithic stones in western wing Inside wall on eastern wing of complex, parallel to W.1 and W.3 East wall of room 2 Section of southern wall of complex (continuation of W.12) North wall of room 1 Eastern wall of the complex entrance (west wall of room 1) Row of monoliths in eastern wing Southern wall of complex (continuation of W.7) Western wall of room 2 Between W.1 and W.4, north wall of room 6 Northern wall of room 9

Avraham Faust and Adi Erlich

Wall

Area

Squares

Phase

W.17

A

J52

2b / 2c

W.19

A

J52

1a

W.20

A

I48

1

W.21

A

J52

1

W.22

A

I52

1a

W.23

A

I52

1a

W.24

A

J52

3

W.25

A

I48

2a

W.26

A

I48

2a

W.27

B

E54, D51

1

W.28

B

G53

1

W.29

B

E54

1

W.30

A

K53 / 54

2a

W.31

E

F49

1

W.32

E

F49

2a

W.33

E

G48

1a

W.34

E

H48

1a

W.35

A

I53

1a

W.36

B

E54

2b / 2c

W.37

Z1

W.38

A

I47

2a

W.39

E

F49

2a

W.40

E

F49

2a

W.41

E

E49

1a / 2a / 2b

W.42

E

E49

1

W.43

B

E54

2b / 2c

W.44

E

H48

1a

W.45

Z2

W.46

AE

H48

3

W.47

A

I52

3

W.48

A

I52

3

W.49

A

J52

2c

W.18

1

Description

Notes

Kirton; 20 cm wide; 1 row; 1 course; 50 cm PH Cancelled Nari; 80 cm wide; 2 rows; 2–3 courses; 95 cm PH Nari; 0.90 m wide; 1 row; 2–3 courses; 60 cm PH Nari; 40 cm wide; 1 row; 1 course; 30 cm high Nari; 40 cm wide; 1 row; 2–3 courses; 95 cm PH Nari; 75 cm wide; 1 row; 2–3 courses; 85 cm PH Nari; 40 cm wide; 1 row; 3 courses; 100 cm PH Kirton; 20 cm wide; 1 row; 1 course; 60 cm PH Nari; 40 cm wide; 1 row; 2–3 courses; 80 cm PH Nari; 50–120 cm wide; 2 rows; 2–3 courses; 120 cm PH Nari; 105–125 cm wide; 2 rows; 3 courses; 150 cm PH Nari; 80–105 cm wide; 2 rows; 2 courses; 80 cm PH Nari; 33–46 cm wide; 1 row; 1 course; 55 cm PH Nari; 80 cm wide; 2 rows; 3 courses; 85 cm PH Nari; 60 cm wide; 1 row; 2 courses; 80 cm PH Nari; 80 cm wide; 2 rows; 3–4 courses; 125 cm PH Nari; 100 cm wide; 1 row; 2–3 courses; 120 cm PH Nari; 57–62 cm wide; 1 row; 3–4 courses; 107 cm PH Kirton; 20 cm wide; 1 row; 4 courses; 75 cm PH Nari; 100–120 cm wide; 2 courses; 100 cm PH Nari; 100–130 cm wide; 2 rows; 2–3 courses; 120 cm PH Nari; 60 cm wide; 1 row; 1 course; 30 cm PH Nari; 40 cm wide; 1 row; 1 course; 20 cm PH Kirton on top of nari; 100 cm wide; 2 rows; 2–3 courses; 100 cm PH Nari; 85 cm wide; 1 row; 3 courses; 110 cm PH Tabun; 24 cm PH Nari; 70–75 cm wide; 2 rows; 2–3 courses; 90 cm PH Nari; terrace wall Nari; 100 cm wide; 1 row; 2–3 courses; up to 120 cm PH Nari; 60 cm wide; 1 row; 3 courses; 120 cm PH Nari; 70 cm wide; 1 row; 2 courses; 85 cm PH Kirton; 50 cm wide; 2 rows; 2 courses; 40 cm PH

Western and northern border of installation, rounded wall

99

Northern wall of room 5 Southern wall of room 13 Runs below W.1 and the blockage of W.15 Northern wall of room 7 Southern wall of room 7 Blockage between monoliths of W.4, western wall of room 7 In room 13 divides cells b from c In room 13 divides cells a from b and c Eastern wall of complex Northern wall of room 2 Southern wall of room 1 External wall, going southward from the southwestern corner of the building Northern wall of room 18, south western wall of room 16 Southern wall of cell 18 Eastern wall of room 15 Western wall of room 14 Western wall of room 3 Installation wall in room 1 Wall south of main building External wall. Continues W.1 northward, outside of complex Eastern wall of cell 18 Western wall of cell 18 Western wall of room 17 South east wall of room 16 In room 1 East wall of room 14 Terrace wall south of Area C Irregular blockage of entrance to room 14 (between W.2 and W.34) Blockage of entrance to room 4 Eastern blockage of room 7 (W.3) Blockage of W.15

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

Wall

Area

Squares

Phase

W.50

A

J53

1a

W.51

E

D49

1

Description

Notes

Nari; 60 cm wide; 1 row; 3 courses; 85 cm PH Nari; 120 cm wide; 1 row; top only cleaned

100

Eastern wall of room 5 Continuation of W.42

Part II: Ceramics

Avraham Faust and Adi Erlich

2.1 The Iron Age and Persian Period Ceramic Finds

R-BLX4 L.322, B.3080/11

Itzhaq Shai

An upper part of a bowl with a flat rim, carinated in the middle of the body. This type was more common in Iron Age IIA, yet it continued in Iron Age IIB as well, for example at Tel Batash (Mazar and Panitz-Cohen 2001:40–41, pl. 24:9).

Twelve pre-Hellenistic indicative forms (Fig. 2.1) were discovered in Kh. er-Rasm (about 40 sherds in all).1 These limited ceramic finds were derived from various loci, many from topsoil or from outside the main building, but not a single sherd came from a clean Iron Age or Persian locus. However, the presence of such a diagnostic find may indicate that from Iron Age I (see below) through Iron Age II and the Persian period, there was some human activity at this site.

R-BLX5 L.174, B.1273/1 A rim of a Judean folded rim bowl. Traces of wheel burnishing are still visible on the rim and on the outside. This type was very common in Judah in the eighth century BCE (e.g., Zimhoni 2004:1793–1794, fig. 26.3:19).

The types below are described one by one, in the following manner: The type number (see Chapter 2.2 for detailed explanation of the numbering system), the locus and basket of the sherd drawn in the plate, a description of the form (including its date) and parallels.

R-BLX6 L.156, B.1159

R-BLX1

A rim of a carinated bowl, which was common in Judah in Iron Age IIB (e.g., Zimhoni 2004: fig. 26.27:17).

L.321, B.3048/7

R-BLX7

A rim of a bell-shaped Philistine bowl, with thin walls and a red band over the rim and another band inside and below the rim. The bowl was not slipped. This type of bowl typically dates to Late Iron Age I, as at Ashdod Stratum XI (Dothan and Ben-Shlomo 2005: fig. 3.46:1, 6).

L.132, B.1077 A rim of a carinated bowl with flanged rim, dated to Iron Age IIB (Zimhoni 2004:1793, fig. 26.3:14).

R-BLX2

R-CUPX1

L.158, B.1147

L.306, B.3019

A rounded bowl with a folded rim, slipped on the inside. This type is common in Stratum III in Tel Batash, although it appears also in Stratum IV and II (Mazar and PanitzCohen 2001:39, pl. 24:14), and at Lachish as well (Zimhoni 1997: fig. 3.66:3, 5, 7). It is dated to Iron Age II.

An upper fragment of an Attic Ware S-shaped cup. Fantalkin and Tal (2004:2187) date this type to the late fifth or early fourth century BCE, as at Lachish (Fantalkin and Tal 2004: fig. 30.8:7).

R-KRX1

R-BLX3

L.814, B.8076/3

L.804, B.8030/7

A fragment of a large decorated Philistine krater, with a white slip and a bichrome decoration, typical of the second stage of Philistine material culture (e.g., T. Dothan 1982). Similar vessels have been found at Iron Age I sites, particularly (although not exclusively) in Philistia, e.g., in Ashdod (Dothan 1971: fig. 1:8, 11).

A fragment of a carinated bowl with a plain rim. Stern (1982:95) dates this type of bowl to the sixth to fourth centuries BCE. A similar bowl was found at Gezer (Gitin 1990: pl. 46:3). 1 Following is the number of sherds that were identified during the initial counting by Débora Sandhaus, according to the serial number of the types as they appear in this chapter: R-BLX1: 1 sherd (L.321); R-BLX2: four sherds (L.150, L.158, L.306, L.804); R-BLX3: one sherd (L.804); R-BLX4: eight sherds (L.109, L.176, L.322, L.323, L.330, L.517, L.518, L.521); R-BLX5: one sherd (L.174); R-BLX6: one sherd (L.156); R-BLX7: one sherd (L.132) R-CUPX1: one sherd (L.306); R-KRX1: five sherds (L.147, L.508, L.814, L.819 [two sherds]); R-KRX2:two sherds (L.808, L.819); R-SJX1: 13 sherds (L.110, L.116, L.149, L.161, L.168, L.186, L.313, L.322, L.330, L.604, L.803, L.808, L.816); R-SJX2:two sherds (L.330, L.818). In addition to the likelihood that a few sherds were missed during the count, we should note that types R-BLX5–R-BLX7 were identified only later, and it is therefore likely that the number of sherds (of those types) was slightly higher.

R-KRX2 L.819, B.8088 An upper part of a krater with two horizontal handles. Stern (1982:99) assumed that this type of krater, with horizontal handles, is a local imitation of East-Greek kraters and should be dated to the sixth–fourth centuries BCE. Similar

103

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

2.1: Plate of Iron Age and Persian Period pottery No.

Type

Basket

Locus

Context

1 2 3 4 5 6 7 8 9 10 11 12

R-BLX 1 R-BLX 2 R-BLX 3 R-BLX 4 R-BLX 5 R-BLX 6 R-BLX 7 R-CUPX 1 R-KRX 1 R-KRX 2 R-SJX 1 R-SJX 2

3048.7 1147 8030.7 3080.11 1273.1 1159 1077 3019 8076.3 8088 1166.1 8077.6

321 158 804 322 174 156 132 306 814 819 161 818

Wall fall in square E54 Mixed (section cleaning) Topsoil Topsoil Natural fill outside the structure south of W12 Topsoil Wall fall in square I48 Topsoil Wall fall in square H48 Wall fall, southeast part of square E49 Fill and occupational debris in square H49 Occupational debris with ash layer

104

Avraham Faust and Adi Erlich

sherds were found at Tel Batash Stratum 1 (Mazar and Panitz-Cohen 2001: pl. 105:5).

follow the same numbering system begun in the Hellenistic typology report.

R-SJX1

Methodology

L.161, B.1166/1

All the pottery was sorted, recorded, and arranged by types defined by ware, function, body and rim forms and eventually, distinctive features. The corpus is divided into general groups according to functional aspects when possible, such as Utility Vessels, Kitchenware, Table and Serving, and Personal Vessels, following Tel Anafa’s system (Berlin 1997a), which was later adopted by BarNathan (2002 and 2006) and Sandhaus (forthcoming). The different vessels were identified according to form and/or rim forms. Each type was designated by the prefix R (for Kh. er-Rasm), with two capital letters defining the vessel’s name, such as SJ (for storage jar), JG (for jug) and so on, and a number, resulting in R-SJ1. Variations in rim form or in capacity are distinguished as subtypes in capital letters (e.g., R-CP7A and R-CP7B).

A rim of a medium-sized bag-shaped jar. There are several subtypes of this jar (for a definition of the type and its subtypes see Mazar and Panitz-Cohen 2001:98–101), which is the most common type of jar in the coastal region during the Late Iron Age, continuing to the Persian period. The jar has been found at many sites including Tell elHesi Stratum V (Bennett and Blakely 1989: fig. 137:3–4), Gezer (Gitin 1990: pl. 28B:11) and Tel Batash (Mazar and Panitz-Cohen 2001: pl. 97:9). The harbinger of this type can be seen earlier, as at Tel es-Safi/Gath Stratum A3 (Shai 2006:115–117).

R-SJX2 L.818, B.8077/6

Once the typology was defined, a second sorting of all the rims was done (Appendix 2.6; at the back of the volume). Closing the Pottery Analysis is a concluding summary.

A fragment of a basket handle of a jar, with a short neck, everted rim and ridge along the neck. According to SingerAvitz (1989b:137) this type was more common at the beginning of the Persian period, and was found at sites such as Tel Michal (Singer-Avitz 1989b: fig. 9.15:5).

The Catalogue includes general information about all the types included under the same name (e.g., storage jars) followed by a fixed structure:

Summary

Type R-SJX

The limited number of sherds and types lacking any stratigraphic context prevents any in-depth study of the status of the site during the Iron Age and the Persian period. However, the fact that a large number of sherds are dated to these periods indicates that there was human activity in Kh. er-Rasm during these eras. The nature of this activity will be discussed in Chapter 4.1.

Description: Type definition, form and ware description. Type date range: Defined by the appearance of the vessels in well-defined chronological sites, stratigraphical data from Kh. er-Rasm and typological analysis. The dates given here represent the writer’s conclusions regarding the dates of this particular type (in contrast to the dates quoted in “Occurrence” below).

2.2. The Pottery from Khirbet er-Rasm: The Typology of the Pottery from the Hellenistic, Roman and Later Periods

Provenance: The stratigraphical context where the vessels (complete and fragments) were discovered. Only complete vessels are mentioned here by loci numbers; for further information see Appendix 2.6 (at the back of the volume).

Débora Sandhaus Introduction

The stratigraphical context of the finds is given here using slightly different terminology from other parts of the report, with the major difference being that what is usually called wall fall is here called collapse. For full stratigraphical analysis the reader is referred to Part 1 of this report; for the purpose of analyzing the pottery, Part 1 should be used together with Appendix 2.6.

This report discusses the typology of pottery from the Hellenistic period onward, recovered from Kh. er-Rasm. The pottery assemblage of Kh. er-Rasm includes a large quantity of Hellenistic pottery, Hasmonean and Herodian sherds, a number of earlier sherds (Iron Age and Persian period, see Chapter 2.1), and some later (Byzantine and Islamic) pieces.

Occurrence: Includes sites with similar vessels, according to a fixed formula: site, stratigraphical context, date (reference). Note that the dates cited here for the various forms usually follow the original publication, and do not necessarily coincide with the conclusions of the present study.

The provenance of the vessels is mostly from the accumulation levels above the floors, and from the occupational levels. The report begins by presenting first the Hellenistic typology, and afterwards the rest of the material, including the Roman, Byzantine and Islamic periods, which

105

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

Discussion: This rubric appears when necessary, and at times the explanations for the type data range (above) are given here.

Pottery from the third century BCE in the region is still an unknown subject since few stratigraphical remains were assigned to that period. The publications of Tel Dor and Tel Gezer present a number of assemblages dated to the Early Hellenistic period. During the last five years, the Modi‛in– Shoham area was intensively excavated, uncovering sites including Early and Late Hellenistic stages. Most of the excavations are in the process of publication and Mr. Yehuda Rapuano, who is in charge of publishing the pottery, kindly showed me pottery from several sites dated to the third century BCE.

The part dealing with the late material follows the same system and numbering as used in the Hellenistic corpus but instead of describing the vessels by function, as in the Hellenistic corpus, a traditional system was adopted, ranging from closed large vessels to open small ones. This decision was made due to the irrelevancy of a system based on function, since the nature of the material is unknown.

Selected Sites for Comparison

2.2.1 The Hellenistic Pottery The Hellenistic corpus is characterized by its local nature. Imports are absent from the assemblage and fine ware is represented by local vessels painted or covered in red, brownish-red, mottled red and black. The group includes utility vessels, kitchenware, table and serving utensils and personal vessels.

Sites were selected based on geographical proximity, welldated Hellenistic contexts and recently published reports. Maresha (Levine 2003:73–136) was the primary source since the site is located at about a distance of only 12 km from Kh. er-Rasm. Although the pottery from Kh. er-Rasm includes most of the local types from Maresha some differences were observed. Certain vessels, such as the everted bowls common in Maresha. are absent from Kh. er-Rasm. Cooking pots share the same form but were made from a different clay, and there are a few types that appeared in the Kh. er-Rasm repertoire but are absent in Maresha.

Utility vessels include storage jars and jugs used for storing liquids, and large coarse bowls which seem to have been used as the earlier mortaria, for grinding cereal or for preparing meals. Kitchenware, including cooking pots, one casserole, and one casserole-lid are well represented in Kh. er-Rasm’s repertoire. The largest group, table and serving ware, includes bowls, plates, jugs, decanters and kraters. Moreover, there is a considerable number of personal containers, mostly used as perfume or perfume oil containers, including unguentaria and juglets. The group also includes flasks, which serve for personal use.

Aderet, a farm dug by Ora Yogev, is situated some five km from Kh. er-Rasm. The structure (yet unpublished; Jon Seligman is preparing the material for final publication) was built during the Iron Age and was in use during the Seleucid period (third century BCE).2 Surprisingly, the repertoire was strikingly different from that of Kh. erRasm. A large number of types are exclusive to Aderet’s repertoire and vice versa. The reason for the differences is probably because Aderet’s repertoire is typical of the third century while the Kh. er-Rasm corpus corresponds to the Late Hellenistic period. We cannot, however, dismiss the option that the difference is due to the ethnic population that lived on the farm.

In order to understand the nature of the corpus, statistical analysis based on complete3 vessels was made (not including lamps), resulting in a distribution where table and serving utensils comprise the bulk of the pottery (21, 41%), followed by kitchenware (14, 27%). Interesting is the small quantity of utility vessels (4, 8%), particularly storage jars (3), which we would expected to find in large quantities. Worthy of note is the appearance of eight complete unguentaria (8, 16%). For the implications of the pottery analysis in relation to the nature of the architectural complex, see Chapter 4.2.

Tel Gezer, located at a distance of about 19 km from Kh. er-Rasm, provides well-dated assemblages from the third to the first century CE (Gitin 1990).

The Hellenistic pottery includes pottery dated mainly to the second century BCE, more precisely to its second half. The presence of earlier types dated to the fourth–third centuries BCE suggest earlier occupation covering this time span.

The assemblages from Tel Dor (Guz-Zilberstein 1995), Tel Anafa (Berlin 1997a) and Tel Iztaba (Sandhaus forthcoming) were selected even though they are far from the site, for one main reason: they are the largest corpuses of Hellenistic pottery recently published. The sites of Jerusalem (Geva 2003), Jericho (Bar-Nathan 2002) and Masada (Bar-Nathan 2006) were also used in spite of their Jewish character, because they are the most updated pottery reports dealing with the period from the second century BCE to the second century CE. Most of the sites published to date are included in the aforementioned publications, and therefore we will not name them. 2

Utility Vessels˝(Storage Jars, Storage Jugs and Mortaria) (Figs. 2.2, 2.3) Storage Jars (Fig. 2.2:1–8) Complete storage jars occur in Kh. er-Rasm assemblage By “complete” vessels we refer to vessels of which at least the entire profile or most of the pot was unearthed. Since no bulk was removed, and no room was excavated in its entirety, we cannot expect to find all the sherds, and hence the above is sufficient to infer that the vessel existed when the site was destroyed (the logic behind an examination of this sort will be further discussed in section 4, Chapter 4.2 below). 3

We thank Jon Seligman for showing us the pottery material.

106

Avraham Faust and Adi Erlich

in scant quantities (three almost-complete vessels). Sherds are widespread all over the site, and include almost all the Hellenistic forms. R-SJ1, the bag-shaped storage jar is the most common type in the site, mainly the R-SJ1C variant. The chronological span for the group is the third–second centuries BCE, but type R-SJ1C is the later variant, found in late second century BCE contexts. The appearance of a complete storage jar, type R-SJ2 in a sealed locus (L.824), provides a good anchor for the occurrence of this type during the late second century BCE. The fact that storage jar R-SJ3, with collared rim, typical of Maresha, appears at Kh. er-Rasm in contexts dated to the late second century BCE and its resemblance to storage jars typical of the first century BCE, points to a late second century BCE date for its first appearance. The class includes a few sherds corresponding to type R-SJ4, dated to the end of the second–beginning of the first centuries BCE.

Occurrence: Jerusalem, Jewish Quarter, main component in Stratum 4, area W and Stratum 6-5 in area X2, both dated to the second half of the second century–early first century BCE (Geva 2003:124, fig. 5.1 SJ 3a, fig. 5.2:25). Tel Dor, appears in all the Hellenistic phases (Guz-Zilberstein 1995:311, fig. 6.36, No. 12, Type SJ1b). Tel Iztaba, 107 BCE (Sandhaus forthcoming).

R-SJ1 (Figs. 2.2:1–5)

R-SJ1D (Fig. 2.2:4)

The type includes vessels with elongated bodies, round shoulders, convex base, and a pair of handles attached to the upper part of the body. A delicate ridging on the body is observable. The neck, which usually flares outward, varies in height and profile, and includes rounded, triangular, modeled and square profiles. They were made of reddish-yellow to pinkish buff clay with white grits. The vast number of variations in rim form is the basis for the subdivision into subtypes, even though this subdivision does not necessarily reflect chronological differences.

Description: Out-folded dropped-down rim.

R-SJ1C (Fig. 2.2:3) Description: Short neck and out-folded rim, squared in profile. Provenance: 72 sherds collected from collapse and occupational levels. Occurrence: Tel Iztaba, 107 BCE (Sandhaus forthcoming).

Provenance: 22 sherds found in collapse and occupational levels. Occurrence: Tel Dor, appears in all the Hellenistic phases (Guz-Zilberstein 1995:311, fig. 6.36, Nos. 7–9, type SJ1b).

R-SJ1E (Fig. 2.2:5) Description: Out-folded rounded rim.

Type date range: Third–second centuries BCE.

Provenance: 21 sherds found in unsealed and disturbed collapse loci and in clean collapse and occupational levels.

R-SJ1A (Fig. 2.2:1)

Occurrence: Tel Dor, the type was sorted in all phases of the Hellenistic period (Guz-Zilberstein 1995:311, fig. 6.35, type JR1a).

Description: Relatively tall neck and thickened modeled rim, triangular in profile. Provenance: One complete storage jar was found in the collapse level. Unfortunately this is a disturbed locus (L.149). Seven sherds were found all over in the surface, collapse and occupational levels.

Discussion: The bag-shaped jars are one of the most popular types during the Hellenistic period. They belong to a long tradition of local pottery extending back to the Persian period (fifth–fourth centuries BCE). Most scholars tend to see these storage jars as a coastal type (Guz-Zilberstein 1995:311).

Occurrence: Tel Gezer, type 160D, Stratum IIB, Late Hellenistic, mid–second century BCE (Gitin 1990: pl. 36:5–6) and late second century BCE (ibid. pl. 39:8–9). Samaria, third century BCE contexts (Zayadine 1966: pls. XXVII, XXVIII:20, 22). Tirat Yehuda, first half of the second century BCE contexts (Yeivin and Edelstein 1970:58; fig. 6:5–6). Tel Dor, the type was sorted in all phases of the Hellenistic period (Guz-Zilberstein 1995:311; fig. 6.36; type SJ1b). Tel Iztaba, 107 BCE destruction level (Sandhaus forthcoming).

The neck, which usually flares outwards, varies in height. A feature which tends to change during the second half of the second century BCE is the form of the rim; storage jars bearing a folded round rim are more typical of the Early Hellenistic period, turning into the squared, triangular profiled rims toward the second half of the second century. This evolution was indicated in the report of Tel Dor (Guz-Zilberstein 1995:311) and Beth-Zur (Lapp and Lapp 1968:71–72).

R-SJ1B (Fig. 2.2:2)

R-SJ2 (Fig. 2.2:6)

Description: Short neck and modeled out-folded thickened rim.

Description: Storage jar with pear bag-shaped body, rounded shoulders and two ridged handles attached

Provenance: 17 sherds discovered from the surface, collapse and occupational levels. 107

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

2.2: Plate of Utility Vessels – Storage Jars No.

Type

Basket

Locus

Context

Ware

1

R-SJ1A

1149.1

149

Fill and occupational debris

2

R-SJ1B

1085.1

130

Wall fall south of wall W19

3

R-SJ1C

1244.2

180

Topsoil

4

R-SJ1D

1214.8

170

Topsoil mixed with wall fall

5 6 7 8

R-SJ1E R-SJ2 R-SJ3 R-SJ4

1061.1 8100.2 1064.1 1075.1

126 824 129 132

Wall fall south of W16 Ash layer Wall fall in corridor Wall fall in room

Pink (7.5YR7/4) Light red (2.5YR6/6); exterior: Pink (5YR8/3) Very pale brown (10YR8/3) Light red (2.5YR6/6), exterior: very pale brown (10R8/3) Pink (5YR7/4) Light red (2.5YR6/6) Pink (7.5YR8/4) Reddish-yellow (5YR6/6), white grits

108

Avraham Faust and Adi Erlich

from shoulder downwards. Its neck is straight or slightly narrowing toward the top and out-folded rim with rounded profile. The usual ware is red or reddish-yellow but it appears also in buff color with many white inclusions.

R-SJ4 (Fig. 2.2:8)

Type date range: Third–second centuries BCE.

Type date range: Late second–early first centuries BCE.

Provenance: One complete vessel was unearthed on the floor, within an ash layer (L.824), which is a sealed, diagnostic locus. 20 sherds were found in collapse and occupational levels.

Provenance: Five sherds were discovered in surface and collapse layers, one sherd was found in the occupational level in the northeastern part (L.816) and one was found in the ash layer covering the occupational level in room 14 (L.823, see also Chapter 2.4, below).

Description: Out-folded thickened modeled rim profile, tall conical neck; made in reddish-yellow ware with white grits.

Occurrence: Maresha, no similar vessel has been published, but the author evaluated the unpublished material and it represents one of the most common storage jars in domestic assemblages in the site.4 At Tel Dor it is included in Type SJ1b (see aforementioned) which appears during the entire Hellenistic period (Guz-Zilberstein 1995:311, fig. 6.36).

Occurrence: Tel Dor, phase 3 dated between 125–63 (?) BCE (Guz-Zilberstein 1995:311, fig. 6.37:4, JR1c).

Utility Jugs (Fig. 2.3:1–3)

Discussion: It is worth noting the presence of one almostcomplete storage jar, type R-SJ2, in a sealed locus of the ash layer, dated to the second half of the second century BCE (see Chapter 2.4).

Two jugs belong to the utility vessels repertoire. Due to the coarse nature of the ware and the dimensions of the vessels, it appears that they served as containers for the storage of liquids rather than for table serving. They include R-JG1 and R-JG2.

R-SJ3 (Fig. 2.2:7)

R-JG1 (Fig. 2.3:1)

Description: Storage jar with ovoid body narrowing toward the bottom ending in a convex base, round shoulders, two ridged handles attached from shoulders, out-folded ridged and flattened rim covering the entire neck surface (based on complete vessels found in other sites, since just one fragment including the rim, neck and shoulders was identified at Kh. er-Rasm).

Description: Large jug with everted neck ending in an out-flaring simple rim. A ridged handle attached from rim to shoulder. Made in reddish-yellow (5YR 7/6) ware with many large white inclusions. Type date range: Third–second centuries BCE. Provenance: One almost-complete jug was found in an unsealed collapse locus (L.170), and seven additional sherds were discovered in unsealed collapse levels.

Type date range: Third–second centuries BCE according to Maresha; typologically it seems to be a later type, probably from the second and even first centuries BCE.

Occurrence: Tel Dor, similar but red glazed, third–second centuries BCE (Guz-Zilberstein 1995: fig. 6.29:14).

Provenance: The upper half of one storage jar was found in the collapse level (L.129) and 14 sherds were unearthed in collapse levels; a few originated in unsealed loci.

R-JG2 (Fig. 2.3:2) Description: Large jug with cylindrical out-flaring neck ending in an out-folded modeled rim. Sometimes an inner groove is observed, probably to fit a lid. A handle is attached from rim to shoulder. The vessel shares the same rim as R-SJ7, and thus, when only rims were unearthed, it was very difficult to distinguish between the two. The ware varies from pinkish-white (7.5YR 8/2) to light reddishbrown (2.5YR 6/4). Sometimes a light reddish-brown (5YR 6/4) or gray core (2.5YR N5) is visible. Many white inclusions and the negative of straw are distinguished.

Occurrence: Maresha, third–second century BCE (Levine 2003:97–98; fig. 6.8:83) and first century BCE (Kloner and Hess 1985:130, fig. 4:9). Discussion: The vessel seems to be a Maresha type since no identical jars were found elsewhere. The type was dated to the third–second centuries BCE based on its appearance in the subterranean complexes, and not on stratigraphical occupational levels. Storage jars with collared rim appear in contexts of the first century BCE, such as Jericho (Bar-Nathan 2002:29–30) and they seem to be characteristic of the period. If so, we should see this type as a second–first centuries BCE type rather than a third–second centuries BCE type.

Type date range: Third(?)–second centuries BCE. It might have continued into the first century BCE. Provenance: Ten sherds occur in collapse and occupational levels. Occurrence: Maresha, third–second centuries BCE (Levine

We would like to thank Prof. Amos Kloner for showing us the unpublished material. 4

109

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

2.3: Plate of Utility Vessels – Storage Jugs and Mortaria No.

Type

Basket

Locus

Context

Ware

1

R-JG1

1268.8

186

Wall fall and occupational debris

2

R-JG2

1264.8

186

Wall fall and occupational debris

3

R-JG4

1214.1

170

Topsoil mixed with wall fall

4

R-MT1

3021.1

309

Wall fall? Equals L305, L307

5

R-MT2

1243.1

176

Wall fall and occupational debris

6

R-MT2

8095.1

816

Wall fall at southwest part

110

Light reddish brown (2.5YR6/4); core: gray (2.5YRN5). Pinkish-white (7.5YR8/2); core: light reddish brown (5YR6/4) Reddish-yellow (5YR7/6); exterior: reddish-gray slip (5YR5/2) White (2.5YR8/2) Pink (5YR8/3); exterior: light red (2.5YR6/6) Light red (10R6/6) with many white grits

Avraham Faust and Adi Erlich

2003:104, fig. 6.11:106–110). Tel Gezer, Stratum IIA, midfirst century BCE (Gitin 1990: fig. 43A:10). Tel Dor, where it appears in earlier loci but it is most common in second century BCE assemblages (Guz-Zilberstein 1995:309, fig. 6.31, type JG4).

Occurrence: Tel Dor, loci of the Early Hellenistic period (Guz-Zilberstein 1995:295, fig. 6.9:1–2, Type BL14b). Tel Anafa, Persian (Berlin 1997a:124–125, pl. 38:341). Although 33 sherds were found, most of them in Hellenistic phases (11 in Hellenistic 1, nine in Hellenistic 2a and nine in Hellenistic 2b). Berlin, however, dates them to the Persian period, calling them surviving Persian period material. Tel Yoqne‘am with heavy rounded rim, Hellenistic, from third– second centuries BCE (Avissar 1996a:51, fig. X.2:4–5).

Discussion: R-JG2, is the most common jug of the second century BCE contexts throughout the country. It develops into the triangular rim jug, typical of the first century BCE (Jericho, type J-JG1, Bar-Nathan 2002:39–40).

For Persian parallels: Lachish (Tufnell 1953: pls. 80.68, 98.568); Tel Gezer (Gitin 1990: pls. 25:20, 30:7–11, 13–16).

R-JG4 (Fig. 2.3:3)5 Description: Large jug with rounded body, wide vertical neck and out-folded modeled rim. A ridged handle is attached from rim to shoulder. The jug is made in a range of colors between reddish-yellow (5YR7/6) and light brown (7.5YR 6/4); sometimes it is covered with red slip.

Discussion: This type is the traditional mortarium of the Persian and Early Hellenistic periods. Bowls of the Hellenistic period do not differ from those of the fifth and fourth centuries BCE (see Stern 1995: fig. 2.2). It seems that the prototype is the flat-bottomed large basins first attested towards the end of the eight century BCE; the ringfooted form appears in the late sixth century BCE (Salles 1985:200–201). Until recently, the consensus was that the type went out of use some time in the fourth century BCE, but nowadays we are familiar with them in contexts dated to the third century BCE (Berlin 1997a:125). Kh. er-Rasm’s specimens were found on the surface and are probably residual material from earlier layers (see section 4).

Type date range: Second–early first centuries BCE. Provenance: One sherd in collapse level and five in upper debris mixed with collapse level. Occurrence: Tel Gezer, mid-second century BCE, common at the end of the century (Gitin 1990: pls. 34B:25–26; 37:3) and a few exemplars in early first century BCE contexts (Gitin 1990: pls. 41:22–23; 44:1). Jerusalem, Jewish Quarter, second century BCE contexts (Geva 2003:127, pl. 5.1, JG1). Beth-Zur, characteristic of the Hellenistic period (Sellers 1933: pl. XI:1; Lapp and Lapp1968: fig. 25). Samaria, second century BCE (Crowfoot, Crowfoot and Kenyon 1957: figs. 42:5–6; 43:10). Yoqne‛am, second half of second century BCE (Avissar 1996a:57, fig. X.7:4).

R-MT2 (Fig. 2.3:5–6) Description: Deep, large coarse bowl with ring rim. The inner surface of the bowl is pink (5YR7/4) due to firing conditions, while the outer surface is in a light red tone (10R6/6). Many big white grits are visible together with straw negatives.

Discussion: Some of the rims of this jug type are very similar to rims of storage jars (SJ7), and it is sometimes difficult to distinguish between the two. The red covering, typical of the late second and beginning of first century BCE, points to a serving rather than a utility function.

Type date range: Third–second centuries BCE. Provenance: 22 sherds appear at Kh. er-Rasm in collapse and occupational levels.

Large Coarse Bowls (Mortaria) (Figs. 2.3:4–6)

Occurrence: Maresha, third–second centuries BCE (Levine 2003:81, 83, fig. 6.2:27). Levine suggested that these plain bowls served as local mortaria (2003:82). Tel Gezer, Stratum IIC, early mid-second century BCE (Gitin 1990: pl. 34:12). Tel Dor, BL14a, mortaria, Hellenistic (GuzZilberstein 1995: fig. 6.9:12).

Two mortaria occur in the repertoire of Kh. er-Rasm. R-MT1 seems to be a residual Persian or Early Hellenistic type. R-MT2 is the mortarium that was in use at the site during the last occupational level.

Discussion: This mortarium seems to be in relatively common use at Kh. er-Rasm. Its appearance in occupational and collapse sealed levels allows us to suggest that the type was still in use during the second half of the second century BCE.

R-MT1 (Fig. 2.3:4) Description: Large and heavy-walled bowls. The bowls are undecorated and usually without handles. They are made of coarse ware and have relatively thick walls, thickened or extended rims.

Kitchenware (Fig. 2.4:1–14)

Type date range: Fifth–third centuries BCE.

Kitchenware comprises an important part of the Hellenistic repertoire from Kh. er-Rasm (22% of the complete vessels), including ten closed cooking pots, one casserole, and many sherds of closed cooking pots and casserole lids.

Provenance: Three sherds were unearthed in topsoil. 5

For R-JG3, see below in the section on pouring jugs.

111

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

The kitchenware corpus sheds light on the chronological determination of the entire assemblage, since it developed through the centuries, and well defined types are characteristic of limited time periods. Cooking pots from the Early Hellenistic period are very different from their counterparts from the late second and early first centuries BCE, and differ from those typical of the first century BCE, and the first and second centuries CE. A few types appear throughout a long time span, from the fourth to the second centuries BCE (R-CP2 and R-CP3). Others appear during the entire second century BCE (R-CP5, R-CP6), while most of the types occurring within the wall falls and occupational debris of the buildings are diagnostic of the late second and early first centuries BCE (R-CP4, R-CP7A, R-CP7B, R-CP7C and R-CP8). There is one residual sherd dated to the transitional phase between the Persian and Hellenistic periods (R-CP1).

thick rim and short neck. The profile of the rim varies from rounded to square. Ware: reddish-brown clay (2.5YR5/4) with many white inclusions and dark red core (2.5YRN4). Type date range: Fourth–second centuries BCE. Provenance: A complete vessel was found within the collapses in room 16 (L.815); 17 sherds appeared in collapse and occupational levels. Occurrence: Jerusalem, Jewish Quarter, common in Stratum five in area W and Stratum seven in area X-2; dated to the Persian–Early Hellenistic period (Geva 2003: figs. 5.2:CP1, 5.4:9). Tel Dor, variation of type CP1, dated from the fourth to the first centuries BCE (Guz-Zilberstein 1995: fig. 6.17:5). Discussion: Geva (2003:133) states that the type is typical of the Early Hellenistic period, although some specimens appeared already during the Persian period. Some specimens still appear sporadically in the second half of the second century BCE. In spite of the aforementioned, a complete example was found in the collapse (L.815). Its occurrence in a late second century BCE context (see Chapter 2.4) is in line with the finds at Tel Dor and reinforces the conclusion that it continues to appear, and not only sporadically, during the late second century BCE.

The Kh. er-Rasm repertoire of cooking pots consists of general ware, slightly smooth, varying from light red to red and even reddish-yellow color with many white grits. Few cooking pots demonstrate the gritty character that is usual in Maresha’s cooking vessels and in Roman and later cooking pots. In general, cooking pots of the Hellenistic period, among them the Kh. er-Rasm specimens, were made in a finer ware than their Persian and their Roman counterparts, although Hellenistic vessels have wider walls than do the thin-walled Roman cooking pots. Still, few types were made in gritty ware such as R-CP4, R-CP5 and some variants of R-CP7A and R-CP7C.

R-CP3 (Fig. 2.4:4) Description: Small globular unribbed cooking pot with convex base, everted neck and simple rim. Two ridged handles are attached from rim to shoulder. Ware: from dark gray (2.5YR 4\1) to reddish-gray (10R5/1).

Closed Cooking Pots (Fig. 2.4:1–12) R-CP1 (Fig. 2.4:1)

Type date range: Second half of fourth–first centuries BCE; specifically, the end of the second century–first century BCE.

Description: Cooking pot with tall wide neck ending in a bolded simple rim with an internal groove on top. Remains of burn marks do not allow us to define the color of the ware; many big white inclusions.

Provenance: One complete cooking pot in the collapse (L.815); seven sherds, mostly in collapse level loci.

Type date range: Transitional phase between the Persian and Hellenistic periods and Early Hellenistic period.

Occurrence: Beth-Zur, Stratum II (Lapp and Lapp 1968: fig. 24:2) and Stratum I (with a squat body; Lapp and Lapp 1968: fig. 27:1–7). Samaria, similar but bigger, first half of the second century BCE (Crowfoot, Crowfoot and Kenyon 1957: fig. 41:1–2). Jericho, standard vessel in Hasmonean 2 contexts, and it still was in use during the Herod 1 period (31–15 BCE) (Bar-Nathan 2002:71–72, pl. 12:145, type J-CP2A). Tel Dor, from the second half of the fourth century BCE to the second century BCE (Guz-Zilberstein 1995: fig. 6.17:2–4, CP1). Tel Anafa, similar but with taller neck, pointed rim, HELL 1 and HELL 2 contexts dated from 250 to 110 BCE (Berlin 1997a: pl. 21:187–190).

Provenance: Five sherds in collapse level loci, unsealed. Occurrence: Tel Gezer, Late Hellenistic Stratum IIB, only in mid-second century BCE (Gitin 1990:258, pl. 37:19). Jerusalem, Jewish Quarter, area W Stratum V and Stratum IV dated to the Persian–Early Hellenistic and Late Hellenistic periods respectively; area X-2, Stratum 7, dated to the Persian–Early Hellenistic period (Geva 2003: fig. 5.2:CP3; pls. 5.1:25; 5.3:1; 5.6:33). Discussion: A. Berlin suggested that it could be an earlier type (personal communication); but according to the parallels it appears also in second century BCE contexts.

Discussion: This cooking pot is part of a large local family developed from the Early Hellenistic to the Roman periods. Different variations in form developed over time. The cooking pots from Early Hellenistic times have thick walls and neck, the rim profile evolved from round to triangular

R-CP2 (Fig. 2.4:2–3) Description: Rim sherd of a cooking pot with everted

112

Avraham Faust and Adi Erlich

2.4: Plate of Kitchen Ware No.

Type

Basket

Locus

Context

Ware

1 2 3 4

R-CP1 R-CP2 R-CP2 R-CP3

8095.2 1159.2 8052.4 1044

816 161 807 115

Wall fall at southwest part Fill and occupational debris Wall fall north of cell Wall fall inside installation

5

R-CP4

1036.2

115

Wall fall inside installation

6

R-CP5

1121.1

147

7 8 9 10

R-CP6 R-CP7A R-CP7A R-CP7B

8080.3 8077.3 8085.12 8077.2

815 818 818 818

Lower occupational debris and kirton wall fall north of W22 Wall fall at northern part Occupational debris Occupational debris Occupational debris

11

R-CP7C

1166.1

161

Fill and occupational debris

12 13 14

R-CP8 R-CS2 R-CS1

8085.15 1172.1 8036.1

818 164 810

Occupational debris Fill/wall fall, equals L165 Wall fall

dark gray (5YR4/1) Reddish-brown (2.5YR5/4) Red (10R5/6) Red (10 R4/6); gritty Weak red (10R5/4); gritty, burnt signs all over the body Red (10R4/6); gritty, mica and white grits Light reddish-brown (2.5YR6/2) Light red (2.5YR6/6) Light red (2.5TR 6/6) Pale red (10R6/4; burnt signs Reddish-gray (10R5/1); core: reddish-brown (2.5YR4/3); big white inclusions Light reddish-brown (2.5YR6/3) Reddish-yellow (5YR7/6) Weak red (10R5/4)

113

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

and the neck from flaring to vertical (Bar-Nathan 2002:72). Kh. er-Rasm’s type finds its closest parallel in Jericho, type J-CP2A, which was dated by Bar-Nathan from the end of the second century BCE, or beginning of the first century BCE up to 31–15 BCE, when it was replaced by J-CP2B (Bar-Nathan 2002:72). Based on parallels in other sites, it seems that this specific type appears as early as the second century BCE, probably during its second half.

reddish-brown (2.5YR6/3) turned into reddish-brown (2.5YR5/3) on the outer face due to firing conditions. Type date range: Third(?)–second century BCE to early first century BCE. Provenance: One complete cooking pot buried in the collapse (L.815); one sherd on floor level below the ash layer.

R-CP4 (Fig. 2.4:5)

Occurrence and discussion:6 The vessel is unusual both in form and ware. The pot was not produced of regular cooking pot ware, which tends to be coarse and gritty; instead it resembles the usual ware of the storage jugs found in the site. Despite the aforementioned, the ware of cooking pots characteristic to the Hellenistic period tends to be less coarse and gritty than the later counterparts; during the late second century BCE, cooking pots became thinner and grittier. Therefore, the clay in which the vessel was produced does not necessarily point to a non-cooking function.

Description: Large cooking pot with sack-shaped unribbed body, ribbed on shoulder, convex bottom, out-flaring neck, sharply incurved rim. The vessels were manufactured in a gritty and coarse weak red ware. Type date range: Second half of the second century–early first century BCE. Provenance: One complete cooking pot was found buried by collapse (L.115) and one sherd in surface. Occurrence: Maresha, the type is dated on the basis of parallels to “no earlier than the second half of the second century BCE” (Levine 2003:94, fig. 6.6:70–71). Tel Gezer, Stratum IIB, late second century BCE (Gitin 1990: pl. 40:26, type 239D). Jerusalem, Jewish Quarter, Stratum 4 in area W and Stratum 5-6 in area X2, both dated to the second half of the second–early first centuries BCE (Geva 2003: pl. 5.1:27, pl. 5.10:20).

The complete vessel shows signs of fire, which resulted from the destruction of the site, but there are burn marks on the base, as one could expect from a cooking pot which was meant to be set on the stove. The form of the neck and handles resemble that of table amphorae, but unlike them, the sack shape of the body suggests storage or cooking utility rather than a tableserving one.

Discussion: The occurrence of this type mainly in southern sites and in relatively late contexts (second half of the second century BCE) points to a southern origin. The absence of the type from Tel Dor, Tel Iztaba and Tel Anafa might reinforce this hypothesis.

Since no identical vessel was found, the closest parallels were found in large jugs, which are similar in form, details and ware but differ in having just one handle. Therefore, the parallels mentioned in the following paragraph were cited mostly for chronological reasons. Large jugs with pear-shaped body and ring base were found in Maresha. The neck, rim, handle and ware are identical to Kh. erRasm’s example. Levine argued that based on the fact that no parallels have been found except in Maresha, and the resemblance of the jugs to Persian prototypes, the type should be dated to the Early Hellenistic period (Levine 2003:100–102, fig. 6.10:92–93). Indeed, the type seems to have its prototype in the Persian period (Stern 1982:117, Type H), but in my opinion, the jugs from Maresha could not be earlier than the middle of the second century BCE, for the following reasons:

R-CP5 (Fig. 2.4:6) Description: Large cooking pot with a globular body, short vertical neck, simple short ledge rim. Two handles are attached from rim to shoulder. Gritty red (10R 4/6) ware including mica and white grits. Type date range: Late second century BCE. Provenance: Five sherds in collapse and unclear loci. Occurrence: Samaria, post-Hellenistic fort wall, Late Hellenistic, wrongly identified as a jar (Crowfoot, Crowfoot and Kenyon 1957: fig. 43:10). Tel Anafa, necked ledge rim cook pot, began in HELL 2A, 125 BCE (Berlin 1997a:90, pl. 23:205).

First, the high base ring which is characteristic of the Late Hellenistic period; second, similar jugs appear in assemblages such as Tel Dor and Judean sites, where local variations have developed and all are dated from the Early Hellenistic (and mostly) to the Late Hellenistic period. The type was named at Tel Dor “globular jugs” and is found both in Early and Late Hellenistic phases (Guz-Zilberstein 1995:308–309, type 11, fig. 6.30:1–6). They all have a

R-CP6 (Fig. 2.4:7) Description: Large cooking pot with pear-shaped unribbed body, vertical neck, and out-flaring, beveled rim, with a smooth inner groove, maybe to fit a lid. Two ridged strap handles are attached from rim to shoulder. Ware: light

Since no identical parallel to Kh. er-Rasm’s type was found, and due to the unusual characteristics of this vessel, the Occurrence and the Discussion will be combined. 6

114

Avraham Faust and Adi Erlich

cylindrical wide neck, everted rims, balloon-like body widening towards a convex base, and they also have a loop handle.

is particularly characteristic of the early first century BCE (Geva 2003:135, pls. 5.1:4, 29; 5.3:11–12; 5.4:10,31–32, type CP8a). Tel Iztaba, 107 BCE (Sandhaus forthcoming). Tel Anafa, HELL 2C, 98–75 BCE (Berlin 1997a:90–91, pl. 24:207–209).

In Judean sites of the second century BCE a similar type of jug was found, but its body has a piriform shape and its base is concave. According to Lapp’s typology, their chronological range in the Judean sites is between 175–50 BCE (Lapp 1961: Type 21.1:F, G, H; Lapp and Lapp 1968:76, fig. 25:5–6).

R-CP7B (Fig. 2.4:10) Description: Small cooking pot with squat unribbed body, vertical, high, slightly concave neck, simple rim with triangular profile, and convex base. Two ridged handles attached from rim to shoulder. The ware is pale red (10R6/4) with an unusual non-gritty ware. It is similar to the regular ware used for jugs, kraters and other serving wares.

On the basis of the finding of similar jugs in cistern I at Beth-Zur (Lapp and Lapp 1968:75, figs. 25:2, 4–6) and the abovementioned characteristics, we can deduce that these jugs were used for drawing water from cisterns or even for storage, but not for serving. However, we cannot conclude that Kh. er-Rasm’s specimen corresponds to a jug or another pot, with cooking or food-storage function.

Type date range: Mid-second to mid-first centuries BCE. Provenance: One complete vessel found in the collapse level lying on top of the occupational debris (L.818) and one sherd.

R-CP7 (Fig. 2.4:8–11)

Occurrence: Maresha, second century BCE (Levine 2003: fig. 6.6:72). Tel Gezer, Stratum IIB, mid-second century BCE (Gitin 1990: pl. 37:13–14). Samaria, a similar pot, Late Hellenistic period (Crowfoot, Crowfoot and Kenyon 1957: fig. 41:5). Tel Anafa, HELL 2C, 98–75 BCE (Berlin 1997a:90–91, pl. 24:210).

Description: Cooking pots characterized by a globular or squat body, a neck varying in height and a triangular rim, more or less pronounced. At Kh. er-Rasm, three different variants were distinguished, all related to the Late Hellenistic type. R-CP7A, is characterized by a simple or triangular rim and tall everted neck, CP7B, of smaller size, has a marked triangular profile and a tall concave neck, and CP7C has a shorter concave neck, still tall, and a triangular rim.

R-CP7C (Fig. 2.4:11) Description: Large globular ribbed cooking pot with short concave neck, thickened rim with triangular or rounded profile. Gritty ware ranging from red (2.5YR4/6) to reddishbrown (2.5YR4/4), and reddish-gray (2.5YR5) wares. The clay has a great number of large white inclusions.

R-CP7A (Fig. 2.4:8–9) Description: Medium-sized cooking pot (average high: 20 cm) with bi-conical, un-ribbed body, vertical, tall, slightly everted neck, convex bottom, and a simple triangular or rounded rim. Two ribbed strap handles are attached from rim to shoulder. The pots were manufactured in a relatively smooth, light red (2.5YR6/6) ware with a few white grits. Some variations were made in a different coarser ware typical of cooking vessels, gritty ware ranging from weak red (10R5/4) to light reddish-brown (5YR6/3) color; many white and gray grits.

Type date range: Mid-second to early first centuries BCE, with sporadic appearances in third century BCE contexts. Provenance: Two complete vessels found, one in the collapse debris (L.813) and the other in occupational debris (L.818); 13 sherds found in all Hellenistic levels. Occurrence: Maresha, second century BCE (Levine 2003: similar to fig. 6.6:74). Tel Gezer, late second century BCE (Gitin 1990: pl. 40:26). Tel Dor, found in third, second and first half of first century BCE contexts (Guz-Zilberstein 1995:299, pl. 2.19). Jericho, late second century BCE–15 BCE (Bar-Nathan 2002: pl. 11:124, 128, type J-CP1).

Type date range: Mid-second to mid-first centuries BCE. Provenance: Two complete vessels in occupational debris (L.818 – sealed critical loci) and 14 sherds in collapse and occupational sealed levels.

Discussion: Cooking pots with triangular rim belong to a long tradition of local vessels, whose roots can be traced from Canaanite-Israelite cooking pots and extending throughout the Byzantine period. The group is characterized by a globular or squat body, a neck varying in height and a triangular rim, more or less pronounced.

Occurrence: Maresha, second century BCE (Levine 2003: fig. 6.6:72). Tel Gezer, Stratum III, late third–early second centuries BCE (Gitin 1990: pl. 33:21), Stratum IIB, mid-second century BCE (Gitin 1990: pl. 37:18). Beth-Zur, Stratum I and Stratum II, second–first centuries BCE (Lapp and Lapp 1968: figs. 24:1–2; 26:8; 27:6–7). Tell el-Fûl, Stratum VIa and Stratum VIb, 175–100 BCE (Lapp 1981:104, pl. 79:1). Jerusalem, Jewish Quarter, Stratum 4, area W, late second century BCE; it

The Early Hellenistic pots were made in local coarse ware and had out-turned short necks ending in a simple or triangular profile. Over time the walls became thinner.

115

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

During the Late Hellenistic period (second half of second century–first half of first century BCE) the neck became taller, still with variations from slightly everted rims through vertical and concave ones, and the body presents globular or bi-conical characteristics; ribbing on the outer face starts to appear, but not on a large scale, and two strap handles instead of ribbed ones are common. The pots were made in a variety of wares including cooking, local coarse ware and even fine and painted wares.

base, everted rim, groove on inner side to take a lid, two vertical handles attached from rim to shoulder. Red gritty ware.

During the Hasmonean period, the walls of the vessels developed into thinner ones, the neck is still high, and ribbing became more common.

Occurrence: Maresha, third–second centuries BCE contexts (Levine 2003: fig. 6.6:77). Samaria, Hellenistic Fort Wall, dated up to 150 BCE (Crowfoot 1957a:229–230, fig. 41:11). Jerusalem, Jewish Quarter, areas W and X2, type CA1, third–first centuries BCE (Geva 2003:135–136, fig. 5.2, CA1, pl. 5.8:37). Tel Dor, dominant type in the second century BCE (Guz-Zilberstein 1995: fig. 6.21:9, CP6). Tel Anafa, 125–75 BCE (Berlin 1997a: pl. 29:241).

Type date range: Second century BCE, most probably second half of the century. Provenance: One complete casserole in collapse level in room 15 (L.810).

The transition to the Roman period is characterized by vessels with very thin ribbed walls and shorter necks.

CP8 (Fig. 2.4:12) Description: Very small squat unribbed cooking pot with convex base, everted neck, turned-in rim with a groove on top. One strap handle attached from rim to shoulder. It was made in local light reddish-brown (2.5YR6/3) ware. On the outer surface it has a different tone of reddish-brown (2.5YR5/4) color. It has a few white grits but not the regular gritty texture of the cooking pots.

Discussion: The tendency to flattened rims can also be observed in Athenian casseroles of the mid-second century BCE onwards (Thompson 1934: D72, E141), at Samaria, and at Tel Anafa. At Tel Dor it is infrequent in earlier contexts but it is the most predominant cooking pot during the second century BCE (Guz-Zilberstein 1995:300).

R-CS2 (Fig. 2.4:13)

Type date range: Late second–first centuries BCE.

Description: Saucer/casserole lid with angular walls and angled-cut rim to fit the casserole. Ware: reddish-yellow (5YR7/6).

Provenance: One complete cooking pot was found in occupational debris (L.818). Occurrence and discussion: Small pots are common in Hellenistic sites, though they differ in rim treatment from site to site. It is worth noting that in spite of being catalogued as cooking pots, none of them were produced in cooking ware, but rather in the local juglet wares. Small pots are well known from sites in phases dated to the end of the second century BCE, such as Maresha, in occupational debris of the buildings (dated to the end of the second century BCE, Levine 1999:66, pl. 15:122), Tel Iztaba, where a number of complete vessels were found, catalogued as pots in the destruction level, 107 BCE (Sandhaus forthcoming), and Yoqne‛am, second half of second century BCE (Avissar 1996a: fig. X.3:7). It seems that small pots continued to appear during the first century BCE and at the beginning of the first century CE as indicated by two complete pots found in Jericho (dated to 85/75 BCE–6 CE, Bar-Nathan 2002:72–73, pl. 12:151–152). During the first century CE different variations are discussed in the Masada report (Bar-Nathan 2006:144–145). Bar-Nathan prefers to call them mugs and we tend to accept her nomenclature (Bar-Nathan 2006:144–145). Masada pots differ from the earlier ones in form and handles but they are definitely the evolution of the type common during the first century CE.

Type date range: Second–beginning of first centuries BCE. Provenance: One sherd in collapse debris. Occurrence: Maresha, third–second centuries BCE (Levine 2003: fig. 6.2:29). Tel Gezer, mid-second century BCE (Gitin 1990:94, pl. 38:18). Tel Anafa, Stratum ROM 1A, end of first century BCE–beginning of first century CE (Berlin 1997a:83, pl. 19:PW174). Discussion: Levine proposed a date range for the type from the second century BCE, through the first century BCE and the early first century CE (Levine 2003:83). Although saucer lids were in use during the first century BCE and first century CE, they differ from Kh. er-Rasm’s type morphologically. They became lower, flatter and with a turned-in rim (Bar-Nathan 2002: pl. 16). We propose to date it to the second –early first centuries BCE for the type.

Tableware (Pouring, Serving and Personal Vessels) (Figs. 2.5–2.7) The term “table vessels” refers to all the utensils that were used during a meal. It includes food-serving vessels such as certain types of kraters, pouring vessels and vessels associated with wine, including table amphorae, jugs, decanters, along with vessels associated with eating and drinking, such as plates and cups.

Casseroles and Casseroles Lids (Fig. 2.4:13–14)

R-CS1 (Fig. 2.4:14)

Description: Casserole with squat rounded body, convex

116

Avraham Faust and Adi Erlich

Pouring and Serving Vessels (Table Amphorae, Jugs, Lagynoi, Decanters, Kraters) (Fig. 2.5)

discovered at the site as yet. A few sherds appearing in nonsealed loci represent the type, although they are probably residual sherds.

R-JG3 (Fig. 2.5:3)

Table Amphorae (Fig. 2.5:1–2)

Description: Jug with conical neck narrowing towards the top, an everted thickened rim with a groove on top of it; manufactured in light reddish-brown ware (2.5YR6/4), gray core (2.5YRN5), with white large and small inclusions.

The table amphora is one of the most typical Hellenistic vessels, identified as a wine container, which was a major component during Greek and Hellenistic meals. Table amphorae are a hallmark of these periods, and their occurrence in the site strengthened the hypothesis favoring a “Hellenic” presence rather than a Jewish one. The complete table amphora is practically the only example of fine ware at the site; imitating the Eastern Sigillata fine ware common in richer sites of the second century BCE.

Type date range: Third(?)–second centuries BCE. Provenance: Two sherds in surface, one in upper debris. Occurrence: Maresha, third–second centuries BCE (Levine 1999: pl. 16:135; Levine 2003: fig. 6.11:106–110). Tel Dor, appears in earlier phases but it is most common during the second century BCE (Guz-Zilberstein 1995:309, fig. 6.31:2, JG4).

R-TA1 (Fig. 2.5:1) Description: Sherd of a table amphora with wide neck turned-out thickened rim with triangular profiled section, ridge at the middle of the neck; produced in light red ware (2.5YR 6/8) with white grits.

R-JG5 (Fig. 2.5:4)7 Description: Thin walled jug with vertical short and narrow neck, flattened rim on top; made in white (greenish-white) color (2.5YR8/2).

Type date range: Third–second centuries BCE. Provenance: Two sherds in surface.

Type date range: Indeterminate.

Occurrence: Maresha, third–second centuries BCE (Levine 1999: pl. 16:137). Tel Dor, similar rim, Early Hellenistic to the end of the second century BCE (Guz-Zilberstein 1995:309, fig. 6.32:4, 6).

Provenance: One sherd in occupational level. Occurrence: No parallels found.

R-TA2/Kantharos (Fig. 2.5:2)

Decanters (Fig. 2.5:5–6)

Description: Table amphora with carinated piriform body, sharply carinated in the shoulders and tall wide neck. The everted set rim has an inner groove, probably to take a lid. Two strap handles emerge beneath the rim to the carination point in the shoulders. It stands on a high base ring. Fluting covers the body up to the shoulder line. It was manufactured of apparently local reddish-yellow clay and covered with red slip, imitating Eastern Sigillata ware.

Local decanters seem to take the place of the Hellenistic lagynoi common in fine Hellenistic assemblages. Two types are included in the Kh. er-Rasm assemblage and they are probably part of the same simple but fine ware of the site. The appearance of complete vessels of this kind, unusual in Jewish repertoires, again reinforces the idea of a pagan occupation.

R-DT1 (Fig. 2.5:5)

Type date range: End of second–mid-first centuries BCE.

Description: Decanter with globular body, cylindrical neck and cup-mouth rim. A ridged handle attached from rim to shoulder. Pinkish-gray clay (7.5YR 7/2) with white grits.

Provenance: An almost complete vessel was found in occupational debris (L.821).

Type date range: Late second–first centuries BCE.

Occurrence: Maresha, before the end of the second century BCE (Levine 2003:89–90; fig. 6.4:60). Samaria, third–second centuries BCE (Crowfoot, Crowfoot and Kenyon 1957:239, fig. 45:11). Jerusalem, Jewish Quarter, Eastern Sigillata type, probably dated between 100–50 BCE (Rosenthal-Heginbottom 2003:195, pl. 6.3:38).

Provenance: One almost-complete vessel was found in a sealed locus of the collapse debris (L.160). Occurrence: Jericho (Bar-Nathan 2002:42–43, pl. 9:67, type J-JG4a).

Pouring Jugs (Fig. 2.5:3–4)

Discussion: Only one parallel jug was found, in Jericho, where it is one of the most abundant jugs in the site; it

Jugs used for pouring liquids are common in local table assemblages, though no complete vessels have been

7

117

For R-JG4, see in the section on utility jugs, above.

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

2.5: Plate of Table Ware – Pouring and Serving vessels No.

Type

Basket

Locus

Context

Ware

1

R-TA1

1214.15

170

Topsoil mixed with wall fall

2

R-TA2

8092

821

Occupational debris

3 4 5 6

R-JG3 R-JG5 R-DT1 R-DT2

1268.22 8100.4 1196.1 1176.2

186 824 160 165

Wall fall and occupational debris Ash layer Wall fall Fill/wall fall

7

R-KR1A

8092.5

821

Occupational debris

8

R-KR1B

1114.3

147

Lower occupational debris and kirton wall fall north of W22

9

R-KR2

8075.4

816

Wall fall at southwest part

Light red (2.5YR6/8) Light red (10R6/6); red slip (10R5/6) Light red (2.5YR6/6) White (greenish) (2.5YR8/2) Light red (10R6/6) Pinkish-gray (5YR7/2) Reddish-yellow (5YR7/3); core: pink (5YR7/6); red slip (?); burnt Reddish yellow (7.5YR7/6); core: pink 7.5YR7/2) and gray (5YR7/6) Red (2.5YR5/6); exterior: pink (5YR7/4)

118

Avraham Faust and Adi Erlich

is dated there to the first century BCE and proposed as the prototype for all the narrow-necked, ridged jugs (BarNathan 2002:43). R-DT1 was found almost complete in Kh. er-Rasm in the collapse debris of square I48, together with pottery sherds dated to the second–early first centuries BCE, and is the only complete vessel in the assemblage in room 13. Its appearance in the last occupational level at Kh. er-Rasm, which we believe was destroyed at some point during the end of the second century BCE (see Chapter 2.4), tends to ante-date the first appearance of the type to the late second century BCE.

thickened rim decorated with a pie-crust design. Based on similar vessels, two coil handles are attached from rim to shoulder. They are made in a pink ware (7.5YR7/2) turning to gray (5YR7/6); the outer surface is reddish-yellow (7.5YR7/6) due to firing conditions. They are decorated with various slip and paint variations. The type includes two variations distinguished by different neck heights and different decoration.

R-DT2 (Fig. 2.5:6)

R-KR1B (Fig. 2.5:8): With high, flaring neck. The rim is decorated by a red band and over the body, thin pinkish gray (5YR6/2) strips cover the body.

R-KR1A (Fig. 2.5:7): With short wide vertical neck. Red slip covered the entire vessel.

Description: Decanter characterized by a squat body, narrow everted rim and handle attached from rim to shoulder. Ware: pinkish-gray clay (7.5YR 7/2) with white grits.

Type date range: Third–first centuries BCE (mainly late second–beginning of first century BCE).

Type Date range: Late second–first centuries BCE.

Provenance: Sherds of one vessel found in occupational debris (L.821) and one sherd in sealed locus of occupational debris.

Provenance: One almost-complete vessel was found in collapse (L.165), three sherds found in surface, collapse and occupational debris.

Occurrence: R-KR1A: Although morphologically the parallels cited below correspond to the short neck type, they appeared to be undecorated. Maresha, third–second centuries BCE (Levine 2003:91–92, fig. 6.5:65). Tel Dor, type KR11, third–second centuries BCE (Guz-Zilberstein 1995:297, fig. 6.14:9). Tel Anafa, first century BCE–first century CE (Berlin 1997a:138, PW412). R-KR1B: Tel Anafa, similar vessel but with a tall, vertical neck instead of an everted one, first appears in HELL 2C (98–75 BCE).

Occurrence: No parallels were found. Discussion: Since no parallels were found and the stratigraphical contexts where the type was found are mostly unsealed loci, the date is just a suggestion based in typological criteria. Decanters with a long narrow neck are common during the first century BCE (Bar-Nathan 2002:43). Bar-Nathan believes that they represent the archetype of all the narrow-necked jugs, including those with the ridge typical of the first century CE (Ibid.:43). Although no similar decanter was found at other sites, I suggest a date during the end of the second–beginning of the first centuries BCE, based on the appearance of this vessel at this point of time – the everted rim (still characteristic of the second century BCE) – which evolved during the first century BCE into a more triangular rim and over time into a vertical neck.

Discussion: Both Levine and Regev maintain that this krater is local in nature (Levine 2003:92; Regev 2003:172–173, form 53). Painted decoration should be seen as a second century BCE feature, which continues to develop during the first century BCE and the first century CE. Pie-crust decoration seems to be a hallmark of later assemblages also dated to the late second–first centuries BCE. In summary, this type should be defined as a late second century–early first centuries BCE type.

Kraters (Fig. 2.5:7–9)

R-KR2 (Fig. 2.5:9)

Kraters were traditionally used as wine containers and mixing vessels; they played the most important role in the symposium–the Greek and Hellenistic traditional dining custom. Although “kraters” are a Greek invention, as is the term, many scholars studying Israel’s earlier pottery (Bronze and Iron Ages) borrowed the name to define big, deep bowls, similar in form to the original krater. The function of these is very different from the original one that was associated with mixing wine; some of the local types were used for serving food. The repertoire of Kh. er-Rasm includes two different types, each one related to one of the two different functions we discussed above.

Description: Sherd of the everted ledge rim of a krater, produced in red (2.5YR5/6) clay turned to pink (5YR7/4) on the outer face. Type date range: Second–first centuries BCE. Provenance: One sherd in the upper debris. Occurrence and Discussion: The sherd is too small to enable us to find a parallel vessel; however the everted ledge rim krater is well documented in assemblages dated from the second century BCE, such as Maresha (Levine 2003:90, fig. 6.4:58), Tel Dor (Guz-Zilberstein 1995:296, fig. 6.11:5) and Tel Iztaba (Sandhaus in press); through sites dated to the first century BCE, when it became the

R-KR1 (Fig. 2.5:7–8) Description: Krater with vertical out-flaring wide neck and

119

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

most common type, for example in Hasmonean complexes at Jericho (Bar-Nathan 2002: pl. 13, 170–171). Variants in rim treatment occur simultaneously during the first century BCE in Jericho, as stated by Bar-Nathan, but the ledge type disappeared in the first century CE (Ibid.:78).

(L.136),8 and two in different occupational levels, L.186 and L.821. 91 sherds were found in all levels.

R-BL1A (Fig. 2.6:1–2) Description: Large bowls with high ring base (average rim diameter: 20–30 cm, average height: 10 cm). The ware varies from reddish-yellow (5YR6/6) to pale yellow (2.5YR7/3) with white grits.

Personal Tableware (Bowls and Plates) (Fig. 2.6) Bowls (Fig. 2.6:1)

Occurrence: Maresha, third–second centuries BCE (Levine 2003:81, 84–85, fig. 6.2:34–35). Tel Dor, second half of fourth century BCE–end of second century BCE (GuzZilberstein 1995:290, fig. 6.1:25–26, type BL8b).

Kh. er-Rasm’s corpus of pottery includes 14 complete bowls and 196 sherds, meaning that together with cooking pots, this is the most abundant class of pottery found in the site. The most common type is R-BL2, characterized by angular walls, followed by R-BL1, the incurved rim bowls. Both types are typical bowls from the second half of the fourth century to the second century BCE. Worthy of mention are the sherds belonging to types R-BL3 and R-BL4, since in spite of being scarce in number, most of the finds were buried in rooms 4 and 14, matching the complete vessels distributional pattern. Most of the complete bowls were in use in the southwest part of the building (see Chapter 4.2).

R-BL1B (Fig. 2.6:3–5) Description: Small bowls (average rim diameter 10 cm, average height: 5–6 cm). The ware varies from light red (10R 6/6) to reddish-yellow (5YR 7/6), well levigated clay with few white grits. A wide variety of decoration including half-painted bowls and undecorated ones is common. Occurrence: Maresha, third–second centuries BCE (Levine 2003:81, 84–85, fig. 6.2:36–38). Tel Dor, BL8a, second half of fourth century BCE–end of second century BCE (Guz-Zilberstein 1995:290, fig. 6.1:15–21). Aderet, third century BCE (J. Seligman, personal communication).

Bowls are described in this report as vessels of a personal nature because although usually they served as personal utensils during meals, as dishes for food or cups for drink, sometimes they were employed for cosmetic purposes.

Discussion: The large bowls (BL1A) are less common than the small ones (BL1B) but are still well known in Hellenistic assemblages dated from the fourth to the second centuries BCE. During the first century BCE a similar type appears in Hasmonean assemblages, with a disc base rather than with the usual ring base. Apparently, the hemispherical bowls became deeper during the first century BCE, and with a less in-curved rim.

R-BL1 ‘Incurved Rim Bowls’ (Fig. 2.6:1–5) Description: Hemispherical wheel-made bowl with thin walls, incurved rim and ring base. It seems to have developed from the Attic black glazed bowls, first appearing in the early fifth century BCE. This type was common all over the Hellenistic realm and it includes bowls, manufactured in a great variety of sizes, wares and decoration.

Although Geva suggested that the disc base appears to be a Judean characteristic, since almost all the bowls originating in Hasmonean contexts belong to the disc-base type (Geva 2003:138, type BL1A, pl. 5.3:19), the appearance of a large number of similar bowls at Maresha seems to indicate a southern specification rather than a Judean one. This conclusion seems to be reinforced by the absence of bowls with disc bases in northern sites, such as Beth-Shean and Tel Anafa.

The earlier vessels tend to be closer to the Attic prototype and are black-slipped, while the typical bowl of the second century BCE tends to be covered with red slip or mottled red paint or is even left bare. At Kh. er-Rasm there are both big and small bowls, and this is the basis for the subsequent definition of subtypes. The big bowls were made in a coarser ware than the small ones and are undecorated.

R-BL2 (Fig. 2.6:6–10) Description: Angular, thin walled bowls with everted walls becoming vertical in the upper part of the vessel, ending in an upturned simple rim. The bowls stand on a high ring base. They are made in a wide range of wares, ranging from pinkish-white (5YR 8/2), pink (7.5YR 7/3), light reddishbrown (2.5YR 6/4) to reddish-yellow (5YR 7/6). All the vessels were manufactured in well levigated clay, even

Type date range: Second half of fourth–second centuries BCE. Provenance: R-BL1, 245 sherds. R-BL1A, two complete bowls found in the collapse debris (L.805) and in occupational debris (L.176); 25 sherds in all levels.

L.136 is topsoil of the third season of excavation in this square (J53), and it is the lowest locus of topsoil. That explains the finding of a complete bowl in a surface locus. 8

R-BL1B, three complete vessels discovered, one in topsoil

120

Avraham Faust and Adi Erlich

2.6: Plate of Personal Table Ware – Bowls and Plates No.

Type

Basket

Locus

Context

Ware

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19

R-BL1A R-BL1A R-BL1B R-BL1B R-BL1B R-BL2 R-BL2 R-BL2 R-BL2 R-BL2 R-BL3 R-BL4 R-PL1A R-PL1B R-PL1C R-PL2 R-PL2 R-PL2 R-PL3

1246.5 8013.1 1099.1 8092.7 8092.8 8076.2 8101.9 8092.9 1064.2 1064.3 1258.1 1117.1 1030.1 1137.1 1139.1 8076.1 8086.5 1128.1 1052.1

176 805 136 821 821 814 823 821 129 129 186 149 109 154 150 814 822 150 119

Wall fall and occupational debris Wall fall in cell Topsoil, equals L108, L118 Occupational debris Occupational debris Wall fall Ash layer Occupational debris Wall fall in corridor Wall fall in corridor Wall fall and occupational debris Fill and occupational debris Wall fall in room Fill and occupational debris Wall fall Wall fall Occupational debris with ash layer Wall fall Topsoil mixed with wall fall

Reddish-yellow (5YR6/6) Pale yellow (2.5YR7/3) Reddish-yellow (5YR6/6); white grits Reddish-yellow (5YR6/6) Ligth red (2.5YR6/6); black slip Pink (5YR7/4) Light red (2.5YR6/6) Light red (2.5YR6/6) Light reddish-brown (2.5YR6/4) Light reddish-brown (2.5YR6/4) Light red (2.5YR6/6) Very pale brown (10YR7/4) Red (10R 5/8) Light red (10R6/6) Reddish-yellow (5YR6/6) Reddish-yellow (5YR6/6) light reddish-brown (2.5YR6/4) Reddish-yellow (7.5YR7/6) Reddish-yellow (5YR7/8); white grits

121

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

if white grits are occasionally visible. The bowls include large and small bowls, and have various decorative patterns, ranging from undecorated bowls to half-painted bowls and entirely painted bowls. The most common color is red but black decoration occasionally occurs as well.

this type of bowl is most frequent in second century BCE contexts, with two exceptions appearing in earlier strata (Guz-Zilberstein 1995:294). It apparently developed into an undecorated type which appears in Jericho, dated to the first decades of the first century BCE (Bar-Nathan 2002:87, pl. 15:229, type J-BL4).

Type date range: Although at Tel Dor the type is registered as early as the fourth century, it seems to be a typical product of the second and first centuries BCE.

R-BL4 (Fig. 2.6:12) Description: Out-curved carinated small bowl. Ware: surface, reddish-yellow (5YR 6/7); core, very pale brown (10 YR 7/4) with small white grits.

Provenance: Nine complete bowls were discovered: two in the collapse of L.129, two in the occupational debris of L.147, two in the occupational debris of L.186, one in the collapse mixed with occupational debris of L.324, one in the collapse of L.814 and one more underneath the latter, in ash layer and occupational debris L.823. 65 sherds were found in all levels.

Type date range: Third–second centuries BCE. Provenance: No complete bowls were discovered; seven sherds, most of them in collapses. Occurrence: Tel Anafa, context dated to HELL A and HELL 1B third–second centuries BCE (Berlin 1997a: pl. 16:PW141, everted rim bowl). Tel Dor, appears in contexts of the third century BCE (Guz-Zilberstein 1995: fig. 6.2:11–12).

Occurrence: Maresha, third–second centuries BCE (Levine 2003:81, 85–86, fig. 6.2:39–44). It seems to be a local imitation of Knidian vessels (Levine 2003:85). Tel Dor, fourth–second centuries BCE (Guz-Zilberstein 1995: fig. 6.1:14, 27, misinterpreted together with hemispherical bowl). Jericho, Hasmonean (Bar-Nathan 2002: pl. 14: 198 with disc base, discussed together with hemispherical bowls).

Plates (Fig. 2.6:13–19) Plates are common in Hellenistic assemblages, and yet at Kh. er-Rasm they are rare. Only five complete saucers and 24 sherds were discovered. The complete vessels do not draw a distributional pattern inside the building. In spite of the scarce nature of the class, the representative saucers of the Hellenistic assemblage are present and include plates made in the fish plate tradition (R-PL1), a large and deep plate (R-PL2) and shallow, flat plates with different kinds of ledge rims (R-PL3 and R-PL4).

R-BL3 Bowls with Horizontal “Pinched-bow” Handles (Fig. 2.6:11) Description: Thin-walled, globular small bowl/cup with simple everted short rim with an inner groove, probably to fit a lid, and a relatively small ring base. They used to have two pinched bow handles attached below the rim, but these did not survive. The clay was well levigated and ranges from pinkish-white (7.5YR 8/2) to light red (2.5YR 6/6) tone. Two examples were recovered, with a gray to reddish-gray slip.

The plates conform to the general date of the Hellenistic last level of occupation in the second century BCE.

R-PL1 (Fig. 2.6:13–15)

Type date range: Second century BCE–first half of first century BCE.

This group comprises the typical saucers of the Hellenistic period. The spots are shallow plates, which seem to have been produced during the Hellenistic period in “fish plate” tradition. Different variations of materials, sizes, rims and base forms have been identified throughout the Mediterranean area and beyond.

Provenance: No complete vessels were found; three sherds were discovered in sealed occupational debris. Occurrence: Maresha (Levine 2003:75, 77, 87, figs. 6.1:3– 4 and 6.3:50 for the un-slipped variation). At Maresha the painted examples are considered to be imports and based on parallels dated to the second century BCE (Levine 2003:75). Tel Gezer, contexts dated from the second to the mid-first centuries BCE (Gitin 1990: pl. 35:6, pl. 38:9-10; pl. 40:10; pl. 43:9, 12; pl. 42:6–8). Tel Dor, second century BCE (Guz-Zilberstein 1995: fig. 6.7:1–8).

Athens, Corinth and also Tel Dor, show an evolution from a shallow flat bowl with vertical rim and broad base to a deeper bowl with out-turned rim.9 The different subtypes are due to differences in rim and base treatment.

Discussion: Most scholars assume that this bowl is a derivation of the Attic and Hellenistic skyphos which was still in use during the first half of the second century BCE. Indeed, bowls parallel to ours have come from contexts dated to the second century BCE. They are widely distributed but do not appear in great quantities. At Tel Dor,

R-PL1A (Fig. 2.6:13) Description: Shallow saucer with a central depression. See the differences between BL4A and BL4b at Tel Dor (GuzZilberstein 1995:291, 295). 9

122

Avraham Faust and Adi Erlich

BCE area W and area X2, second century BCE (Geva 2003:138, pl. 5.3:BL2) and in Area A, first century BCE (Geva 2003:138, pl. 6.3:1–10). Maresha, second century BCE (Levine 2003:83, fig. 6.2:30). Jericho, J-PL1A3, very common in Hasmonean complexes and appears in Herodian period. It does not continue into the first century CE (BarNathan 2002:92–94, pl. 16:252–263).

Generally a raised ridge surrounds it. Ware: from pink (5R 7/4) to light red (10 R 6/6). A dull red matte slip covers the entire vessel. Type range date: This type of fish plate appears as early as the third century BCE–early second century BCE and is common through the second century BCE.

R-PL2 (Fig. 2.6:16–18)

Provenance: Two complete saucers were found in collapses (L.129; L.109); five sherds in collapse levels, mostly in the same area.

Description: Large, carinated, shallow plate with short outturned ledge rim, some pulled down, others rounded. Made in reddish-yellow (7.5YR 7/6) ware.

Occurrence: Maresha, 284–114 BCE (Levine 1999:45, pl. 9, No. 47). Samaria, context dated late in the second century BCE (Crowfoot, Crowfoot and Kenyon 1957:261, fig. 54:5–6). Jerusalem, Jewish Quarter, second century BCE (Geva 2003:144–145, fig. 5.3:30, Type FP3). Tel Dor, type BL4b, typical of the second century BCE (Guz-Zilberstein 1995:291–292, photos 6.7-6.8, 349, fig. 6.3:6–16; references to Tel Michal, and Ashdod there). Tel Iztaba, 107 BCE (Sandhaus in press).

Type date range: Third–second centuries BCE. Provenance: One complete vessel found in collapse debris (L.814); eight sherds in surface and collapse levels scattered widely in the site. Occurrence: Tel Dor, Type 14c, resembles kraters from the third and second centuries BCE (Guz-Zilberstein 1995:295, fig. 6.9:8–9).

R-PL1B (Fig. 2.6:14)

R-PL3 (Fig. 2.6:19)

Description: Flat shallow saucers with convex bottom, ring base, angular profile and thickened infolded rims. Produced in light red ware (10R 6/6) with white grits.

Description: Shallow flat plate with short ledge rim. Ware: reddish-yellow (5YR7/8) with white grits.

Type date range: Late fourth/third century BCE–second century BCE.

Type date range: Second century BCE. Provenance: One sherd found in surface.

Provenance: One complete plate in occupational debris L.154; two sherds in wall fall mixed with occupational debris.

Occurrence: Maresha, second century BCE (Levine 1999: pl. 8:27–29).

Occurrence: Samaria, Hellenistic Fort Wall, 220–150 BCE (Crowfoot, Crowfoot and Kenyon 1957:221, fig. 37:9–10) and in second century BCE contexts (Crowfoot, Crowfoot and Kenyon 1957:252, fig. 51:5–8). Tel Dor, in late fourth–third centuries BCE context and in second century BCE context (Guz-Zilberstein 1995:293, pl. 6.4:1–9, type BL5A). Maresha, 285–114 BCE, Type 5 (Levine 1999:43, pl. 8:34); Tel Iztaba, second half of second century BCE, type BS-NY.PL2b (Sandhaus in press). Jerusalem, Jewish Quarter, a few examples, second century BCE (Geva 2003:145, fig. 5.3, type FP4).

Personal Ware (Juglets, Unguentaria, Flasks) (Fig. 2.7) Juglets-Perfume Containers (Fig. 2.7:1–4) Three complete juglets and twenty four sherds are included in the Hellenistic corpus of pottery from Kh. er-Rasm. Most of the sherds came from surface or unstratified loci, but the complete vessels were discovered in collapses and sealed floor levels. The contribution to the dating of the last occupational level of the Hellenistic period is worthy of mention, since R-JT2 and R-JT3 first appear during the late second century BCE, thus providing a limited time frame during the last decades of the second century BCE for the Hellenistic assemblages (see Chapter 2.4).

R-PL1C (Fig. 2.6:15) Discussion: Thickened beveled rim and disc base. It has a very slight sunken depression on the inner side of the base that is almost invisible; manufactured in reddish-yellow ware (5YR6/6) with white and gray grits.

R-JT1 ‘Maresha Juglets’ (Fig. 2.7:1–2)

Type date range: Second century BCE–first century BCE.

Description: Angular pear-shaped body, disc base, handle attached from rim to shoulder, no neck, simple or thickened rim. The containers could be ribbed or plain, with no decoration or covered by reddish-brown slip. The ware

Provenance: One complete plate in collapse (L.150), seven sherds scattered widely in collapse levels. Occurrence: Jerusalem, Jewish Quarter, second century

123

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

tends to be fine, varying from reddish-yellow (5YR 6/6) to white color (10YR8/2).

Berlin (1997a:50) suggested a pouring purpose for these juglets, but I still prefer a personal use.

Type date range: Third–second centuries BCE.

R-JT3 (Fig. 2.7:4)

Provenance: One complete juglet in collapse (L.121); eight sherds scattered widely in collapse and surface levels.

Description: Juglet with globular body and cup-mouth rim. A twisted handle is attached from rim to shoulder; made in a range of colors varying from pink (5YR 7/3), reddishyellow (5YR 6/6) and red (2.5YR6/6). An almost invisible ribbing covers the body.

Occurrence: Maresha, cave system dated to the second century BCE (Levine 2003:110–112, fig. 6.13:136), 113/112 BCE (Kloner 1994:270).

Type date range: Late second century BCE–first century CE.

Discussion: This juglet is an apparently a Maresha product. Scholars have suggested that it was manufactured for a specific liquid; Kloner (1994:270) claimed that the juglets served for drawing liquids from collecting vats at Maresha, and Levine proposed that it was used for measuring (Levine 1999:78).

Provenance: One complete juglet was found in the sealed ash layer (L.818); six sherds widespread in debris and surface levels, mostly in Area C. Occurrence: Maresha, late second century BCE (Levine 2003:109, 111, fig. 6.13:126–127). Tel Gezer, Late Hellenistic, mid-first century BCE (Gitin 1990:87, pl. 44:2, type 187 A). Jerusalem, Jewish Quarter, typical in late second–beginning of first centuries BCE and in contexts dated to the first century CE (Geva 2003:130, pl. 5.2, type JT2). Tel Iztaba, 107 BCE (Sandhaus in press). Jericho, first century BCE to first century CE (Bar-Nathan 2002:53–54, pl. 10:87, JT1A1).

Concerning the origin of the form, a Persian prototype, a south Phoenician prototype and metal vessels have been suggested (Levine 1999:78). According to Levine, the juglet appears already in third century BCE contexts in the oil press and it was still in use during the late second century BCE, discovered with a coin hoard whose latest coins are dated to 111/112 BCE (Levine 1999:79; see also Kloner 1991:40).

Discussion: An extensive discussion of the origin, function, development and chronology appears in the Jericho report, summarizing almost all the information about the type (BarNathan 2002:48–49, 52–55). It seems that this type follows a local tradition, started during the end of the Iron Age and Persian period, when the juglets shared a “globular body form with thick ring rim” as suggested by Stern (1982:121– 122) and by Kahane (1953:48–50). Kahane raised a second possibility, which defines this type of juglet as a form made in the tradition of the Greek lekythos of the fifth century BCE (1953:50). Still, the former hypothesis, of an origin in local tradition, seems to be the favored one.

R-JT2 (Fig. 2.7:3) Description: Squat juglet with neck narrowing toward the top ending in a flaring everted thickened rim and a high disc base. A handle is attached from rim to shoulder. It was produced in a color ranging from red (5YR 5/6) to reddishyellow (5YR6/6) with small white grits. Sometimes the rim or half of the body was covered by red slip. Type date range: Second half of second century BCE. Provenance: One complete juglet was found in the collapse debris (L.119) and ten sherds were found throughout the site, mostly in surface and debris levels, eventually in collapse levels.

Unguentaria (Fig. 2.7:5–8) Fusiform unguentaria are spindle-shaped bottles whose use was related to the perfume or unguent industry. These bottles are known from domestic, public and burial contexts across the Greek world (Anderson-Stojanović 1987:106). They were produced in many wares, and according to their distribution throughout the Mediterranean, there seem to have been many manufacturing centers (AndersonStojanović 1987:105 and Hayes 1991:68). This assumption is strengthened by the local variations in each site or region.

Occurrence: Maresha, second century BCE (Levine 2003:109–111, fig. 6.13:130–132). Tel Dor, second half of second century BCE contexts (Guz-Zilberstein 1995: fig. 6.28:9–14, type JT2b). Tel Gezer, mid-second century BCE (Gitin 1990: pl. 37:6). Tel Iztaba, before 107 BCE (Sandhaus in press). Tel Anafa, late second century BCE contexts, HELL 2A–125 BCE (Berlin 1997a: pl. 10:PW53). Discussion: The most common juglet in the second half of the second century BCE. Its first appearance can be traced to the mid-second century BCE, since no sherds were included in pottery assemblages dated prior to the mid-second century BCE. The type disappeared during the end of the century as indicated by the absence of the type in first century BCE contexts.

Apparently, fusiform unguentaria with a tall foot first appear by the late fourth and early third centuries BCE in Greece.10 During the second century BCE the most common type was the elongated bottle with a long stem and a disc or string-cut base (R-UN1 and R-UN2). Many scholars tried In Athens in group B (Thompson 1934: B6) re-dated to the first quarter of the third century BCE (Rotroff 1987:102). 10

124

Avraham Faust and Adi Erlich

2.7: Plate of Personal Vessels

No.

Type

Basket

Locus

Context

Ware

1 2 3

R-JT1 R-JT1 R-JT2

1057.1 8101.8 1040.1

121 823 119

Wall fall in corridor Ash layer Topsoil mixed with wall fall

4

R-JT3

8077.1

818

Occupational debris

5 6

R-UN1 R-UN1

8071.1 8070.1

813 814

Wall fall Wall fall

7

R-UN1

8086.9

822

Occupational debris with ash layer

8

R-UN2

1036.1

115

Wall fall inside installation

Light red (2.5YR6/6) Exterior: white (10YR8/2); burnt Light red (2.5YR6/8) Pink (5YR7/3); exterior: red (2.5YR6/8) Light reddish-brown (2.5YR6/4) Reddish-yellow (5YR7/6) Light reddish-brown (2.5YR6/4); exterior: Reddish-brown (2.5YR5/4); burnt Light reddish-brown (2.5YR6/4); red slip

9

R-FK1

1083.1

134

Occupational debris and wall fall in corridor

125

Reddish-yellow (7.5YR7/6)

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

to propose different chronological developments in form but with dubious success.11 Different variations coexist in the same assemblages, the heavy type together with the finer one, coarse and fine ware, painted, slipped and plain. A change is evident during the first century BCE, when the unguentaria tend to be more elongated and thinner than the earlier ones (see at Jericho, type J-UN1, Bar-Nathan 2002:57–58, pl. 10:93–102).

third through the second centuries BCE (Guz-Zilberstein 1995: fig. 6.26:37–39, type UG 2e).

Flasks (Fig. 2.7:9) Two sherds of one flask (including rim, neck and handles, comprising about 1/4 of the vessel) were unearthed in a well defined stratigraphical context. Its importance lies in the fact that it is a diagnostic type of the second half of the second century BCE, providing a chronological anchor for the general dating of the Hellenistic assemblage (see Chapter 2.4).

UN1 (Fig. 2.7:5–7) Description: Fusiform unguentarium with oval body shape. It stands on an elongated stem, ending in a string-cut or disc base. The tall neck ends in an out-folded thickened or triangular rim. It is made in a variety of fabric colors from pink (7.5YR 8/3) to light red (10R 6/8). Some were made in a coarse ware including gray grits and some are well levigated. It is remarkably heavy; sometimes, on the upper body and lower neck, red-painted bands decorate the vessels.

FK1 (Fig. 2.7:10) Description: Flask with cylindrical wide neck and thickened rim. Two twisted handles are attached from neck to shoulder; made in reddish-yellow ware (7.5YR7/6) with white grits.

Type date range: Second century BCE.

Type date range: Mid-second century BCE–beginning of first century BCE.

Provenance: Three complete vessels were found in collapse debris: L.130 and the two equal loci L.813 and L.814. Four complete vessels were found in occupational debris, L.186, L.8181, L.822, and L.821; two rims in collapse debris.

Provenance: Two pieces including rim, neck and handles were discovered in collapse level above the floor (L.130) and on occupational debris (L.134); six rim sherds widely scattered throughout all the levels.

Occurrence: Maresha, complexes 21 and 70, third–second centuries BCE (Levine 2003:114–115, fig. 6.14:148–149); typical of second century BCE contexts (Levine 1999: fig. 19:175–178). Tel Dor, second century BCE phases, most common in second half of second century BCE (GuzZilberstein 1995:305, fig. 6.26:23–31, type UG2c). Tirat Yehuda, second century BCE (Yeivin and Edelstein 1970: fig. 7:17). Jerusalem, Jewish Quarter, second century BCE contexts (Geva 2003:131, pl. 5.2:46, type FU2). Tel Iztaba, 107 BCE (Sandhaus forthcoming, type BS-NY. UN4). Pella, west cut, 150–85 BCE (McNicoll, Smith and Hennessy 1982:140, pl. 130. 11).

Occurrence: Maresha, third–second centuries BCE (Levine 2003:99–100, fig. 6.9:91). Tel Gezer, contexts dated to the late second century and early first century BCE (Gitin 1990: pls. 39:18, 41:24–25). Jerusalem, Jewish Quarter, Stratum 4 in area W and Stratum 6-5 in area X-2, dated to late second–early first centuries BCE (Geva 2003:128, fig. 5.2: type FK2). Discussion: Flasks are well known vessels, which first appear during the Iron Age. In the Persian and Early Hellenistic periods the flask was thick-walled, its wide neck flaring outwards and the handles were attached close to the rim (Stern 1982:166, type A and Geva 2003:128, photo. 5.2). During the course of the second century BCE a change took place, the vessel became thinner though still thick and rather coarse, the neck became vertical and the handles tend to descend to the middle of the neck. The next development of the type is represented by the flasks of the first century BCE and first century CE which became fine vessels, made with thin walls and well levigated clay, the neck is now tall and narrow and the handles are attached to the middle of the neck (Bar-Nathan 2002:66, J-FK1). During the first century CE a pronounced ridge appears in the attachment point of the handles and the necks.

UN2 (Fig. 2.7:8) Description: Fusiform unguentarium with oval body shape. It stands on a long stem, ending in a disc base. The tall neck ends in a triangular rim. It is made of a light reddish-brown (10R 6/8) well levigated ware and belongs to the heavy type. It is covered by red slip. Type date range: From the end of the third–second centuries BCE; dominant type during the second century BCE. Provenance: One complete vessel was found in collapse (L.115) and one rim in an unsealed locus.

2.2.2 The Pottery from Roman, Byzantine and Later Periods (Fig. 2.8)

Occurrence: Maresha, second century BCE (Levine 2003:114–115, fig. 6.14:146). Tel Dor, from the end of the

Storage Jars (Fig. 2.8:1–5) R-SJ6 (Fig. 2.8:1)

Lapp 1961:21, type 91.1 in comparison to type 91.2; Thompson 1934:472–474. 11

126

Avraham Faust and Adi Erlich

Description: Sherd of collared rim storage jars; made in pink ware with gray grits.

Description: Hole-mouth jar with no neck, but with thickened rim and ridged body. The vessel was manufactured in light red ware.

Type date range: Herodian period, 31 BCE–6 CE.

Type date range: End of sixth–seventh century CE.

Provenance: Five sherds found in surface and debris.

Provenance: Only one sherd was discovered (chance find on the surface).

Occurrence: Jericho, Herodian I, 31–15 BCE (Bar-Nathan 2002:33–34, 256, pl. 6:39, Type J-SJ7a). Masada, 37/31 BCE–6 CE, and reused by the Zealots 66–73/4 CE (BarNathan 2006:54, pl. 4:19, Type M-SJ7a).

Occurrence and Discussion: The “Gaza” jar is a well known type of jar produced from the first century BCE to the seventh century CE. During these centuries the vessel changed its shape. The sample from Kh. er-Rasm is the typical later type, dated to the end of the sixth–seventh centuries CE.

R-SJ7 (Fig. 2.8:2) Description: Thin walls, simple cut-in rim, ridge on the lower part of the neck; produced in light red ware.

The “Gaza” jar has been studied and published intensively during recent decades. The first typology was offered by Ann Killebrew but was never published. Gaza jars from Ashkelon have been published by Johnson and Stager (1995:95–109); Majcherek (1995) offered a new typology based on the finds from Alexandria and Oked offered an economic overview and a more detailed typology based on the finds from north Sinai (Oked 2001:227–250).12

Type date range: End of first century BCE–beginning of second century CE. Provenance: Four sherds discovered in various types of loci. Occurrence: Jericho, Herodian contexts, 31 BCE–73 CE (Bar-Nathan 2002:34, pl. 7:42, type J-SJ7B2).

“Gaza” jars were manufactured on the southern coast, from Sheikh Zuweid in north Sinai up to Ashdod in the north (Oked 2001:227).

Discussion: This is the most common storage jar at the end of the first century BCE and it continued to appear during the first century CE, when additional rim forms joined it (Bar-Nathan 2002:34). After 70 CE, the type appeared in Jericho (Bar-Nathan and Gitler-Kamil 2002:152) and in the Cave of the Treasure (Bar Adon 1980:209, Nos. 2–3). According to Bar-Nathan and Gitler-Kamil, during the Second Temple period this type is widespread in Judea, as well as in the coastal area and the Galilee (Bar-Nathan and Gitler-Kamil 2002:152).

The importance of the jars lies in their contents. From written sources and archaeological remains such as agricultural installations, wine production, fish pools from the Byzantine period, we learn that they were used to hold several types of contents (rather than only wine), including fish and perhaps olive oil (Mayerson 1992:79). Workshops for Gaza jars discovered close to agricultural facilities suggest that they were in use by farmers for the purpose of exporting locally produced goods. But it seems that the goods carried in these jars originated from Judea and the Negev, too (Mayerson 1992:76; Blakely 1988:38).

R-SJ8 (Fig. 2.8:3) Description: Thin walls, triangular profiled rim, ridge on the lower part of the neck. Manufactured in pinkish-white ware with a light red core, including white grits.

The type found in Kh. er-Rasm was found all over the Mediterranean, from Alexandria, Egypt (Majcherek 1995:166) through Caesarea, on the Israeli coast (Magness 1995:144; Riley 1975:31, no. 13), to Spain (Keay 1984:121, no. 1) and more.13

Type date range: First century CE to early second century CE (130/32 CE). Provenance: One sherd in surface. Occurrence: Jerusalem, Jewish Quarter, area A, Stratum 5–4, first century CE and in destruction level dated to 70 CE (Geva and Rosenthal-Heginbottom 2003: pl. 6.5:7). Masada, contexts dated to 66/7–73/74 CE, but the type date ranges from the first century CE to the first third of second century CE (Bar-Nathan 2006: pl. 14:39–42, type M-SJ8).

R-SJ10 (Fig. 2.8:5) Description: Storage jar with thin walls, conical long neck and out-folded pulled down rim with triangular profile; produced in reddish-brown ware. Type date range: Undefined.

Discussion: The ridge at the bottom of the neck is a distinctive feature of first century CE storage jars.

Provenance: One sherd in Area C, L.509.

R-SJ9 “Gaza” Jar (Fig. 2.8:4)

For further information including updated references, see Oked 2001:227–250. 13 For further parallels, see ibid.:241. 12

The term “Gaza” jars I used here for products of both Gaza and Ashkelon. 127

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

2.8: Plate of Diagnostic Sherds from Later Periods

No.

Type

Basket

Locus

Context

Ware

1

R-SJ6

3048.1

321

Wall fall

2

R-SJ7

3095.2

330

Topsoil

3

R-SJ8

8097.1

820

Topsoil

4 5 6 7 8 9 10 11

R-SJ9 R-SJ10 R-JG6 R-CP9 R-CS3 R-BL5 R-BL6 R-BL7

3049.1 1153.1 1114.1 1217.2 1101.1 9901.1 9999.1 9901.2

322 155 148 172 143 990 999 990

Topsoil Topsoil Topsoil mixed with wall fall Topsoil mixed with wall fall Topsoil Surface Surface Surface

12

R-BL8

7004.1

705

Fill in cave

13

R-BS1

3095.1

330

Topsoil

128

Pinkish-gray (7.5YR7/2); core: pale red (2.5YR6/2 Light red (2.5YR6/6) Light red (10R6/6); exterior: pinkishwhite (5YR8/2) Light red (2.5YR6/6) Reddish-brown (2.5YR 4/6) dark reddish-gray (5YR4/2); gritty Red (2.5YR4/8) red (10R5/8); gray and mica grits reddish-yellow ware (5YR7/6) Brown Reddish-brown (2.5YR 4/6); red slip Pale yellow (2.5Y8/3); pale yellow glaze White (10YR8/1)

Avraham Faust and Adi Erlich

Jugs (Fig. 2.8:6)

as the ones from Masada (Bar-Nathan 2006: pl. 31:80–84) and developing into the typical casserole with everted walls and two horizontal handles attached to the body. This type continued all over the Byzantine and the Umayyad periods. Differences can be traced through the different periods. While during the first centuries the bowls were made with thin unribbed walls and small loop handles, during the Early Byzantine period the ware tends to be coarser and ribbing starts to cover the walls and base, alongside casseroles without ribbing which continued to appear during the entire Byzantine period. Late Byzantine examples became deeper and coarser. The handles grew in proportion to the vessel. Umayyad casseroles became deeper and the handles grew again. It is not always possible to distinguish the different nuances, especially concerning the later types. The Kh. erRasm specimen, with the everted flat walls, still thin, and proportionately relatively small handles, seems to belong to the third–early fifth centuries CE variant.

R-JG6 (Fig. 2.8:6) Description: Jug with swollen neck, thin-walled cooking jug with pronounced tall concave neck ending in a simple incurved ridged rim. Type date range: 13th to 15th centuries or even later. Provenance: One sherd in collapse, mixed with topsoil. Occurrence: Avissar and Stern 2005:108–110, fig. 45:4–5.

Cooking Pots (Fig. 2.8:7–8) R-CP9 (Fig. 2.8:7) Description: Large globular cooking pot with thin walls, dense ribbing, short vertical neck and simple rim with a groove on outer side. Two ridged handles attached from rim to shoulder. The type was produced in a reddish-yellow (5YR6/6) gritty ware and an inner reddish-gray (5YR5/2) core.

Bowls (Fig. 2.8:9–12) R-BL5 (Fig. 2.8:9) Description: Sherd of a “rilled-rim bowl” made in reddishyellow ware (5YR7/6) with a groove on the surface of the rim ending in a hook.

Type date range: Late first century BCE–early second century CE.

Type date range: 200 CE to the first half of the sixth century CE.

Provenance: One complete cooking pot in locus L.172, and two almost complete cooking pots in Area C (L.501, L.505, L.506, L.507, L.501, L.515).

Provenance: Survey.

Occurrence: Jericho, 31 BCE–48 CE (Bar-Nathan 2002:171, pls. 26:476–479, types J-CP2C1 and J-CP2C2). Masada, last third of first century BCE to first third of second century CE (Bar-Nathan 2006:154–158, pl. 27–28, nos. 6–25, M-CP1B).

Occurrence: Jerusalem, late third/early fourth–sixth centuries CE (Magness 1993:203); Jerusalem, Binyanei Ha’uma, end of second century CE (Magness 2005:147, fig. 27:3); Jerusalem, Jewish Quarter, 200 CE–sixth century CE (Magness 2006:184–191, pl. 7.1:9).

Discussion: The type is the Herodian cooking pot par excellence. It represents the most popular cooking pot during the Early Roman period. The Masada report presents an exhaustive discussion, which is innovative in that it broadens the range of the type to the first third of the second century CE (for further discussion see Bar-Nathan 2006:154–158).

Discussion: This type is common in Jerusalem and its environs. According to Magness (2005:105), “Jerusalem rilled-rim” deep bowls were inspired by the handle-less Roman cooking cauldrons and the deep basins and mortaria. Updated chronology based on the finds from the pottery workshop at Binyanei Ha’uma in Jerusalem (Magness 2005: fig. 27:3) moved up the date of its appearance to 200 CE (ibid.:105–106). This type of deep bowls or basins, together with other Jerusalem types, continued to be produced with little variation until the first half of the sixth century CE (ibid.).

R-CS3 (Fig. 2.8:8) Description: Casserole with flaring walls, simple flattened rim and two horizontal handles attached to the walls. Ware: red (10R5/8) with gray and mica grits; gritty.

R-BL6 (Fig. 2.8:10)

Type date range: Third–early fifth centuries CE.

Description: Sherd of a big bowl/basin with straight walls, made in brown ware with a ledge rim ending in a hook.

Provenance: One sherd in collapse unsealed debris. Occurrence and Discussion: The casserole is common in all sites with Late Roman-Early Byzantine phases. It belongs to a longer local tradition of casseroles, beginning during the Roman period with various kind of cooking bowls, such

Type date range: Second/early third–sixth centuries CE. Provenance: Survey.

129

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

Basins (Fig. 2.8:13)

Occurrence: Jerusalem, late third/early fourth–sixth centuries CE (Magness 1993:203); Jewish Quarter, late third/early fourth–sixth centuries CE (Magness 2003:428, pl. 18.2:10); Jerusalem, Binyanei Ha’uma, second–third centuries CE (Magness 2005: fig. 29:7–8).

R-BS1 (Fig. 2.8:13) Description: Deep, handmade basin made in coarse white (10YR8/1) ware, decorated with combed horizontal lines and incised perpendicular ones, with bolded rim.

Discussion: Like the type discussed above (R-BL5), this type of big bowls or basins is well known in the Jerusalem area and was dated, after the publication of the assemblages from the pottery workshop in Binyanei Ha’uma, to 200 CE–sixth century CE (Magness 2005:69–194). This type is very common all over the Roman Mediterranean during the second and third centuries CE (ibid.:97). It continued to be produced with little changes until the first half of the sixth century CE (ibid.:97, 105–106).

Type date range: Umayyad–Abbasid periods. Provenance: Surface. Occurrence and Discussion: Handmade basins adorned with combed decoration started to appear during the Late Roman period and they are still produced today. During the Roman and Byzantine times the basins were made mostly in reddish-yellow, orange clay. At the end of the Byzantine period, gray ware handmade basins were produced. During the ninth and tenth centuries the gray clay was replaced by a buff, whitish one (see Bar-Nathan et al. in press). The basins are widely dispersed all over the country and appear on both sides of the Jordan Valley.14

R-BL7 (Fig. 2.8:11) Description: Sherd of a bowl made in reddish-brown clay, covered by red slip, with simple rim and a groove on the outer surface below it. Type date range: Early Islamic period.

Summary

Provenance: Survey.

Although the pottery discovered at Kh. er-Rasm covers a long period of time, most of it dates to the Hellenistic period, and especially to the second century BCE, when the site was destroyed (and see below, Chapter 2.4).

Occurrence: Jerusalem, similar bowl with no slip, second half of the sixth–eighth centuries CE (Weksler-Bdolah 2006:103); Fine Byzantine Bowls (FBW) Form 2B, midseventh to ninth/tenth centuries CE (Magness 1993:199).

The assemblage is characteristic of a domestic household, and it suits the isolated nature of the building; it is completely local in nature, a theory strengthened by the total absence of imports (except for one Attic lamp, see below chapter 2.3).

Discussion: This type is similar to the fine ware bowls “African Red Slip – Form 99” (Hayes 1972:155, fig. 28). The imported prototype was dated from the fifth to the early seventh centuries CE. During the sixth–eighth centuries CE, local fine ware appears all over the area, imitating some of the imported earlier wares which tend to disappear during the eighth century CE. The new local fine wares replaced the imported wares. R-BL7 looks also very similar to “Fine Byzantine Bowls – Form 2B” after Magness (1993:199) and dated to the mid-seventh to ninth/tenth centuries CE. The sherd is too small and it is impossible to define whether it corresponds to either one of the previously named types, but judging by the similarity between them we suggest a date in the Early Islamic period.

The most abundant types of vessels include cooking pots and bowls. Worthy of note is the small number of complete storage jars – only three. A further dig in the unexcavated squares may reveal more storage jars. Personal containers, mostly unguentaria, are well represented in the assemblage. The evidence seems to indicate the domestic nature of the building (see Chapter 4.2 for a further suggestion). Two main focuses of activity can be traced, one in the southwest section, where complete table vessels constitute the majority of the finds; and the second concentrated in the northern row of rooms. Unlike the activity in the southern area, it seems that the northern section was designated for cooking, since cooking pots constitute the most abundant type (and see Chapter 4.2).

R-BL8 (Fig. 2.8:12) Description: Bowl with out-turned rim made in pale yellow ware (2.5Y8/3). The internal surface was covered by a lead glaze in pale yellow that splashes toward the rim. Type date range: Ninth–first half of tenth centuries CE.

To sum up, the evidence emerging from the pottery analysis indicates that during the last decades of the second century BCE the place was destroyed and abandoned. During the Herodian period some kind of activity took place at the site, as evident by a few types recorded in room 8, to the north of W.22, in the western aisle, and in the stepped stone

Provenance: Surface. Occurrence: Caesarea, stage VI, dated to the end of the ninth–first half of tenth century CE (Arnon 2003:141, pls. 90:1–6); Tiberias, ninth century CE assemblages (Stacey 2004:113, fig. 5.23:3).

For more information see Bar-Nathan et al., in press, The Umayyad Workshop at Beisan. 14

130

Avraham Faust and Adi Erlich

building. The residual finds related to the Byzantine and Islamic period have no architectural reference.

Kh. er-Rasm (Nos. 11–19, Figs. 2.9:11, 13, 15, 16, 17), and one almost complete specimen bearing a “radial” decoration (No. 20, Fig. 2.9:20).

2.3 Oil Lamps from Khirbet er-Rasm

Among the “closed” lamps, the earliest specimen identified at Kh. er-Rasm is the lamp commonly known as the globular (Attic) lamp (No. 5). These lamps were abundant in the Hellenistic world and have been found in the Southern Levant: along the Mediterranean coast, as well as at Tel Dor (Rosenthal-Heginbottom 1995:234, fig. 5.13, No. 5,) and at Apollonia (Fischer and Tal 1999:241, fig 5.15:22); inland they have been found at Yoqne‛am (Avissar 1996b:188, Nos. 1–3); Samaria (Crowfoot, Crowfoot and Kenyon 1957:365–367, fig. 85, Nos. 2 and 3); and in the Shephelah at Maresha (Ambar-Armon 2007:73–84) and at Gezer (Macalister 1912: pl. CLXXXIII: 5). These lamps are characterized by a very well levigated fabric and a very glossy black glaze. The lamp fragment found at Kh. erRasm belongs to Howland type 25, and it is characterized, amongst other things, by its small wick-hole and thick nozzle with a flat top.

Einat Ambar-Armon The excavations at Kh. er-Rasm produced 25 oil lamps (complete and fragments) (Fig. 2.9). The lamps belong to two main types: open lamps (Nos. 1–4) and closed lamps (Nos. 5–20), with five additional fragments (Nos. 21–25) that are too small to be identified. The lamps were found in 21 different loci. Below is an introduction to the typology, followed by a table with details on the individual lamps. The assemblage of lamps presented in this report embodies a diversity of types that reflects not only the chronological aspect but other aspects as well.

Open Lamps Four fragments of the open lamp type were found (Nos. 1–4 in the table, Fig. 2.9:4). These fragments are very small, making it extremely difficult to assess their original forms. The shoulders of these lamps were not preserved and no evidence of knife shaving, a phenomenon characteristic of Persian period lamps, was discernible. However, based on the shape of the pinched wick-hole and the relative size of the fragments, it can be presumed that some subtypes could be represented here: the “saucer” lamp characteristic of the Persian period, and the widespread Hellenistic lamp known as the “folded,” “pinched” or “Hasmonean” type. Part of this small group might also be a transition type between the “saucer” lamp and the “folded” lamp, which was also common in the Hellenistic period (for example see Geva 2003:139–140).

Howland, in his volume on the lamps from the Athenian Agora, divides this lamp type into six subtypes (Howland 1958:67–84). The lamp found in Kh. er-Rasm could belong to either one of these two subtypes. The first is subtype 25A which is characterized by a globular body (ball shaped), a deep unglazed groove that encircles the filling hole of the lamp, and a thick heavy base with a raised concave cone-shaped indentation. This subtype is dated by Howland from the middle of second quarter of the fourth century BCE to the first quarter of the third century BCE (Howland 1958:67). The second classification to which this lamp may belong is Type 25B, which has the same characteristics, but with an element new to this period: the addition of a pierced knob attached to the body of the lamp. This type is dated to the second half of the fourth century BCE to the first quarter of the third century BCE (Howland 1958:72–74). Howland’s dating is still commonly accepted, at least for these two types. This should also be the time span for the globular (Attic) lamp from Kh. er-Rasm, even though it could be from a slightly later period.

As mentioned above, saucer lamps are characteristic of the Persian period and are widespread throughout the Levant (i.e., Maresha: Levine 2003:115, fig. 6.15: 151; AmbarArmon 2007:49–52). Apparently this form was still in use during the Early Hellenistic period (i.e., Apollonia: Fisher and Tal 1999:232, fig. 5.8, No. 8) – a fact that makes it difficult to date these fragments more precisely. Folded lamps are commonly found throughout Judaea and are characteristic of the second and first centuries BCE (BarNathan 2002:102–104). Only small quantities of these lamps were found outside Judaea, for example at Ashdod (Dothan and Freedman 1967:23, fig. 4, No. 11), at Samaria (Reisner 1924:318, fig. 187, No. 5a) and at Maresha (Ambar-Armon 2007:55–61). However, as noted above, the fragments discussed here are extremely small, making it difficult to identify them further or to provide a more precise chronology.

Included amongst the Kh. er-Rasm lamps are lamps which imitate the globular (Attic) lamp. On the whole, these local imitations have smaller bodies than their Attic prototypes. Despite the small size of the fragments that have survived, these pieces of lamp bodies are indicative enough to enable their identification as imitations of the globular (Attic) lamps (Nos. 6–10). For instance, they have the characteristic deep groove encircling the filling hole at a distance and the thick flat-topped nozzle with a small wickhole. This type of lamp is common in the Levant, and also in the Judean Shephelah, i.e., Lachish (Tufnell 1953:286, pl. 82, No. 137) and Maresha (Ambar-Armon 2007:80–84). Rosenthal-Heginbottom dates lamps similar to these from Tel Dor to the middle of the fourth century to the second century BCE (Rosenthal-Heginbottom 1995:235). Likewise the lamps of this type from Kh. er-Rasm should be dated to

Closed Lamps This collection includes a globular (Attic) lamp fragment (No. 5, Fig. 2.9:5); five fragments of locally made globular (Attic) lamp imitations (Nos. 6–10 Fig. 2.9:7); nine fragments of local types common to the region surrounding

131

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

2.9: Plate of Lamps

No.

Type

Basket

Locus

Context

4

Open lamp

8030.04

803

Topsoil

5

Globular (Attic) lamp

1212.01

170

Topsoil mixed with wall fall in square I53

8039.01

808

Fill in cell 18

8101.03 8013.01

823 805

Ash layer in square H48 Wall fall in cell 18 Wall fall and occupational debris in square I53 Occupational debris with ash layer in square G48

11 13

Globular (Attic) lamp imitation Wheel-made closed lamps Wheel-made closed lamps

15

Wheel-made closed lamps

1258.01

186

16

Wheel-made closed lamps

8085.02

818

17

Wheel-made closed lamps

1073.01

129

Wall fall in corridor 6

20

Mould-made lamp

8012.01

802

Topsoil

7

132

Ware Reddish brown clay with tiny and crowded white grits Attic Clay (pink to brown) with good black glaze inside and outside. Orange clay with sepia grits, little blackened wick-hole Buff clay, white grits Buff clay, patina Buff clay, Tiny white grits, Blackened wick-hole Reddish brown clay, patina, Blackened wick-hole Brown- gray clay with white grits, Blackened wick-hole Reddish clay

Avraham Faust and Adi Erlich

the end of the fourth–third centuries BCE and perhaps to the beginning of the second century BCE as well.

The collection also includes five fragments, which are unidentifiable due to their small size. It is possible that fragment No. 25 is a fragment of a terracotta figurine, though the shape makes it impossible to exclude it from this collection. This might be the lower part of a small lamp that includes part of the nozzle, base and the lower part of the conventional knob. Whether this is a figurine or a lamp, it was broken at the link between the two sections, which means that it was made with a mould.

The bulk of the Kh. er-Rasm corpus consists of lamps that continue the Hellenistic tradition of wheel-made “closed” lamps (Nos. 11–19). These include two complete wheelmade lamps, one whole and six fragments. These lamps are known as “sandal-shaped lamps” (Stern 1982:129), “neoBabylonian” lamps (Sussman 2004:152), or “Shephelah lamps” (Levine 2003:115). They are characterized by a large filling hole surrounded by a high shelf, a nozzle that narrows towards the end with a base formed by the string cut process. Two subtypes may be distinguished according to the profile of the body. One has a convex wall, displaying a profile reminiscent of that of the globular (Attic) lamps (No. 11). The middle of the body of the other subtype is characterized by a straight wall (Nos. 12–17). Two more small fragments were found belonging to this type; however they cannot be included with certainty in one of the two subtypes noted above (Nos. 18, 19).

*** In conclusion, the lamps found during the course of the excavations at Kh. er-Rasm are predominantly small fragments. However, the diversity of identifiable types reflects the continuity of the site’s occupation as well as implying a regional uniqueness, and providing some knowledge about the characteristics of the population. The lamp finds from Kh. er-Rasm suggest that the site was occupied for most of the Hellenistic period. The earliest lamps could be from the Persian period or from the earlier part of the Hellenistic period. The abandonment of the site, judging by the lamps found at the site, probably took place at the end of the second century BCE. It is possible that the site was deserted earlier, at some point during the second century BCE. The absence of lamps later than the second century BCE indicates that occupation at Kh. er-Rasm could not have continued later than the early first century BCE.

These lamps are characteristic of the Shephelah region, and have been found at sites such as: Tel Beth Shemesh/ Ain Shems (Grant and Wright 1939: pl. 50:23), Gezer (Macalister 1912: pl. CLXXXIII: 4), Lachish (Tufnell 1953:286–287, pl. 82, No. 138), Tirat Yehuda (Yevin and Edelstein 1970:66, fig. 11:5–6, pl. 7), and Tell Sandahanah\ Maresha (Bliss and Macalister 1902:129, pl. 62, No. 1; Ambar-Armon 2007:92–98). Rosenthal and Sivan dated these lamps to the second and first centuries BCE (1978:78). In Maresha, these lamps were in use until the end of the second century BCE, at the time of the conquest of John Hyrcanus I. The earliest dating of these lamps is still not determined, but the majority of them were in use during the second century BCE. We are not aware of any site from the first century BCE which includes lamps of this type.

The lamp findings represent different phases of the Hellenistic period. There is a distinction between the Early Hellenistic period lamps and lamps that are dated later in the Hellenistic period. From the Early Hellenistic period (and possibly the last part of the Persian period) the lamps represented at Kh. er-Rasm are the open, globular (Attic) lamp type, and imitations of the globular (Attic) lamp type. These lamps are similar to contemporaneous lamps from other areas in the Land of Israel, for example coastal sites such as Tel Dor, Apollonia, inland sites such as Yoqne’am, and in the Shephelah at Maresha, Tirat Yehuda, and Gezer. The Kh. er-Rasm lamps from the latter part of the Hellenistic period are characterized by a large filling hole surrounded by a high shelf. These lamps are characteristic of sites in the Shephelah and are rarely found beyond this region. These lamps, which were also found in good preservation conditions, represent the last phase in the site.

The Kh. er-Rasm corpus includes one almost complete mould-made lamp (No. 20) bearing a radial decoration that circles the filling hole of the lamp. The fragment of the preserved lamp incorporates the characteristic knob typical of such lamps from the Hellenistic period. This knob was commonly formed as a S-Coil (line in the shape of the letter S), and on the Kh. er-Rasm lamp only the remnants of a S-Coil can be discerned. The knob is found on the bottom of the lamp (when the nozzle faces left) and its presence together with the general outline implies that this lamp may have borne a similar knob on its upper side, e.g., as in parallel lamps from Maresha (Ambar-Armon 2007:119–121) and from the Jewish Quarter excavations in the Old City of Jerusalem (Rosenthal-Heginbottom 2003:196, pl. 6.4:10). Although only part of the lamp’s base has survived, it is still possible to make out that its form was flat with a groove in the center. In spite of its extremely poor condition, it is possible that the base of this lamp was decorated with a sun pattern, like in Maresha (Ambar-Armon 2007: fig. 81:17–19). A similar lamp was found in area 100 in Maresha and was dated by context to the second half of the second century BCE (Ambar-Armon 2007:123, lamp no. 1780).

An additional point of interest about the lamps from Kh. er-Rasm is related to the manufacturing technique. The percentage of mould-made lamps is significantly lower than that of wheel-made lamps (one, or possibly two out of 25). An examination performed as part of the author’s doctoral thesis about oil lamps in the Land of Israel during the Hellenistic period in light of the finds from the Maresha excavations (Ambar-Armon 2007), shows that there are sites dated to the second century BCE which contain a significantly large number of mould-made lamps, while other sites contain a larger number of wheel-made lamps. For example, at Tel Dor (Rosenthal-Heginbottom 1995)

133

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

and Samaria (Reisner, Fisher and Lyon 1924; Crowfoot, Crowfoot and Kenyon 1957) the finds consist mostly of mould-made lamps, while in other sites, like the Jewish Quarter (Geva 2003) and Mount Gerizim (Magen 1993) there are mainly wheel-made lamps.

way, while other populations preferred to use mould-made lamps, which was a relatively “new” technique, at least in its use for oil lamp manufacturing (for “conservatism” as a cultural phenomenon, see Levy 1998; Faust 2005c, and additional references).16 If this assumption is correct, then we may conclude that the population at Kh. er-Rasm preferred the “traditional” technique over the relativity new technique of lamp manufacturing. This interpretation is reinforced by the absence, or the small quantity, of imported pottery, mould bowls and terracotta figurines that were found at this site. The lamps from Kh. er-Rasm are also interesting, compared to similar research on the oil lamps from Maresha (the distribution of lamps at Maresha shows areas in which more wheel-made lamps were found while in other areas more mould lamps were unearthed). The lamps from Kh. er-Rasm, together with wheel-made lamps from Maresha and other sites, can contribute to the research on the population and its culture during the Hellenistic period in the Shephelah. In any case it is curious that in the period when mould lamps were common, the use of wheel-made lamps still prevailed at Kh. er-Rasm.

There are several possible explanations for these differences. They can be related to regional proximity, economic considerations or manufacturing areas. However, an examination of these explanations shows that they are not reasonable. Regional proximity examination shows that mould-made lamps were used mainly in the coastal area, while wheel-made lamps were used mainly in the mountainous area, such as Jerusalem. Furthermore, detailed site examinations indicate differences in sites which are close to each other. At Mount Gerizim, mainly wheel-made lamps were found, while in the nearby site of Samaria there were mainly mould-made lamps. An examination of the economic aspect also eliminates this as a possible explanation. In the Jerusalem area, where most of the oil lamps were wheel-made lamps, the population was mixed, wealthy and poor. Similarly, it is not reasonable to assume that in major coastal sites like Akko, Tel Dor and Jaffa the economic status of the population was homogeneous.

2.4 Dating the Hellenistic Ceramic Assemblage of Khirbet er-Rasm Débora Sandhaus, Adi Erlich and Avraham Faust

An examination of the manufacturing areas does not provide a reasonable explanation either. Adan-Bayewitz proposed, regarding the Roman period in the Galilee, that mould-made lamps were made in the cities and wheel-made pottery, which required a larger amount of clay, were made in the rural areas (Adan-Bayewitz 1995). While in the Roman period (excluding the beginning) the main use was of mould-made lamps, in the Hellenistic period there is evidence for widespread use of both types of lamps. Thus, if the solution was related to the manufacturing areas, we would expect to find lamps of different techniques in all areas, while de facto, this is not the case. It seems that production was not affected by the workshops, but that the workshops were affected by demand; the population could easily manufacture objects using either method. Wheel technique was known in all the areas and was used to manufacture pottery vessels. The mould technique was already known in ancient times and the population in Jerusalem or Mount Gerizim had at least the same technological capabilities as other populations.15 Thus it seems that the non-use of a specific type of lamp is a result of a certain fashion, i.e., cultural inclinations, and not the lack of capabilities.

The present chapter aims at dating the last phase of usage in the main building and the destruction of the Hellenistic occupation at Kh. er-Rasm. While the pottery discovered at Kh. er-Rasm covers a very long period of time, including the Iron Age, as well as the Persian, Hellenistic, Roman, Byzantine and the Medieval periods (see Shai, Chapter 2.1 above and Sandhaus, Chapter 2.2 above. See also Chapters 4.1–4.3, below), it is clear that the main era in which remains were unearthed is the Hellenistic period. Almost all the intact and complete17 pottery vessels unearthed at Kh. er-Rasm (well over 50 vessels) are dated to this period, and this assemblage of complete vessels is the subject of the present chapter (Figs. 2.10–2.14). The rich Hellenistic assemblage of complete vessels is important for dating the last phase of occupation in the Hellenistic period. During this phase the site was abandoned and the cheap pottery vessels were left behind, unlike more precious objects, and later on the site was destroyed and pottery was smashed and buried in the destruction. Notably, while the pottery sherds unearthed at the site

We would therefore like to suggest another possible explanation for these differences. The number of lamps that were made by each technique is mainly related to the preference of the local population. It is possible that cultural or ethnic preferences caused certain populations to use wheel-made lamps which were made in a “traditional”

Interestingly, the stone objects unearthed at Kh. er-Rasm (Peshin, Chapter 3.3) include technically sophisticated millstones, which according to Peshin indicate Hellenistic influence. We should therefore note that a group can adopt advanced technology while at the same time maintaining a conservative material culture (see Faust 2005c, and additional references). 17 By “complete” vessels we refer to vessels of which the entire profile or most of the pot was unearthed. Since no balk was removed, and no room was excavated in its entirety, we cannot expect to find all the sherds. Hence the above is sufficient to infer that the vessel existed when the site was destroyed (the logic behind an examination of this sort will be further discussed in section 4, Chapter 4.2 below). 16

Strengthening this conclusion are the finds from Maresha, where there is evidence of local manufacturing of wheel made lamps as well as mouldmade lamps. 15

134

Avraham Faust and Adi Erlich

Table 2.1: Open Lamps (Saucer Lamps/Pinched Lamps) Color Grey clay with tiny white grits, blackened wick-hole Greenish clay with grey grits

Measures L. (P) 3.8 H. (P) 2.2 W. (P) 0.6 L. (P) 2.9 W. (P) 1.8

Bright clay with pink core and white grits, blackened wick-hole

L. (P) 3.9 W. (P) 1.9

Reddish brown clay with tiny and crowded white grits

L. (P) 6.5 H. (P) 5 W. (P) 5.2

Description

Basket

Locus

Area and Square

Serial Number

Fragment of spout

3052.01

326

B, E54

1

Lip fragment

3111.01

330

B, E51

2

1136.01

153

A, H49

3

8030.04

803

E, E49

4

Lip fragment, includes part of the spout Lip fragment with part of the base, restored

Table 2.2: Closed Wheel-made Lamps Color Attic Clay (pink to brown) with good black glaze inside and outside Buff clay Orange clay with sepia grits. Somewhat blackened wick-hole Brown clay with few white grits and dark red slip Buff to pinkish clay, somewhat blackened wick-hole Pink clay Buff clay, white grits, Reddish brown clay with few white grits Buff clay, patina Buff brown clay Buff clay, tiny white grits, blackened wick-hole Reddish brown clay, patina, blackened wick-hole Brown-gray clay with white grits, blackened wick-hole

Measures L. (P) 4.7 H. (P) 2.9 W. (P) 3.3 L. (P) 2.2 H. (P) 1 W. (P) 1.2 L. (P) 5 H. (P) 3.1 W. (P) 4.3 L. (P) 4.3 H. (P) 2.8 W. (P) 4.5 L. (P) 4.2, H. (P) 2.1 W. 3.9 L. (P) 4 H. (P) 1 W. (P) 0.6 L. (P) 4.1 H. (P) 3 W. (P)3.8 L. (P) 3.3 H. (P) 2.9 W. (P) 0.9 W. lip 0.6 L. (P) 5.2 H. (P) 3.8 W. (P) 4.9 L. (P) 4.3 H. (P) 1.2 W. (P) 1.4 L. (P) 8.1 H. 3 W. 5 L. 8.3 H. 3.6 W. 4.8 L. 8.3 H.4.9 W.3.5

Description

Basket

Locus

Area and Square

Serial Number

Nozzle

1212.01

170

A, I53

5

Small fragment of shoulder

1268.23

186

A, I53

6

Nozzle and beginning of shoulder

8039.01

808

E, F49

7

Shoulder, restored.

8084.01

821

E, H48

8

Nozzle and beginning of shoulder

1198.01

169

A, I53

9

Part of the shoulder

5001

501

C,C5

10

Part of body, including base

8101.03

823

E, H48

11

Shoulder

8002.02

802

E, G48

12

Shoulder with filling hole and nozzle

8013.01

805

E, F49

13

Part of body

8088.01

819

E, E49

14

Nearly complete, tip of nozzle missing

1258.01

186

A, I53

15

Intact

8085.02

818

E, G48

16

Intact

1073. 01

129

A, J52

17

135

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

Reddish clay, white grits Reddish clay, with a few massive grits

L. (P) 2.9 H. (P) 2.1 W. (P) 3.9 L. (P) 3.9 H. (P) 2.4 W. (P) 3.3

Nozzle except the tip

8028.01

811

E, F49

18

Nozzle, except the tip

8101.01

823

E, H48

19

Table 2.3: Closed Mould-made Lamp Color

Measures

Description

Basket

Locus

Area and Square

Serial Number

Reddish clay

L. (P) 5.2 H. (P) 2.4 W. (P) 4.8

Part of body, include base, restored

8012.01

802

E, G48

20

Table 2.4: Unidentifiable Lamp Fragments Color Brownish-reddish clay Reddish clay, white grits, a little blackened wick-hole Blackened Reddish clay H. – Height W. – Width L. – Length (P) - Preserved

Description

Basket

Locus

Area and Square

Serial Number

Small fragment Lower part of Nozzle fragment

1214.01 8030.03

170 804

A, I53 E, E49

21 22

Nozzle fragment

1162

159

A, G53

23

Small fragment Small fragment, figurine?

8046.01 8051

807 814

E, F49 E, H48

24 25

represent pottery discarded during the entire Hellenistic period, the large corpus of intact and complete vessels are representative of a single episode. All the complete vessels unearthed at Kh. er-Rasm clearly “lived” together at the same time,18 and became de facto refuse when this “phase” came to an end (for the formation processes of Kh. er-Rasm, see mainly in section 4, Chapters 4.2, 4.3).

and Vince 1993:190), Therefore, while encompassing two centuries and even more, the possible dates of all the various vessels do not necessarily coincide with one another. Examining the combined possible date of all the various vessels – the assemblage – would give us a narrower possible date in which they could have all coexisted. The period of time in which all the vessels “lived together” is much more limited than the life of each of the vessels, and the more vessels we have (i.e., the larger the assemblage), the narrower the possible date of the assemblage would be (e.g., Orton, Tyres and Vince 1993:25).

Dating Pots and Assemblages: Each pottery form, when examined in itself, may have a long life. Different pots have different rates of appearance, dominance and disappearance, but most cover a long period, with gradual appearance and disappearance, known as the “battleship curve” (see, e.g., Deetz 1967:26–33; Sinopoli 1991:74–75; Orton, Tyres

The main part of this chapter will therefore discuss the possible date of the entire assemblage of complete vessels in order to date the abandonment/destruction of the Hellenistic site. The date of each type/form is based on the discussion in Chapter 2.2, and the Type date range for pottery forms, as established there as well as for oil lamps in Chapter 2.3. We will also compare the finds in the various wings of the main building in order to see if there could be a chronological difference between them. In the second part

Note that three complete vessels (all CP9) are excluded from the discussion. Those pots are later in date (Early Roman period, up to early second century CE), and were all discovered in other contexts: one was discovered in another, higher, level at the main structure (in a later occupation in one room in area A), and two were unearthed on the topsoil, outside the stepped structure in Area C (see Chapter 1.2). These are clearly irrelevant for the present discussion, and belong to a later occupational phase at the site (Chapter 4.3). 18

136

Avraham Faust and Adi Erlich

2.10: Plate of Assemblages – Storage Jars, Pouring and Serving Vessels No.

Type

Locus

Basket

1 2 3 4 5 6 7 8

SJ1A SJ2 SJ3 JG4 DT1 DT2 TA2 KR1A

L149 L824 L129 L170 L160 L165 L821 L821

1149.1 8100.2 1064.1 1214.1 1196.1 1176.2 8092 8092.5

137

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

2.11: Plate of Assemblages – Kitchen Ware No.

Type

Locus

Basket

1 2 3 4 5 6 7 8 9

CP3 CP4 CP6 CP7A CP7A CP7B CP7C CP8 CS1

L115 L115 L815 L818 L818 L818 L813 L818 L810

1044 1036.2 8080.3 8077 8085.12 8077.2 8071.9 8085.15 8036.1

138

Avraham Faust and Adi Erlich

2.12: Plate of Assemblages – Bowls and Plates No.

Type

Locus

Basket

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

BL1A BL1A BL1B BL1B BL2 BL2 BL2 BL1B BL2 BL2 PL1A PL1A PL1B PL1C PL2

L176 L805 L136 L186 L814 L129 L129 L821 L823 L186 L129 L109 L154 L150 L814

1246.5 8013.1 1099.1 1258.5 8076.2 1064.2 1064.3 8092.8 8101.9 1270.23 1064.2 1030.1 1137.1 1139.1 8076.1

139

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

2.13: Plate of Assemblages – Personal Vessels and Lamps No.

Type

Locus

Basket

1 2 3 4 5 6 7 8 9 10 11 12 13 14

JT1 JT2 JT3 UN1 UN1 UN1 UN1 UN1 UN1 UN1 UN2 Lamp Lamp Lamp

121 119 818 130 186 L813 L814 L818 821 L822 L115 L186 L818 L129

1057.1 1040.1 8077.1 1084.2 1264.6 8071.1 8070.1 8085.2 8092.2 8086.9 1036.1 1258.01 8085.02 1073.01

140

Avraham Faust and Adi Erlich

2.14 Complete vessels from the site

of the chapter we will examine all the finds, including the sherds from two rooms,19 locus by locus, in order to see if the picture is the same as the one observed by the study of the complete vessels.

century BCE. This is supported by other types of finds, including coins and 14C (see Barkay, Chapter 3.2 and Boaretto, Chapter 3.9, and see also a detailed discussion in Chapter 4.2).

The Date of the Khirbet er-Rasm Assemblage

Dating the Assemblages of the Two Wings

The corpus discussed here consists of some 53 vessels, including three storage jars, one jug, two decanters, three juglets, ten cooking pots, one casserole, one table amphora, fourteen bowls, five plates, one krater, eight unguentaria and four lamps (Table 2.5).

Since the two main wings of the structure functioned differently (Chapter 4.2), we have examined the finds in each of them separately in order to see whether the different findings cannot be attributed to different history or chronology, rather than function. Finds that do not appear before the second half of the second century or after the second century BCE, were unearthed in both wings: the first group of types which do not appear before the second half of the second century, includes the following: JT3, DT1, CP8 and TA2, which were discovered in the northern wing, and JT2, CP4 (see also CP3) that were discovered in the southern wing. The second group of types which do not continue after the second century BCE, includes numerous vessels in both wings, e.g., the “Neo-Babylonian” lamps, UN1, and BL1 (both A and B).

Intermediate Summary An examination of the possible date(s) of the above corpus shows that the only possible date for the entire assemblage is the late second century BCE. Some types do not appear prior to the mid-second century (e.g., JT2 CP4 CP7A, CP7B, CP7C) or even before the late second century (e.g., JT3, DT1,20 DT2, CP8, TA2). Other forms, however, do not continue beyond the end of the second century (e.g., JT1, JT2, SJ1A, SJ2, CS1, BL1A, BL1B, PL1A, PL1B, PL2, UN1, UN2, and the “Neo-Babylonian” Lamps), thus pointing to the only possible date in which the assemblage functioned together. The pottery, therefore, shows that the structure was abandoned and destroyed in the late second

The finds, therefore, clearly show that in the available chronological resolution, activity at both wings ended contemporaneously, and the entire building went out of use in the late second century (as seems to be indicated by the architectural history of the building, i.e., the blocking of the doorways, etc. see above Chapter 1.1, and Chapter 4.2 below).

The rooms were selected randomly out of a few rooms that could have been used for this exercise. 20 See note 21, below. 19

141

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

Table 2.5: The Dating of the Complete Vessels of the Hellenistic Period Form

Type

Loci

General date; based on Chapters 2.2, 2.3

Storage jars

R-SJ1A R-SJ2

L.149 L.824

R-SJ3

L.129

R-JG4 R-CP2

L.170 L.815*

R-CP3

L.115

R-CP4

L.115

R-CP6

L.815

R-CP7A R-CP7B

L.818, L.818* L.818

R-CP7C

L.813, L.818

R-CP8

L.818

R-CS1

L.810

Third–second centuries BCE Third–second centuries BCE Second century, and even first half of first century BCE (see discussion above) Second–early first century BCE Fourth–second centuries BCE Second half of fourth–first centuries; specifically the end of the second–first centuries BCE Second half of second– early first centuries BCE Third(?)–, second–early first centuries BCE Mid-second–mid-first centuries BCE Mid-second–mid-first centuries BCE Mid-second–early first centuries BCE (sporadically, also third century) Late second–first centuries BCE Second century BCE (most probably the second half)

Jug Cooking Pots

Casserole

L.821

End of second–mid-first centuries BCE

Kraters

R-TA2 (Kantharos) R-DT1 R-DT2 R-KR1

L.160 L.165 L.821*

Bowls

R-BL1A

L.176, L.805

R-BL1B

L.136, L.186*, L.821*

Late second–first centuries BCE21 Late second–first centuries BCE Third–first centuries BCE Second half of fourth–second centuries BCE Second half of fourth–second centuries BCE

Table Amphora Decanters

R-PL1A R-PL1B R-PL1C R-PL2 R-JT1 (Maresha) R-JT2

L129, L.129, L.147*, L.147*, L.186*, L.186*, L.324*, L.814, L.823 L.109; L.129 L.154 L.150 L.814 L.121 L.119

R-JT3

L.818

R-UN1

L.130, L.186, L.813, L.814, L.818, L.821 L.822

R-BL2 Plates

Juglet

Unguentaria

Second–first centuries BCE (at Tel Dor, it appears as early as the fourth) Third–second centuries BCE Late fourth–second centuries BCE Second–first centuries BCE Third–second centuries BCE Third–second centuries BCE Second half of second century BCE Late second BCE century to first century CE Second century BCE

Late third–second centuries BCE (mainly second century) Second century BCE (no examples “NeoL.129; L.186; L.818 are known after the end of the second Lamps Babylonian” century) “mould-made” L.802* Second century BCE * The asterisk (*) sign means that these specific vessels were not restored (and their profile does not appear in the “restored vessels’ plates”; see Figs. 2.10–2.13). Naturally, there is no need to actually restore the vessels in order to treat them as complete (see also Schiffer 1987:285). R-UN2

L.115

The only parallel found to DT1 is dated to the first century BCE (see Chapter 2.2). Since, however, the context in Kh. er-Rasm leads to an earlier date in the late second century BCE we added the “late second century” possibility in the table. It should therefore be stressed that the dating is based on the pottery found in this Stratum, and not on typological considerations. 21

142

Avraham Faust and Adi Erlich

A more detailed discussion of the chronology of the site during the various periods, on the basis of additional lines of evidence, will be conducted in Chapters 4.1 and 4.2.

the overall finds concerning the date of the building. The main phase of occupation, i.e., when the site was destroyed (and hence the date of most finds), is the late second century BCE. Some sherds date to earlier periods and are remains of earlier periods of occupations, while a few sherds from the wall fall are later in date, and result from the processes of collapse which took place in the Early Roman period (section 4, Chapter 4.3).22

Dating All the Finds in Selected Rooms The above discussion of all the complete vessels is sufficient to date the final phase of the Hellenistic settlement. Still, we preferred to examine all the finds, locus by locus (from the bottom up), in selected rooms. While it is clear that some sherds (unlike complete vessels) can “move around” in the various layers due to various mechanisms (e.g., Schiffer 1987:201–212, 300; see more in section 4, Chapters 4.1–4.3), we felt that such an examination should still be conducted.

Conclusions The examination of the possible date for the abandonment of Kh. er-Rasm according to the ceramic assemblage clearly shows that it took place during the late second century BCE. This date is also supported by other types of finds, as will be shown below in section 4, Chapter 4.2, and is also in accordance with the available historical data (below, section 5, Chapter 5.1).

Several rooms could have served this purpose, since the stratigraphy is very clear, and we have selected two of these: room 14 (square H48; Tables 2.6-2.8) and room 15 (square G48; Tables 2.9-2.11).

Interestingly, although it is less accurate, the examination of all the finds in the loci in rooms 14 and 15 support this date. The finds in the ash layer and occupational debris below the wall fall in room 15 (some 22 vessels and rims) clearly show that the only possible date for the destruction was during the second half of the second century BCE (one earlier sherd is a result of earlier activity, and probably came from within the floor; see section 4 for discussion).

Below we have examined the finds from those rooms, from the lowest locus, which was an ashy layer, all the way to the upper part of the wall fall (not including the topsoil of course). Summary: The finds from this room are in accordance with

Room 14 Table 2.6: The Pottery from Room 14, L.823 (Ash layer; “clean,” below the wall fall) Form

Type

Quantity

General date; based on Chapters 2.2, 2.3

Storage jars

R-SJ1C

2

Jugs

R-JG2

1

Juglets

R-JT1

Cooking pots

R-CP6

1 1

Third–second centuries BCE Third (?)–second centuries BCE. It might continue into the first century BCE Third–second centuries BCE

Bowls

R-BL2

1 (complete)

Lamps

Neo-Babylonian lamp

2

Third (?) to second–first centuries BCE Second–first centuries (at Tel Dor, it appears as early as the fourth) Second century BCE

Possible date for the locus: second century BCE. Table 2.7: The Pottery from Room 14, L.821 (below the wall fall; occupational debris/ash layer) Form

Type

Quantity

General date; based on Chapters 2.2, 2.3

Storage jars

R-SJ1B

1

Jugs

R-JG2

1

Unguentarium Cooking pots Table amphora Mortaria Kraters Bowls

R-UN1 R-CP7A R-TA2 R-MT2 R-KR1 R-BL1B

1 (complete) 1 1 (complete) 1 1 (complete) 1 (complete)

Lamps

Local copy of an Attic lamp

1

Third–second centuries BCE Third (?)–second centuries BCE. It might continue into the first century BCE Second century BCE Mid-second–mid-first centuries BCE Late second–mid-first centuries BCE Third–second centuries BCE Third–first centuries BCE Second half of fourth to second centuries BCE Second half of the fourth to the mid (or early) second centuries BCE

Possible date for the locus: late second century BCE (with the exception of the lamp, which is slightly earlier).

143

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

Table 2.8: The Pottery from Room 14, L.813 and L.814 (Kirton Wall Fall) Form

Type

Quantity

General date; based on Chapter 2.2

Storage jars

R-SJ1B R-SJ1D R-SJ2 R-CP5

1 1 1 1

R-CP7C

1 (complete)

R-CS3 R-JT1 R-UN1 R-KRX1 R-PL2 R-PL3 R-BL1B

1 1 2 (complete) 1 Complete 1 1 2 complete and 2 sherds 1

Third–second centuries BCE Third–second centuries BCE Third–second centuries BCE Late second century BCE Mid-second to early first centuries BCE (sporadically, also third century) Early fifth–third centuries BCE Third–second centuries BCE Second century BCE Iron Age I Third–second centuries BCE Second century BCE Second half of fourth to second centuries BCE Second–first centuries (at Tel Dor, it appears as early as the fourth) Second–mid-first centuries BCE

Cooking pots

Juglets Unguentarium Kraters Plates Bowls

R-BL2 R-BL3

Possible date for the locus: second half of second century BCE (two sherds are earlier). Summary: The finds from this room are in accordance with the overall finds from the building, and date the building mainly to the late second century, with some orphan sherds,

indicating that the structure was built earlier (see also Chapters 4.1–4.3).

Room 15 Table 2.9: The Pottery from Room 15, L.824 (ash layer/floor) Form

Type

Quantity

General date; based on Chapter 2.2

Storage jars Jug Bowls

R-SJ2 R-JG5 R-BL1B

1 (complete) 1 2

R-BL2

2

Third–second centuries BCE Hellenistic – no parallel found Second half of fourth–second centuries BCE Second–first centuries BCE (at Tel Dor, as early as the fourth)

Possible date for the locus: second century BCE (though an earlier date cannot be precluded, based on the evidence from Tel Dor).

Table 2.10: The Pottery from Room 15, L.818 (below wall fall/occupational debris mixed in ash layer) Form

Type

Quantity

General date; based on Chapters 2.2, 2.3

Storage jars

R-SJ1A R-SJ1B R-SJ1C R-SJX1 R-DR2 R-JT3 R-UN1 R-CP7A R-CP7B

1 1 1 1 1 1 (complete) 1 (complete) 1 (complete) 1

R-CP7C

1

R-CP8 R-PL1A R-MT2 R-BL1B

1 (complete) 1 1 3

Third–second centuries BCE Third–second centuries BCE Third–second centuries BCE Late Iron Age–Persian period Late second–first centuries BCE Late second century BCE–first century CE Second century BCE Mid-second to mid-first centuries BCE Mid-second to mid-first centuries BCE Mid-second to early first centuries BCE (sporadically, also third century) Late second to first centuries BCE Third–second centuries BCE Third–second centuries BCE Second half of fourth century to second century BCE

Decanter Juglet Ungentarium Cooking pots

Plates Mortaria Bowls

144

Avraham Faust and Adi Erlich

Lamps

R-BL2

1

Neo-Babylonian

1

Second–first centuries BCE (at Tel Dor, it appears as early as the fourth century) Second century BCE (no examples are known after the end of this century)

Possible date for the locus: late second century, with some earlier finds, probably from within the floor.23 Table 2.11: The Pottery from Room 15, L. 810 (Kirton Wall Fall) Form

Type

Quantity

General date; based on Chapter 2.2

Storage jars

R-SJ1D R-SJ6

1 1

Cooking pots

R-CP9

1

Bowls

R-CS1 R-BL1A

1 (complete) 1

R-BL2

1

Third–second centuries BCE Herodian (31 BCE–6 CE) Early Roman period (late first century BCE–early second century CE Second century BCE (most probably its second half) Second half of fourth–second centuries BCE Second–first centuries BCE (at Tel Dor, it appears as early as the fourth)

Possible date for this locus: This locus was defined as a wall fall. Forms unearthed are mainly Hellenistic (until the second century BCE), but it is interesting that within the fallen stones a few later sherds were unearthed. This might be important for understanding the formation processes at the site (see also Chapters 4.1–4.3 below).

2.5 Petrographic Results

The results of the analysis of the finds in room 14 are similar. In this room eighteen vessels and rims were unearthed below the wall fall. All of them, with one exception, existed in the late second century BCE, hence supporting the date. The one exception is an imitation of an Attic lamp, which is usually dated to the mid-second century BCE, and in the Shephelah perhaps even until the early second century. This sherd is probably a remnant of the earlier occupation of the site (see Chapter 4.2).

Anat Cohen-Weinberger Background This study presents the petrographic results of 19 Hellenistic vessels and one Early Roman pot from Kh. er-Rasm. Petrographic studies of Hellenistic pottery from sites in Israel are rare and include studies of pottery from Maresha, a prominent Hellenistic site in the Shephelah. This pottery includes daily vessels,24 figurines (Erlich and Kloner 2008) and oil lamps,25 which are an important comparison with the K. er-Rasm assemblage.

It should be noted that in both rooms not a single late sherd, later than Hellenistic, was unearthed below the wall fall (in or above the floor). This also supports the basic stratigraphical scheme of the excavation. The analysis of the complete vessels, in any event, enables us to narrow down the date of the abandonment and destruction of the site and to pinpoint it to the late second century BCE. The overall processes of abandonment and destruction will be discussed in details below (section 4).

The petrographic results of Maresha’s pottery show preference of terra rossa, loess and rendzina soils as well as marl and clay of the Moza Formation as the raw material. In addition, several pottery kilns with dates ranging from the Middle Bronze Age to the Roman period, such as Horvat Zafit,26 Yesodot,27 Tel Malot,28 Tel Miqne-Ekron (Killebrew 1996, 1998a, 1998b), and Kfar Menahem (Ben-Shlomo 2005), have been excavated in the Shephelah and the inner Coastal Plain. Pottery found in some of theses kilns has been petrographically examined (Killebrew 1998b; BenShlomo 2005).29 The characterized raw material from the kilns serves as reference data for a local production in the western Shephelah and helps to differentiate between the various production sites in the same geographic region. All the above sites are located 8–20 km west or northwest of Kh. er-Rasm in the outlet of Nahal Soreq and Nahal Ha’Elah to the Coastal Plain. There are no excavated kilns, nor comprehensive petrographic studies of pottery from sites in the inner Shephelah, which can be used as references for

22 A nearly complete mould-made lamp was unearthed on top of the collapse. Since it was part of a locus which was still defined as topsoil (L.802), it is not discussed here. It should be stressed, however, that it is in accordance with the general dating proposed here (the lamp is typical of the Shephela, see Chapter 2.3 above). 23 A Ptolemaic coin (#2 in Barkay’s list, see Chapter 3.2) was also discovered in this locus. But since a coin, like tiny sherds, could very easily “move around” the layers through various processes, it is difficult to establish, on the basis of the available evidence (from pottery), that this coin was still in use during the second century, nor would it be possible (on the basis of this single coin) to push the date of some pottery forms (e.g., DR2, CP7A [unearthed complete], CP7B, CP7C [one unearthed complete], CP8 [complete], etc.) back to the third century (in order to conform with such a date). For discussion of the coins, see also Barkay (Chapter 3.2, below) and Chapter 4.1 (below).

145

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

a local production. Geological maps and previous studies of this region help to define the potential raw materials for pottery manufactured in the area (Buchbinder 1969; Dan et al. 1977; Sneh, Bartov and Rosensaft 1998).

pottery production in antiquity (Glass et al. 1993:272, 276; Goren 1995:301; Goren, Kamaiski and Kletter 1996).

Geologic Setting

The analyzed pottery clustered into petrographic groups according to the characteristics of the raw material (see Table 2.12):

Results

An assessment of the geologic setting of the site of Kh. er-Rasm and its surrounding area was carried out with the aim of identifying the potential provenance of the raw material. The following is based on the 1:200,000 geological map of Sneh, Bartov and Rosensaft 1998. Kh. er-Rasm is located on chalk and chert rocks of the Eocene Adulam Formation (Israel coordinates 14355/12185). Chalk rocks of the Maresha Formation are also exposed very close to the south and west of the site. Close to the site of Kh. er-Rasm, on the flanks of Nahal Ha’Elah and in the surroundings of Bet-Shemesh marl, clay and chalk of the Paleocene Taqiye Formation are exposed. Calcareous caliche crust, known by the local name nari, develops on the Shephelah chalky rocks (Buchbinder 1969:1). Nahal Ha’Elah drains the Judean Hills and the Shephelah and is characterized by accumulated alluvial beds (Buchbinder 1969). The area of the site is also characterized by brown rendzina and grumusolic soils (Ravikovitch 1969; Dan et al. 1977; soil map of Beth Shemesh, 1:50,000). East of the site, the Senonian Mishash and Menuha formations, which are characterized by chert and chalk, are exposed. Further east, Turonian and Cenomanian formations, characterized by hard limestone and dolomite rocks with some marl and clay units, are exposed. Terra rossa soils developed out of the limestone and dolomite rocks and are widely exposed in about 9 km east of Kh. er-Rasm. West of Kh. er-Rasm (in about 4 km) conglomerates of the Pliocene Pleshet Formation (Buchbinder 1969:9) and (in about 8 km) Quaternary hamra soils and Kurkar ridges are exposed. Grumusols soils are widely exposed in the area west of the site in the eastern Coastal Plain. These soils developed from aeolian or alluvial clay or silty clay (Dan et al. 1977:12, 136).

Group A1: This group is characterized by fine clay, somewhat calcareous, containing iron oxides. The nonplastic components are homogenous, densely spread (f:c ratio {0.062mm}= ~75:25) and dominated by well sorted idiomorphic rhomboid dolomite crystals, which were exposed to some decarbonization during firing. This fabric belongs to a well-defined petrographic group that has been published in detail (e.g., Goren, Finkelstein and Na’aman 2004:263 and additional references there). Based on the extensive body of reference material, this petrographic group is identified as originating from clay of the upper member of the Moza Formation, mixed with dolomite sand quarried from the overlaid ‘Amminadav Formation. The Moza and ‘Amminadav formations are exposed in the Cenomanian section of the Judean Mountains (Bentor 1945). Only one vessel (Table 2.12:7) is related to this group. Group A2: This group is characterized by the same clay type as group A1 with sparsely spread silty idiomorphic rhomboid dolomite crystals. Three vessels are related to this group (Table 2.12:2, 12, 20). These samples are devoid of any non-plastic components except for one vessel (Table 2.12:12) which is characterized by ~5% sand-sized (200–500µm) surrounded by quartz grains sometimes with andulose extinction. Group A3: This group is characterized by yellowish-tan carbonatic matrix rich in tiny (below 50µm), rhombohedral dolomite crystals. This clay is identified as marl of the Moza Formation. Three vessels are related to this group (Table 2.12:4, 6, 17).

Sediments rich in clay are essential for pottery manufacture. The following potential raw materials are found in the vicinity of the site: (1) the marl of the Paleocene Taqiye Formation; (2) soils such as rendzina, alluvial and grumusolic soils; (3) From Elah Valley eastward, terra rossa soils and clay and marl members of the Moza Formation (at least 10 km from the site), which were widely used for

Groups A1 to A3 are identified as quarried out of the marl and clay members of the Moza Formation. The use of this clay for pottery production is well known from sites of different periods spread throughout the ridge of the central hills (see e.g., Goren, Finkelstein and Na’aman 2004:262–264, with additional references there).

The petrographic study of the daily used vessels was conducted by Marcia L. Okun and was not published. 25 Study of the raw material used for Hellenistic oil lamps from Maresha was conducted by the author as a contribution to Einat Ambar-Armon’s Ph.D. dissertation (2007). 26 Unpublished kiln excavated by Daniel Verga, IAA excavation license A2954/1998. 27 Unpublished, most likely MB kiln excavated by Angelina Degot, IAA excavation license A3260. 28 Unpublished MB kiln excavated by Giora Parnos, IAA excavation license A3641/2002. 29 Petrographic Study (unpublished) of the pottery from Horvat Zafit, Yesodot and Tel Malot was conducted by the author.

Group B: This group is characterized by ferruginous opaque clay rich in silt-sized quartz grains. The silt content is about 30% and consists mainly of quartz grains with accessory zircon, feldspar and oxihornblende. Opaques are common in the silty fraction of the paste. The non-plastic components (f:c ratio {0.062mm}= ~90:10) are dominant in three samples (Table 2.12:1, 3, 16) by quartz grains of 200–500µm with some chert (Table 2.12:16) or poorly sorted silicified rock fragments of up to 1.2mm (Table 2.12:1). In the other samples related to this group (Table 2.12:5, 9, 10 and 14) the non-plastic components (f:c ratio {0.062mm}= ~95/90:5/10)

24

146

Avraham Faust and Adi Erlich

Table 2.12: Petrographic Results #

Basket

Locus

Type

1 2 3 4 5 6

8085 8092 8100 8076/2 8013/1 1214

818 821 824 814 805 170

CP6 TA2 SJ2 BL2 BL1A JG4

7

1083

134

FK1

A1

8

11 12 13 14 15 16 17 18

1037 1064 8013 1217/2 1196 8080 1137 1036 8085 1030 8086

121 129 805 172 160 815 154 115 818 109 822

JT1 SJ3 Neo-Babylonian CP9 DT1 CP6 PL1B CP4 CP8 PL1A UN1

C B B B

19

1149

149

SJ1A

D

20

8084/01

821

“Attic lamp imitation”

A2

9 10

Petrographic Group

Soil/Formation

B

Terra rossa soil Clay of the Moza Formation Terra rossa soil Marl of the Moza Formation Terra rossa soil Marl of the Moza Formation Clay of the Moza Formation and dolomitic sand of the ‘Amminadav Formation Loess soil and nari Terra rossa Terra rossa soil Terra rossa soil Clay of the Moza Formation Loess soil and quartz Terra rossa soil Hamra soil Terra rossa soil Marl of the Moza Formation Loess soil and nari Mixed unidentified carbonatic clay with marl of the Paleocene Taqiye Formation

A2 B

A3 B

A3

A2 C B E B

A3 C

Clay of the Moza Formation

pot and the bottle (Table 2.12:13, 18) paste include well sorted rounded to sub-rounded quartz grains of ~300µm and rarely chert fragments. Based on a bulk of comparative material and published data (e.g., Porat 1987:112–115; Goren 1996b), the matrix is identified as loess soil. By using the term “loess” I refer to a set of aeolian and alluvial carbonatic silty-clay sediments. Loess soil occurs mainly in the northern Negev and southern Shephelah regions (Ravikovitch 1969). However, petrographic studies of pottery from kilns in the inner Coastal Plain, such as Yesodot, Horvat Zafit and Kfar Menahem show use of clay that cannot be distinguished petrographically from loess used for vessels made in the northern Negev and southern Shephelah. The carbonatic inclusions accompanying the matrix and the relatively fine quartz sand point to the Shephelah or the inner Coastal Plain as the possible provenance of this group. The loess-made vessels in the coastal Negev sites are characterized by coarse quartz grains and other coastal sand-sized grains of accessory minerals, such as hornblende and feldspar and kurkar fragments (Goren 1996a:54).

include rounded micritic limestone fragments. One sample is devoid of any non-plastic components (Table 2.12:11) and in another sample decomposed carbonatic sands appear (Table 2.12:5). This petrographic group is identified as terra rossa soil, which is commonly found in Judea, Samaria, and the Galilee Mountains. The parent material is a hard limestone, dolomitic limestone, or dolomite. A large amount of aeolian particles from external source was incorporated into the soil (Wieder and Adan-Bayewitz 2002:395). Terra rossa soil was extensively used as a raw material for pottery in antiquity (see e.g., Goren, Kamaiski and Kletter 1996). This soil type is not local to the area of Kh. er-Rasm. Terra rossa soils appear east of the Elah Valley (Dan et al. 1977; soil map of Beth Shemesh, 1:50,000). Although the terra rossa soil is not local to the Shephelah, vessels made of terra rossa are common in the ceramic assemblages of sites in the Shephelah region. For example, in the Iron Age pottery assemblages of Beth Shemesh and Batash, the terra rossa group is dominant (Goren 1996a; Mazar and PanitzCohen 2001:18–19). Group C: Three examined vessels (Table 2.12:8, 13, 18) are related to this group, which is characterized by silty carbonatic b-fabric matrix. The silt is well sorted and contains mainly quartz grains but also a recognizable quantity of other minerals including hornblende, mica, and feldspars. The vessels are characterized by different non-plastic assemblage. The non-plastic components (f:c ratio  {0.062mm}= ~95:/5) in the juglet (Table 2.12:8) paste include mainly rounded, poorly sorted carbonatic nari (caliche) fragments of 200µm–1.5mm. The non plastic components (f:c ratio  {0.062mm}= ~90:/10) of the cooking

Group D: One examined vessel is related to this petrographic group (Table 2.12:19), which is characterized by silty carbonatic matrix. The silty components comprise of mainly quartz grains with some appearance of silty heavy minerals and feldspar grains. This clay is unidentified and was mixed with another clay type, which appears in the former clay in a pellet structure of b-fabric clay minerals and contains tertiary foraminifera. The second clay is identified as the Paleocene Taqiye Formation. The nonplastic components contain idiomorphic rhomboid dolomite

147

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

voids which are evidence of straw that vanished in the process of firing also appear in the matrix. The nature of this group clearly points to a coastal origin. This raw material is identified as hamra soil. Hamra soil is found along the Coastal Plain of Israel from the Ashdod area northward.

crystals which were quarried from the ‘Amminadav Formation (see Group A1) and carbonatic nari (caliche) fragments. This vessel was made by a mixture of two clay types. The dolomite was quarried in the central hill of Judea and the provenance of the Taqiye Formation clay as well as the nari fragments is in the Shepelah region.

Coastal sand of the classification described here does not extend beyond the Acre area on the northern coast of Israel (Issar 1968; Sivan 1996:107–110). Therefore, this group should be attributed to the Coastal Plain of Israel, between Ashdod and Acre. In terms of reference, hamra-made, quartz-rich pottery is distributed in Israel in sites located mainly along the central Coastal Plain (see e.g., CohenWeinberger and Goren 2004).

Group E: One examined vessel is related to this petrographic group (Table 2.12:15), which is characterized by ferruginous clay with some silty quartz grains and rarely silty feldspar and mica grains. The non-plastic components (f:c ratio {0.062mm}= ~80: 20) are comprised of mainly rounded to sub-angular quartz grains of 200–300µm with some coarser grains of up to 1mm, and rarely feldspar and heavy minerals such as hypersthene and mica laths. Elongated

148

Part III: Other Reports

Avraham Faust and Adi Erlich

3.1 The Flint Collection

Table 3.1. Waste Material Frequencies

Ofer Marder During the excavation a small sample of flint artifacts (n=29) was collected from various loci at the site of Kh. er-Rasm. The artifacts are made of low quality breccious flint (n=13) or high quality fine-grained chalcedony flint (n=11). The remainder is burnt flint, of a type that is not identifiable (n=5). Interestingly, all the tools and the core that were recovered are made of chalcedony flint. Most of the artifacts are waste material including flakes, primary flakes, blades, chips and chunks (Table 3.1). In addition, one amorphous core (Fig. 3.1:1) and three tools were found (Figs. 3.1:2–4). The tools consist of two retouched blades (Figs. 3.1:2–3) and one truncated blade

Type

n

Primary Elements Flakes Blades/lets Total Debitage Chunks Chips Total Debris Debitage Debris Cores Tools

4 12 2 18 2 5 7 18 7 1 3

Total

29

3.1 Flint artifacts Figure

Type

Square

Locus

Basket

3.1:1 3.1:2 3.1:3 3.1:4

Amorphous Core Retouched Blade Retouched Blade Truncation

H49 I53 I53 I53

161 186 186 186

1159 1261 1264 1261

151

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

(Fig. 3.1:4). Interestingly, all the tools were found in locus L.186. This is the ‘lowest’ locus in Square I53, and is located below the Hellenistic period wall fall (L.170, L.184). L.186 was defined as the transition between occupational debris and the accumulation below it, and since the tools were found within the lowest two baskets, it is clear that they came from within and below the Hellenistic floors.

southwestern slopes of the hill, Chalcolithic pottery (e.g., three cornets and a jar) and undiagnosed flint artifacts were collected, therefore it is possible that a few flint artifacts, which were found during the excavation, are part of Chalcolithic occurrences on the site.

It is difficult to ascribe this flint collection to a particular industry, since the sample is small and not a single formal tool (fossil guide) was found. However, during a survey conducted by Yehuda Dagan (IAA site 14-12/31/1) on the

3.2 The Coins Rachel Barkay

3.2 Coins

152

153

5

“

4

Ptolemy III Eurgetes 246-221 BCE

“

Ptolemy II Philadelphos 285-246 BCE

Ruler

3

2

No.

c. 265 BCE and after

c. 285-266 BCE

Laureate head of Zeus r.

Head of Alexander the Great r., with horn of Zeus Ammon, clad in elephant’s scalp and aegis.

“

Laureate head of Zeus r.

Obverse

1

Date

Head of Alexander the Great r., clad in lion’s scalp.

c. 330c. 320 BCE

Alexander Philip III

PTOLEMIES

Obverse

Date

Ruler

No.

THE HELLENISTIC PERIOD

“

“

Eagle l. on thunderbolt closed wings; on shoulder, cornucopiae. [PTOLEM]AIOU [BASILEWS]

“

Alexandria

Mint

Mint

Eagle l. on thunderbolt with wings open. [PTOLEM]AIOU BASILEWS

Eagle l. on thunderbolt with wings open. In front, monogram. PTOLEMAIOU BASILEWS Eagle l. on thunderbolt, wing open (worn out). Countermark, trident. [PTOLEMAIOU BASILEWS]

Reverse

Above: Quiver and bow to l. Below: club of Heracles to r. ALEΞ[ANΔEP]

Reverse

4.38

6.72

14.78

15.01

Weight (gr.)

8.36

Weight (gr.)

15-17

21

26-27

27

Diam. (mm.)

20-21

Diam. (mm.)

Æ

Æ

Æ

Æ

Metal

Æ

Metal

6

10

12

12

12

Axis

Axis

Area E; Sq. E49; L. 826; Bas. 8102.01

G. 27/2003 L. 999; Surface

Sq. I-50 Surface

Area E; Sq. G48; L. 818; Bas. 8093.01

SNG 1977: no. 191.

SNG 1977: nos. 159-162; Svoronos 1904: no. 451, pl. XVII:17

Ibid., no. 129

SNG 1977: no. 91.

references

e.g. Price 1991, 431 no. 3427.

Area B; Sq. D51; L. 320; Bas. 3055

Dig Ref.

references

Dig Ref.

This type lacks the flan’s central cavity, which started to appear c. 265-261 BCE

Notes

Worn out mint mark

Notes

Avraham Faust and Adi Erlich

”

”

”

”

7

8

9

154

175-164 BCE

Ruler

John Hyrcanus (?) [135] 127104 BCE

No.

12

HASMONEANS

11

Antiochus IV Epiphanes

Date

”

”

”

6

”

From 198 BCE

Antiochus III the Great 222-187 BCE

10

Date

Ruler

No.

SELEUCIDS

” (off center)

Illegible Paleo-Hebrew inscription surrounded by a wreath.

Obverse

Diademed radiate head of Antiochus r.; to l. monogram.

Laureate head of Apollo r., hair in krobylos with wavy lock falling behind ear, another on back of neck.

” (worn out)

Mint

Jerusalem

Double cornucopias with pomegranate set between the horns.

Akko Ptolemais

Veiled female figure standing facing, holding long scepter or torch. [BASILEWS] ANTIOCOU

Reverse

Uncertain, in Southern Coele Syria

Nike advancing, holding long palm in extended r. hand. [B]ASILEWS [A] NTIOCO[U]

”

”

”

” [BAS]ILE[WS ANT]IOCOU

Laureate head of Antiochus III as Apollo r.

” [BASILEWS ANTIOCOU] ” [BASILEWS ANTIOCOU]

Akko Ptolemais

Nude Apollo standing l., holding arrow and resting l. hand on bow, to l. [BASILEWS ANTIOCOU]

” (worn out)

Mint

Reverse

Obverse

2.24

Weight (gr.)

2.48

8.88

1.54

1.82

2.54

2.84

Weight (gr.)

14-15

Diam. (mm.)

1717.5

18

10-11

10

10-12

10-11

Diam. (mm.)

Æ

Metal

Æ

Æ

Æ

Æ

Æ

Æ

Metal

12

Axis

12

12

-

-

-

12

Axis

references e.g. Meshorer 2001: 207, Group D, pl. 11.

Area A; Sq. I47; L. 182; Bas. 1248.01

Houghton and Spaer 1998: nos. 1130-1137.

Ibid.: p. 418, no. 1100.

Ibid.

Ibid.

Notes

serrated

”

”

”

Crude style. No control mark

Houghton and Lorber 2002: pp. 416-417, no. 1096. pl. 90. Ibid.

Notes

references

Dig Ref.

Area A; L. 124; Bas. 1059.01

G 27/2003 L. 999 Surface

Area B; Sq. D51; L. 320; Bas. 3084..01 Area B; L. 998B; Bas. 3116..01 Area E; L. 998E; Bas. 8091..02

Area E; Sq. G49; L. 804; Bas. 8010..01

Dig Ref.

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

42 CE

Agrippa I 37-44 CE

“

13

14

155

Obverse

Illegible Arabic legend

Date

12th-15th cent. CE

Ruler

Ayyubids or Mamluks

No.

17

Segments of Arabic legend «There is no God but Allah»

Reverse -

Mint

-

Cross Illegible legend

«

«

«

-

Worn out

Diademed and draped bust of emperor r. Illegible legend

4th-5th cent. CE

unidentified

Mint

Reverse

ISLAMIC

16

“

“ [AGRIPA] BACILEWC

Obverse

15

Jerusalem

Three ears of corn issuing from between two leaves; on l. and r. date: L [V ] (year 6=42 CE)

Canopy [AGRI]P[A] BACILEWC

“

Mint

Reverse

Obverse

Date

Ruler

No.

LATE ROMAN PERIOD

“

Date

Ruler

No.

HERODIANS

1.02

Weight (gr.)

0.48

1.78

Weight (gr.)

2.40

2.06

Weight (gr.)

15-18

Diam. (mm.)

10

12-14

Diam. (mm.)

16

16

Diam. (mm.)

Æ

Æ

Æ

Metal

Æ

Æ

Metal

Metal

-

Axis

12

-

Axis

12

11

Axis

Surface Sq. H-52

Dig Ref.

G 27/2003 L 999 surface Area E; Sq. G50; L. 809; Bas. 8043.01

references

references

“

Area E; Sq. H48 L. 803; Bas. 8009.01

Dig Ref.

Meshorer 2001: 231 no. 120, pl. 52.

references

Sq. I-49 Surface

Dig Ref.

Very worn out

Notes

«

Very worn out

Notes

Notes

Avraham Faust and Adi Erlich

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

3.3 Stone Vessels

Millstones (Nos. 3–5) This type of millstone represents the first, or earliest, stage of the mechanization of grinding technology. This tool, which was activated by a pole that was used as a crane, freed humans from the need to manually activate the upper grinding stone. This advance enabled the enlargement of the dimensions of the grinding stones and consequently increased the quantity of flour produced. The pair of basalt stones includes a square upper millstone with a concave depression on the top. On the bottom there is a narrow crack formed as a sort of funnel through which the grain was fed during the grin ding process. On the frame of the upper millstone there are parallel depressions in which the pole was positioned. In the grinding areas of the pair of millstones are grooves that are intended to facilitate the even distribution of the grain poured from the upper funnel. This technique of grinding using the millstone is depicted in a scene found on a Homerian bowl (Rostovtzeff 1959: pl. 25).

Rinat Peshin The present chapter describes the stone vessels and artifacts unearthed at Kh. er-Rasm (Fig. 3.3). We shall present the details of all the objects in Table 3.2 (below), but would like to begin with a discussion of some of the more significant finds.

Alabaster Vessel – Small Ledge-Rim Bowl (No. 1) Excavations in Israel have produced numerous alabaster vessels dating to the Persian and Hellenistic periods. Egypt was the production center for such vessels, and during the Persian period and well into the Hellenistic and Roman periods, the import of alabaster objects into the region was resumed (Ben-Dor 1945:93). Made of a semi-transparent white-yellow calcite, these vessels sometimes include darker-toned veins. The small bowl found at Kh. er-Rasm was used for the grinding and mixing of cosmetic powders and creams.

Several suggestions have been put forth concerning the dating of this type of millstone. Amiran places the earliest use of this type to Iron Age II based on the finds at Tel ‘Azekah (Tell Zecharia), Tel Goded, Samaria and Tel Halaf (Amiran 1956:47). White and Robinson date the invention of the “grinding machine” to the fifth century BCE (White 1963:205; Robinson 1938:332). However, their suggestion conflicts with evidence from sunken ships predating the fifth century BCE (Kingsley and Raveh 1996:16).

Additional vessels of this type, dated to the Persian period, have been found at Tel Michal (Clamer 1989), Tel Dor (Stern 1994:187), Tel ‘Ira (Beit-Arieh 1999:465, fig. 14.22:12) and from the Hellenistic period at Maresha (Peshin 2001:67–9), Ashdod (Dothan and Freedman 1967:27, fig. 9:3; Dothan 1971: fig. 29), Tel Beersheba (Derfler 1984:126, pl. 24), Shechem (Kerkhof 1969:99, fig. 30:10), and Tell Abu Salima (Sheikh Zuweyid) (Petrie 1937a:10, pl. 26:12, 13, 15; Petrie 1937b:14, pl. 36:936–938).

Millstones such as these have been found in numerous sites throughout the region and their origin seems to have been Hellenistic, e.g., in Delos (Deonna 1938:126–127), Thera (Von Gaertringen and Wilski 1904:181, figs. 193, 195), Olynthus and Priene (Robinson 1938:227–334, pl. 80), Anatolia and Syria (Oppenheim 1939: pl. xxxviii:b), Phoenicia (Pritchard 1988:116, fig. 32:4 Dunand and Duru: pl. xxxii:3) and southern Italy (White 1963:202, pl. 47:5).

Examples of this type of bowl have been found in burial assemblages in Alexandria (Breccia 1912:98, figs. 60, 62), in Persepolis, where an extraordinarily large alabaster bowl was found, in El Mina in Syria (Woolley 1938: fig. 19), and in Byblos (Dunand 1950:66–7, fig. 42:7109).

One of the identifying signs for these millstones is the fishbone (idra) motif, or other motifs that decorate the friction board. Robinson, in his discussion on the findings from Olynthus, claimed that the similarities in form, material and size among the Greek millstones from different sites may indicate a single production center, possibly in Thera, from where they were distributed to the world markets (Robinson 1938, 330). Evidence of marketing of these millstones throughout the Mediterranean can be seen in the ancient anchorage at Yavneh Yam (Galilli, Sharvit and Bahat-Silberstien 1996:118) where an upper millstone of this type was found. In light of the fact that the stone shows no evidence of friction, we can conclude that it was part of a shipment. Throughout the region these imported millstones can be found together with specimens that lack explicit signs of Greek production. The latter were probably local imitations of Greek forms utilizing local material, an example of which can be seen in the finds from Maresha (Peshin 2001:117).

Stone Ledge-Rim Bowl (No. 2) The stone ledge-rim bowl is common in Hellenistic assemblages. Sometimes this bowl is found on a trumpet base (Cahill 1992:193–4). It is commonly produced from local limestone, imitating the Egyptian alabaster bowls that begin to appear in Israel during the Persian period. Bowls of this type were found at Maresha (Peshin 2001:80– 84); in Jerusalem in the City of David (Cahill 1992:191–3, fig. 14:1–18) and in the Upper City (Reich 2003:279, pl. 8.1 4–7), Tell el Ful (Lapp 1981:110), Samaria (Reisner, Fisher and Lyon 1924:336, fig. 209:6i), Lachish (Tufnell 1953: pl. 64:5), Shechem (Kerkhof 1969:102, fig. 32:6), Ashdod (Bahat 1971:202–203, fig. 96:12), and Tel Michal (Singer-Avitz 1989a:352–53, fig. 31 4:9–12).

156

Avraham Faust and Adi Erlich

Table 3.2: Stone Objects Area

Square

Locus

Basket

Material

1

E

G48

802

8002.01

Alabaster

2

E

F49

806

8033.01

Limestone

3

B

E50

999 Surface find

3120.01

Basalt

4

B

D50

333

3108.01

Basalt

5

E

H48

814

8070.01

Basalt

6

B

G27

998 Surface finds in Area B

7

A

I53

186

1268.03

Basalt

8

B

E51

330

3083.02

Basalt

9

E

G48

802

8024.03

Limestone

Grinding tool

10

C

C1

512

5026.01

Limestone

Stopper

11

B

G53

303

3010.01

Limestone

Weight (?)

12

A

I53

177

1221.01

Limestone

Fragment of weight

13

B

F53

306

3019.01

Chalk

Weight (poss. loom weight?)

14

A

I53

159

1156.01

Chalk

Weight (poss. loom weight?)

15 16 17 18 19 20 21 22 23 24 25 26

B B B A B E E A A E E E

F51 E51 E51 I52 D51 F49 E49 F51 I53 E49 E49 H48

332 330 330 113 332 801 804 128 177 819 804 814

3080.01 3091.03 3115.01 1034.01 3097.01 8007.62 8030.03 1067.01 1221.02 8082.01 8030.02 8076.01

Limestone Limestone Limestone Limestone Limestone Limestone Limestone Limestone Limestone Limestone Limestone Limestone

27

C

C8

523

5059.01

Chalk

28

A

I49

155

1157

Chalk

29

D

Cave

704

7002

Chalk

30

E

G48

802

8024.02

31 32 33

A A A

I52 I52 I52

175 175 175

1219.01 1219.02 1245.01

Basalt

Type

Dimensions

Fragment of small ledge-rim bowl Fragment of bowl with ledge-rim bowl Fragment of upper millstone. The lower part is decorated with parallel grooves Fragment of upper part of millstone (?) Fragment of lower part of millstone

Width of rim 2.3 cm; Fig. 3.3:1 2 cm height

2.3 cm width

Fig. 3.3:5

Grinding stone fragment

Area 9 x 10.5 cm; 4.5 cm width

Fig. 3.3:6

Diameter 6 cm

Fig. 3.3:7

Grinding stone fragment with depression Grinding stone fragment

Mosaic tile Mosaic tile Mosaic tile Mosaic tile Mosaic tile Mosaic tile Mosaic tile Stone sphere Stone sphere Stone sphere Stone sphere Stone sphere Unidentified, perforated Flat wall perforated at corner (ossuary?) Flat wall (ossuary?) Stone disc

Chalk Chalk Chalk

Cup Bowl Disc base

157

Width of rim 2.5 cm

Notes

Fig. 3.3:2

Fig. 3.3:3

Fig. 3.3:4

Length 7 cm; width 4.5 cm Length 25.5 cm; width 8 cm Length 6.5 cm; width 2 cm Length 5.5 cm; width 2.3 cm; diameter of perforation 2 cm Diameter 11 cm; width 5.5 cm Diameter 8.5 cm; width 3 cm; diameter of perforation 1.5 cm Diameter 2.6 cm; diameter of perforation 1 cm 2 x 1.5 cm 2 x 1.5 cm 2 x 2 cm 2 x 2cm 1 x 1 cm 1.8 x 1.8 cm 1.6 x 2.2 cm Diameter 11 cm Diameter 11 cm Diameter 11.5 cm Diameter 11 cm Diameter 9.5 cm Length 2.5 cm; width 2 cm Length 6 cm; width 5.5 cm Length 11 cm; width 7.3 cm Diameter 1.8cm; width 2mm Height 11 cm Width of rim 1.5 cm Width 1.2 cm

Not illustrated Not illustrated Fig. 3.3:8 Fig. 3.3:9 Fig. 3.3:10 Fig. 3.3:11 Fig. 3.3:12 Not illustrated Not illustrated Not illustrated Not illustrated Not illustrated Not illustrated Not illustrated Not illustrated Fig. 3.3:13 Fig. 3.3:14 Not illustrated Fig. 3.3:15 Fig. 3.3:16 Fig. 3.3:17 Not illustrated Not illustrated Fig. 3.3: 18 Fig. 3.3: 19 Not illustrated

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

3.3 Stone vessels

158

Avraham Faust and Adi Erlich

3.3 Stone vessels (continued)

159

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

3.3 Stone vessels (continued)

160

Avraham Faust and Adi Erlich

Millstones of this type were discovered in different sites in Israel dating to the Persian, Hellenistic and Roman periods at Tel Mevorakh (Stern 1978: pl. 44:5), Shikmona (Elgavish 1974: pls. xviii:240, xxiv:254), Tel Dor (Stern 1994: fig. 13; Kingsley and Raveh 1996:17, fig. 17), Tel Michal (SingerAvitz 1989a:350–351, fig. 31.2:9–10), Samaria (Reisner, Fisher and Lyon 1924:338, fig. 212), Ashdod (Dothan and Freedman 1967: fig. 4:15), the Upper City of Jerusalem (Kingsley and Raveh 1996:16), Tel Zecharia and Tel Goded (Bliss and Macalister 1902:143, pl. 73), Maresha (Peshin 2001:119–120) and Yavneh Yam (Galilli Sharvit and BahatSilberstien 1996:118).

destruction of the Temple, the laws of purity lost their significance and the production of stone vessels in the Jerusalem quarries disappeared. They ceased to appear some time during the first half of the second century CE, coinciding with the defeat in the Bar Kokhba revolt, which seemed to erase any hope of return to Jerusalem (Magen 2002:162–64). However, Weiss notes the use of stone vessels in Jewish settlements in the Galilee continued up to the third–fourth centuries CE, based on their discovery in the synagogue at Sepphoris (Weiss 2005:310, pl. FIV).

The millstones from Kh. er-Rasm are poorly preserved and the sections that would have included possible signs of foreign production are indecipherable, making it difficult to determine whether they are local or imported. However, their presence at the site does indicate a Hellenistic influence demonstrated in the use of this new production technique similar to the findings at Maresha.1

Stone cups are the most common item among the stone vessel assemblages from the end of the Second Temple period up to the Bar Kokhba revolt. These cups have one or two square-shaped vertical handles with a large perforation in the center. These cups were formed by hand and in most cases vertical lines left by the leveler are visible on the outside of the vessel, while the insides are smoothed. Sometimes a spout is added to the cup.

Stone Cup (No. 31)

Mosaic Tessera (Nos. 15–21)

This type of cup is commonly found in sites such as Jerusalem, in the City of David (Cahill 1992: fig. 20:1–5), adjacent to the Temple Mount (Mazar 1971:18, fig. 12:1–1), and in the Upper City (Magen 2002:23, fig. 2.6:4–6, 41, fig. 2.33: 8, 98, fig. 3.60:1–3; Magen 1988: pl. 17). Geva dates the Jerusalem cups to the second half of the first century BCE and suggests that they are an early phase of stone cup development (2006a:231, pl. 9.1:10–15). They were also discovered around the city, on Mount Scopus (Amit, Seligman and Zilberbod 2000:358) and Sanhedria (Rahmani 1961: fig. 5:26).

The mosaic stones that were found at Kh. er-Rasm are simple and white. These stones were discovered in Areas A, B and E always in ‘topsoil’ or ‘mixed with topsoil loci.’ Their simple form is indicative of a possible industrial structure in the area at some stage.

Ballistra Stones (Nos. 22–26) Five solid round stones that were found at the site may be identified as ballisteria. It should be noted that three of them were found together in Area E.

Similar vessels were found outside of Jerusalem in assemblages dating to this period: in Horvat Hilel in Gush Etzion (Amit 1993:221, fig. 7), Herodion (Bar-Nathan 1981: pl. 11:1–5), Ahuzat Hazan (Kloner and Tepper 1987:348, fig. 163:5), Horvat Tabak (Sagiv and Zissu 1998: pl. 6:1, 2, 4, 5), Ashdod (Dothan and Freedman 1967: fig. 12:11–12), Gibeon (Pritchard 1964: figs. 34:15, 38:14), Wadi Murabat (De Vaux 1953: fig. 4:10), Ein Gedi (Mazar and Dunayevsky 1967: pl. 34:9), Masada (Yadin 1966:152–153), and in the Galilee at Kfar Hanania (AdanBayewitz 1988-89) and Yodfat (Aviam 1999:97).

Stone Vessels from the End of the Second Temple Period Stone vessels of this type are characteristic of the Jewish population from the second half of the first century BCE until the Bar Kokhba revolt. According to the rules of Jewish halakha, stone vessels cannot become impure and hence their distribution increased during this period when Jewish ritual, in particular the laws of purity and impurity, constituted a central focus in the daily life of the Jewish population. These vessels were used daily by the Jews and became a characteristic marker of Jewish populations; they are not found in pagan settlements. The production of stone vessels concentrated around Jerusalem in an area with abundant veins of soft chalk (quarries for stone vessel production have been found in the areas of Hizma, Mount Scopus, and the Mount of Olives) although production centers have been found in the Galilee as well (Gal 1991:25). Evidence of production has been found in the large deposits of production waste and damaged vessels.

Stone Bowls (Nos. 32-33) A flat open bowl with a slanting side, the rim is angled with an engraved decoration. It appears that the design of these bowls was intended to imitate the ceramic bowls of the Terra Sigillata type. Bowls of this type were found in Jerusalem in the City of David (Cahill 1992: fig. 15:5) and adjacent to the Temple Mount (Mazar 1971:18, fig. 12:13; Magen 1988: fig. 7:10– 11); in the Judean Shephelah at Ahuzat Hazan (Kloner and Tepper 1987:348, fig. 163:2) and Horvat Tabak (Sagiv and Zissu 1998: fig. 6:3); and at Ashdod (Dothan and Freedman 1967: fig. 12:11–12).

These stone vessels were manufactured partially by hand and partially by using lathes of different sizes. With the For conservative elements in Kh. er-Rasm’s material culture, see Chapter 2.3. 1

161

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

3.4 Metal artifacts

162

Avraham Faust and Adi Erlich

Finger Ring M 4. Area A, L.120, 1059.02;

The stone cup and the stone bowl from Kh. er-Rasm were found in the same locus, and belong to the same unique category of Jewish vessels from the end of the Second Temple period. The discovery of these two vessels may therefore shed light on the religious observance of the occupants or, at the very least, indicate a Jewish presence at the site at the time (phase 4 of the main building).

topsoil

Copper-alloy finger ring; fragment; diam. (estimated): 2 cm; the ring has a flat oval bezel (length: 2 cm; width: 1.2 cm; thickness: 0.1 cm); on the disc there is an engraved decoration. Parallels: From the first century BCE to the first century CE: Qalandia (Magen 2004: pl. 14:13); from the Early Byzantine period: Sardis (Waldbaum 1983: pl. 48:827) and Corinth (Davidson 1952: pl. 103:1861–1862, 1855–1856).

3.4 Small Finds 3.4.1 The Metal Artifacts

Implements M 5–M 7

Ravit Nenner-Soriano Introduction

M 5. Area D, L.701, 7001.01; topsoil

This report includes seven metal objects from various areas of excavation.

Iron ring; complete; diam.: 3 cm; thickness: 0.5 cm; this ring was probably used as a hinge.

Catalogue (Fig. 3.4)

M 6. Area E, L.818, 8085.03; Hellenistic

Iron Arrowhead M 1. Area E, L.810, 8027.01; Hellenistic

Copper-alloy ring; fragment; diam. (estimated): 1.8 cm; width: 0.7 cm; it is composed of two very thin layers connected to one another (each is 0.05 cm thick).

Iron arrowhead; complete; length: 6 cm; width: 1.2 cm; thickness: 0.5 cm; the arrowhead is a square cross-section and pointed tang. It can also be identified as a spearhead or a catapult projectile. Parallel: From the first century BCE: Jerusalem, the Jewish Quarter (Gutfeld and Nenner-Soriano 2006: pl. 12.1:M 1).

M 7. Area E, L.823, 8101.02; Hellenistic

Cosmetic Implements and Jewelry M 2–M 4

Iron nail head; fragment; diam. of the head: 4 cm; width of the leg: 1.5 cm; this nail is rounded in section. Usually nails were used for carpentry work, but because of its size this nail could also have been used as a tool pin.

Rods M 2–M 3

3.4.2 Glass

Yael Gorin-Rosen

Two rods were found in Kh. er-Rasem. On one of the ends there is a small spoon. They were used as cosmetic implements (kohl sticks). A large group of kohl sticks was found in Masada. Some of them were found in the casemate rooms in the wall, adapted as living quarters by the Jewish defenders (for an expanded discussion on the subject, see: Krause 2005:15–25).

Five small fragments of glass vessels were found, of which only two, although very small (and therefore not illustrated), are indicative. The fragments are all of blown vessels. One small fragment of a rounded upright rim of a bowl, wine-glass or a bowl-shaped oil lamp made of bluish-green glass (L.808, B.8023.01) was found in a fill of a cell. The second is a tiny fragment of a small concave base, probably from a bottle made of greenish glass (L.179, B.1251.01), which was found in topsoil outside of the main building. The material of the glass fragments, as well as their shape, places them in the Byzantine period. In addition, a small disk made of faience was found in Area B (L.333, B.3118.02). The disk is of unidentified type and measures 1.6 cm in diameter and has two holes on each side.

M 2. Area B, L.321, 3117.01; Hellenistic Copper-alloy rod; almost complete; Length: 15.5 cm; thickness: 0.3 cm; elongated rod, rounded in cross-section. One end of the rod is shaped like a small spoon (1.7 X 1.1 cm). Where the rod meets the spoon there is a decoration of an engraved spiral.

M 3. Area B, L.321, 3117.01; topsoil

3.4.3 Beads

Copper-alloy rod; complete; Length: 13.8 cm; thickness: 0.3 cm; elongated and twisted rod, rounded in cross-section. One end of the rod is shaped like a small spoon (1.2 X 0.7 cm). There are lines engraved in disarray along the rod.

Deborah Cassuto Three artifacts described as beads were found at Kh. erRasm (Fig. 3.5). These were discovered in clear contexts in different parts of the site. Two of the specimens are definitely beads while the third, significantly larger, a round

Parallels for M 2 and M 3: From the Hellenistic period: Qalandia (Magen 2004: fig. 128a, pl. 14:1–2); from the first century CE: Masada (Krause 2005: pl. 2:15).

163

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

presence at the site, as simple as they are, indicates that the occupants of the ancient structure possessed such objects of adornment as beaded jewelry, which they took with them upon abandoning the site, leaving behind the few beads that had fallen at some time during the site’s occupation. Since no comparative quantitative data are available, we cannot pursue this line of research.

3.5 Faunal Remains Inbal Shoam, Noa Raban-Gerstel and Guy Bar-Oz Introduction This report presents the faunal remains from the excavations at Kh. er-Rasm. All bones provided by the excavator were examined and documented. The faunal remains yielded domestic livestock bone fragments. Deposits, mainly fills, from an adjacent cave (2001 excavation season) were also examined and are included in a separate table. The complete research protocol and dataset for each identified bone can be acquired from the authors. A general summary of the fauna is given.

3.5 Beads

perforated stone object, may have had a different function altogether. Table 3.3: The Beads No.

Locus

Basket

No. 1 No. 2 No. 3

186 518 821

1268.02 5050.01 8092.01

Faunal Analysis Procedures Identified bone elements were sorted and coded in an Excel worksheet according to area of excavation, locus, square and basket. Recorded elements were inspected for cut marks, burning activities and other macroscopic bone surface modifications, such as signs of animal activity (Lyman 1994).

Beads 1 and 2 are both biconical monochrome beads in a deep blue with an iridescent weathering, similar to Spaer’s type 46 (Spaer 2001:64, 74, pl. 5). Although this form goes back to as early as the second millennium BCE, by the Hellenistic and Byzantine periods it can be found dispersed widely throughout the eastern Mediterranean (ibid.:64).

Bone remains were identified to bone element and species using the comparative collection of the Laboratory of Archaeozoology at the University of Haifa. The relative abundance of the different taxa were quantified using the number of identified specimens (NISP) for each taxon together with the minimum number of individuals (MNI) from which the remains could have originated. These values were calculated using the assumptions described in Klein and Cruz-Uribe (1984). Age at death was analyzed based on epiphyseal closure and tooth wear (Klein and Cruz-Uribe 1984).

Bead No. 1 is biconical with one end higher than the other. End-to-end it measures 0.45 cm with the higher end measuring 0.35 cm and the shorter end 0.1 cm. It measures 0.695 cm at its widest part and the perforation is 0.8 cm at the taller end and 0.22 cm at the shorter one. Bead No. 2 has two equal ends and measures 0.47 cm end-to-end. It measures 0.58 cm at its widest part with the perforation at one end 0.13 cm and at the other 0.186 cm.

The Finds

Bead No. 3 is a large biconical worked bead; a stone carver’s marks can be discerned along the sides. End-toend it measures 2.24 cm and at its widest part 1.97 cm. The perforation measures 0.64 cm at both ends. Technically this could have functioned as a spindle whorl; however, this specimen is narrower in the middle than typical biconical spindle whorls of the Hellenistic period.

A small assemblage of 49 complete and fragmentary animal bones was recovered from the combined excavated area (Table 3.4). In addition, 34 bones were found in the adjacent cave, many of which seem to be of recent origin (Table 3.5). The faunal remains from the excavated area (the main building) consist of cattle, sheep and goats (Table 3.6). Butchery marks were found on a single sheep/goat humerus. Signs of burning were observed on two sheep/ goat specimens.

It is interesting that despite the extensive sifting of soil conducted during the excavation, so few beads were found. Used for adornment, beads could have been considered luxury items, and their absence may reflect the poverty of the residents at Kh. er-Rasm. On the other hand, their

The assemblage contained immature cattle (one femur shaft

164

Avraham Faust and Adi Erlich

Table 3.4: Distribution of Faunal Remains According to Area, Locus and Basket Area

A

B D

Locus

Basket

# Bones / Basket

# Bones / Locus

157

1168

1

1

160

1174

2

160

1180

1

186

1258

1

186

1268

7

186

1270

6

311

3026

2

2

703

7003

10

10

7004

23

7005

1

8013

1

705 805

E

# Bones / Area

3 18 14 2 34

24 1

808

3039

1

808

8019

3

808

8023

6

808

8039

1

808

8057

1

816

8081

2

816

8095

3

817

8073

1

818

8072

1

818

8077

1

818

8085

1

819

8088

2

2

823

8101

3

3

824

8100

2

2

12

5

29

1 3

Table 3.5: Number of Identified Specimen (NISP) and Minimum Number of Individuals (MNI) of Each Taxon Represented at Kh. er-Rasm Cave (Area D) Sheep/Goat

Equus sp.

Testudo graeca

Aves (Gallus sp?)

Common Name

Cattle

Head:

 

Horn

 

1

 

Cranium

2

4

 

Mandible

1

4

 

Teeth

 

Body:

 

Ver: Lumbar

 

Rib frag.

 

Carpace/plastron

 

Forelimb:

 

Scapula

1

Humerus

 

Radius

 

1

1

2

  1

 

  1

3

 

165

Total

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

Sheep/Goat

Cattle

Ulna

 

Hindlimb:

 

Pelvic frag.

 

Femur

 

Tibia

1

3

 

Metatarsus

 

1

 

Astragal

1

 

Coboid

 

1

Equus sp.

Testudo graeca

Aves (Gallus sp?)

Common Name

 

1

1

 

Toes:

 

Phalanx 1

 

1

Phalanx 2

2

 

Metapod cond.

 

NISP

8

17

1

1

7

%NISP

24%

50%

3%

3%

21%

MNI

1

1

1

1

3

 

Cattle

Sheep/Goat

Head:

 

Mandible

1

5

Teeth

2

9

Body:

 

Ver: Lumbar

 

1

Rib frag.

 

1

Forelimb:

 

Humerus

 

4

Radius

 

3

Ulna

 

1

Hindlimb:

 

Pelvic frag.

 

2

Femur

1

4

Tibia

3

5

Metatarsus

3

 

Toes:

 

Phalanx 1

 

1

Metapod cond.

 

3

NISP

10

39

%NISP

20%

80%

MNI

1

3

34 7

and one tibia shaft), and a smaller percentage of immature sheep and goats (one proximal humerus). This high ratio of juvenile cattle suggests that they were slaughtered primarily for their meat while sheep and goats were probably used for secondary products, such as milk and/or wool. An animal economy based on large herds of sheep and goats exploited primarily for milk or wool, and less for meat, is the “traditional Middle Eastern subsistence pattern” (sensu Horwitz, Tchernov and Dar1990).

Table 3.6: Number of Identified Specimen (NISP) and Minimum Number of Individuals (MNI) of Each Taxon Represented at Kh. er-Rasm (Areas A, B and E) Common Name

Total

Total

3.6 Plant Remains Ehud Weiss Introduction This report presents the plant assemblage from Kh. erRasm. 99 samples arrived at the lab for analysis and all of them went through the analysis process. Most samples were of one bucket (ca. 10 liter) of dirt and some were the contents of several vessels. From these vessels, only one cooking pot (Area A, Square G48, Basket 8077) produced plant remains – one lentil seed and two olive pips (Appendix 3.6.1). Inside the rest of the vessels – three oil lamps, four bottles, one juglet, and an additional cooking pot – nothing was found. The plant assemblage retrieved from these samples is included in this report. As a whole, the Kh. erRasm plant assemblage is small, a total of 111 seeds, with a low density of plant remains across the excavated loci.

49

Methods

4

We will provide a short description of the procedures that

166

Avraham Faust and Adi Erlich

were employed to retrieve and analyze the Kh. er-Rasm plant assemblage.

The generally low frequency of plant remains and the prevalence of tough olive stones accords with the low level of preservation observed throughout the plant assemblage. Appendix 3.6.2 also presents the relative density of plant remains in the excavated sediment, which is quite low – 0.31 seeds/liter. In the different samples the density ranges between 0.125 seeds/liter to 2.125 seeds/liter.

1. Sampling and Recovery The samples arrived at the archaeobotanical lab after the excavation came to an end. Samplings for flotation of charred plant remains were done by the excavator, who collected buckets of dirt from most loci. Samples were collected in 10-liter buckets, not completely filled, approximately 8 liters, each sample was of one bucket. A total number of 99 flotation samples, totaling ~792 liters of soil, were collected during the 1998–2003 seasons (mainly during the 2003 season, which was also the largest one); only 43 samples yielded charred plant remains.

As a result of these factors, most plant remains were identified to the genus or even family level only (Appendix 3.6.1). One of the most interesting issues in the work on the Kh. er-Rasm plant assemblage was the abundance of recent plants within the samples. Many of the samples contained recent, uncharred plants, and hosts of dead ants or other insects. The general feeling was that some ant colonies were dug out, although the excavation team did not notice such phenomena while excavating. Moreover, many of the samples features lay beneath a thick layer of fallen walls with no visible burrows between the stones. Since the material was collected and packed immediately after flotation, we can assume that the material was not placed near an active colony after the samples were taken.

Charred remains were recovered using flotation machine technique. A 1 mm screen was used to hold the heavy residue within the flotation tank and a 0.3 mm sieve caught the floated-off light fraction. After the silt was shaken through the screen, the heavy fraction was recovered and laid out to dry in the sun. The light fraction was taken indoors to dry slowly. After the light and heavy fractions had dried, they were sorted according to their content. The plant remains, mainly from the light fraction, were packed in cardboard boxes and sent to the lab.

The plant assemblage is comprised predominantly of cultivated plants (88.3%) and to a lesser extent of weeds (9.9%); some of the plant remains were unidentifiable (1.8%) (Appendix 3.6.1). It is noted that this assemblage is relatively ‘clean’ in term of annual crop processing – there are no chaff remains in any loci.

2. Laboratory Work General The laboratory work was carried out in the archaeobotanical laboratory in Bar-Ilan University, Israel. The plant remains were sorted out and identified under a stereo-microscope using magnifications up to x50. They were identified (mainly on the basis of their external morphology) by sorting types, size, measurements, shape and surface texture. In addition, for comparative purposes we used manuals and seed atlases (such as Bertsch 1941 and Beijerinck 1947 that were sometimes used as a preliminary guide), together with the “computerized key for grass grains of Israel and its adjacent regions” (Kislev et al. 1995; 1997; 1999). But ultimately, a reference collection was used to identify the plant remains. Latin plant names are according to Danin (2004), and common names are after Fragman et al. (1999).

Due to the very limited plant assemblage, reconstructing the diet and environment of Kh. er-Rasm inhabitants is problematic, and only a rough observation can be made. Generally, this plant assemblage hints at a typical Mediterranean diet, encountered already at other sites in the area. It includes crop plants: cereals, pulses and fruit, as well as weeds of cultivated plants and wild plants. Cereals are represented by wheat and barley, four grains each, pulses by three lentils seeds, and fruits by 83 olive stones and three grape pips. In addition, there are 11 seeds of wild plants and weeds.

Appendices (at the back of the volume)

Quantification

3.6.1: Number of identified seeds of each taxon represented at Kh. er-Rasm.

The following method was used for counting the finds – seed fragments with more than half the seed were recorded as one; every tip of a grape pip was counted as one.

3.6.2: Distribution and density of plant remains retrieved from Kh. er-Rasm.

The Plant Assemblage

3.7 Dendroarchaeological Investigations

A total of 111 seeds were analyzed from the Kh. er-Rasm excavation. The frequencies of plant remains in the different samples are very low, between one and 17 seeds/sample. Distribution of plant remains retrieved according to area and square are detailed in Appendix 3.6.1. The most common find was olive stones that comprise some 75% of all finds.

Nili Liphschitz During the excavations, numerous charred wood remains and a single carbonized seed were collected from three Areas: A, B, and E.

167

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

in the oven for one week. Blocks were made in paraffin. Cross sections as well as longitudinal tangential and radial sections of 12 micron thickness were made with a rotary microtome. The identification of the samples up to the species level, based on the three-dimensional structure of the wood, was made microscopically from these sections. Comparison was made with reference sections prepared from systematically identified recent trees and shrubs and with anatomical atlases. The carbonized seed was identified morphologically.

Table 3.7: Location of tree species at Kh. er-Rasm Area

Locus

Basket

Tree Species

A “ A “ “ “ “ “ “ “ “ “ “ “ B “ E “ “ “ “ “ “ “ “ “ “ “ “ “ “ “ “ “ “ “

Square I48 “ “ “ “ “ “ A48 J51 “ “ I49 “ I 53 G53 “ F48 “ “ “ “ F49 “ “ “ G48 “ G48 “ “ “ “ H48 “ E49 “

132 “ 160 “ “ 162 164 162 126 “ “ 165 “ 186 323 “ 814 “ 821 “ 823 807 808 817 “ 808 “ 810 “ “ 818 824 813 814 822 825

1075 1081 1171 1174 1180 1177 1165 1181 1061 1067 1071 1176 1182 1268 3050 3053 8070 8076 8084 8092 8101 8062 8057 8073 8078 8072 8077 8044 8052 8065 8085 8100 8071 8063 8094 8099

A

I50

163

1164

A

I48

160

1160

E

G48

810

8044

Olea europaea “ “ “ “ “ “ “ “ “ “ “ “ “ “ “ “ “ “ “ “ “ “ “ “ “ “ “ “ “ “ “ “ “ “ “ Pistacia palaestina Pistacia sp. Stone of Olea europaea

In all, 38 wood samples were retrieved from the excavations. As can be seen from the results (Table 3.7) 36 of the 38 wood samples were of Olea europaea (Olive), a single sample was of Pistacia palaestina (Terebinth) and another sample could be identified only as Pistacia sp. due to its poor state of preservation. A single carbonized olive stone was also found. The fact that 94% of the wood assemblage found at Kh. er-Rasm was of Olea europaea wood testifies to the presence of olive orchards in the vicinity of the site during the Hellenistic period. Trees that ceased to give fruits were used as a source for construction timber and for an energy supply. Olea europaea and Pistacia palaestina trees grow today in the environs of the site.

3.8 Shells Daniella E. Bar-Yosef Mayer This assemblage consists of 49 shells, including 40 shells of landsnails, and nine of marine bivalves. 23 of the landsnails are complete specimens consisting of Buliminus, Calaxis, Helix and Levantina. The rest are broken or fragmentary landsnail shells of Helix, Levantina, Sphincterochila and one possible Trochoidea. All are local species that probably inhabited the site any time between its occupation during the Hellenistic period and the present. Indeed, some specimens derive from occupation levels, and others were recovered in the topsoil (for the database, see Appendix 3.8.1 at the back of the volume). The nine marine specimens, all bivalves, consist of five valves of Donax trunculus, a common Mediterranean species, one valve of Glycymeris insubrica, as well as two fragments of Glycymeris and one unidentifiable fragment of a large bivalve.

The site is located in a Mediterranean area, with a mean annual precipitation of ca. 450 mm (measurements of mean annual precipitation were made at Netiv Ha-Lamed Heh – map reference 149/122, elevation: 400 m: 433 mm and at Beit Jimal – map reference 147/125, elevation: 360 m: 493 mm) (Meteorological Notes, 1967). The present day arboreal vegetation of the area is of Mediterranean maquis, garigue and batha.

None of the shells were processed or used as artifacts. Moreover, Glycymeris, which are the most common shells on Israeli shores, are remains of a population on the verge of extinction (Sivan et al. 2006). It is worth mentioning, however, that Donax is an edible species and may have been collected as food, although the five valves that are the remains of three specimens at most are not a strong enough indication for this activity.

Samples of 0.5–1 cubic cm were taken from the charred wood remains for botanical identification. The samples were aspirated in absolute ethyl alcohol, dipped in Celloidin – clove oil solution for 24 hours, rinsed in absolute ethyl alcohol and transferred to paraffin 50–55 C

The same species were discovered in other sites of roughly the same period: Fairly large numbers of Donax were encountered in the Hellenistic site of Kom Ge‘if (Ancient Naukratis) in Egypt, where it is also considered food debris 168

Avraham Faust and Adi Erlich

Table 3.8: Sample Details and Radiocarbon Results Number

RTT

5358

Type

Olive pit

14C Age ± 1σ Year BP

2210 ± 50

RTT

5227

Olive pits

2190 ± 45

RTT

5226

Olive pits

2170 ± 45

RTT

5252

Olive pits

2130 ± 45

RTT

5359

Seeds

2055 ± 35

Calibrated Age 68.2% probability 370 BC ( 9.8%) 340 BC 330BC (58.4%) 200BC 95.4% probability 400BC (95.4%) 160BC 68.2% probability 360 BC (40.2%) 270 BC 260 BC (28.0%) 190 BC 95.4% probability 390 BC (92.9%) 150 BC 140BC ( 2.5%) 110 BC 68.2% probability 360 BC (35.1%) 280BC 260 BC (33.1%) 160BC 95.4% probability 380 BC (95.4%) 90BC 68.2% probability 350 BC ( 7.1%) 320 BC 210 BC (57.9%) 90 BC 70 BC ( 3.2%) 50 BC 95.4% probability 360 BC (18.1%) 280 BC 260 BC (77.3%) 40 BC 68.2% probability 160 BC ( 6.2%) 130 BC 120 BC (62.0%) AD 95.4% probability 170 BC (95.4%) 30 AD

(Reese 1997). Interestingly, Donax is only rarely mentioned as a food preference in the Aegean region (Karali 1999:14– 17). A few specimens of Glycymeris were retrieved in the Hellenistic levels of the City of David (Mienis 1992).

Collection Site

Sample ID

δ13C ‰ PDB

Kh. er-Rasm, Area E, Square G48, L.824, B 8100

68/1

-20.8

Kh. er-Rasm, Area E, Square F49, L.808, waste location below RTT 5226

89/1

-20.6

Kh. er- Rasm, Area E, Square F49, L.808, waste location

86/1

-20.95

Kh. er-Rasm, Area E, Square H48, L.823, B 8101

54/1

-20.3

Kh. er-Rasm, Area E, Square G48, L.818, B 8077

35/1

-20.6

would introduce a further uncertainty in the dating due to the old wood effect, as it would have shifted to older dates by an unknown amount. The amount of seeds collected was very small. None in the sample can be regarded as a cluster. Since single seeds could be redeposited in the archaeological record, the above observation should be considered for the radiocarbon age interpretation.

Unfortunately, our knowledge of shell assemblages from Hellenistic sites is very limited, therefore it is premature to draw any definite conclusions on their use, or the reason for their transportation to the site, approximately 30 km away from their source in the Mediterranean.

The five samples analyzed for radiocarbon were all identified. For details see Table 3.8. Samples RTT 52265227 are from the same waste pit in L.808 with RTT 5227 deeper in the pit than RTT 5226. In the pit, many small (unrestorable) pottery sherds and a small number of small bones were recovered.

3.9 Radiocarbon Dating Elisabetta Boaretto Introduction

Samples RTT 5252, RTT 5358 and RTT 5359 are from different contexts related to a destruction event identified as a large accumulation of ash. RTT 5252 and RTT 5358, though from different rooms, both came from the thick ash layer just below, or within, the level which was defined as a floor and on which complete pottery vessels were found. Sample RTT 5359 was a few cm above RTT 5358, and the locus was defined as occupational debris below the wall fall (it was also covered with ash).

Radiocarbon dating was performed on five charred samples collected during the 2003 excavation season at Kh. erRasm. Material for dating was collected from various loci and a few samples were selected for dating from different areas and contexts. The carbonized samples were all short-lived samples, either olive stones or seeds, collected by flotation of the floor sediment material. Although there was abundant charcoal at the site, to improve the accuracy of the date only shortlived samples were selected for the dating. Wood charcoal

According to field observation three of the samples were

169

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

Table 3.9: Location and context of the sediments analyzed with FTIR Area

Locus

Square

Basket

Locus type

E E A A

824 802 185 173

G48 G48 I52 J53

8100 8005 1256. 1203

Ash layer below wall fall and occupational debris Topsoil Occupation level below wall fall Topsoil outside of the structure

3.6 Probability distribution of the radiocarbon dates

related to floors made of unpacked sediments with ceramic deposited on them. The ceramic is typologically dated to the end of the second century BCE. RTT 5252 and RTT 5358 appear to have come from inside the floors, while RTT 5359 came from the occupational debris on the floor; the two other samples, RTT 5226 and RTT 5227 came from a waste pit.

calibrated ages are given for ±1 and ±2 standard deviation. The probabilities are given in the table for each interval.

Sample Preparation

The sediments analyzed are given in Table 3.9:

Samples were pre-treated using the procedure presented in Yizhaq et al. 2005. All seeds were cleaned in order to eliminate the presence of possible carbon contaminants that could interfere with the dating. The pre-treated material was then oxidized to CO2 in the presence of copper oxide at 900ºC in vacuum. The extracted CO2 was then prepared as graphite for the Accelerator Mass Spectrometry measurement.

Sediments from loci 824, 802 and 185 were related to archaeological contexts. Sediments from locus 173 are considered a control sample as it came from topsoil outside the structure.

Sediments from different loci were analyzed using Fourier Transform Infrared analysis (FTIR). The FTIR analysis was performed on the sediments before and after an acid dissolution step aimed at removing the carbonate fraction.

The major fraction in the raw material was represented by carbonate and clay minerals, with the carbonate constituting 53% to 60% of the mass of the sample. After carbonate dissolution all the samples show only the presence of clay. Only sediment from L.824, described as ash layer, provided a spectrum indicating a significant presence of charcoal. This is also the only sediment sample related to a radiocarbon sample (RTT 5358).

Results All the information about samples and the results obtained is presented in Table 3.8. 14C ages (4th column) are reported in conventional radiocarbon years (before present =1950) in accordance with international convention (Stuiver and Polach 1977). Thus all calculated 14C ages have been corrected for fractionation so as to refer the results to be equivalent with the standard d13C value of -25‰ (wood). Calibrated ages (5th column) in calendar years have been obtained from the calibration tables in Reimer et al. 2004 by means of the 2005 version OxCal v. 3.10 of BronkRamsey (Bronk-Ramsey 1995; Bronk-Ramsey 2001). The

Discussion Radiocarbon dates obtained with samples details and calibrated dates, are given in Table 3.8. The probability distributions of the calibrated dates for the five samples are given in Fig. 3.6. Radiocarbon determinations from the different samples are

170

Avraham Faust and Adi Erlich

3.7 Probability distribution of the average for the samples RTT 5227 and 5229

very similar, indicating a relatively short occupation of the site between the fourth and first centuries BCE.

Assuming that the charred material is in situ, the latest date (RTT 5359) will provide the terminus post quem for the destruction. RTT 5359 was found in a layer described as occupational debris/ash layer. The sediments from this layer, also described as ash layer, were not available for the FTIR analysis. Based on the above consideration the assumed destruction event would be dated later than sample RTT 5359 or contemporaneous with it. This would be between 160–0 BCE with 68.2% probability (±1σ) and between 170 BCE–30 CE with 95.4% probability (±2σ).

The dates from the waste pit (RTT 5226-5227) are essentially the same, showing that the pit was used only for a short period. The average of the two results is: 2180±32 14C year BP. The calibrated ranges are for ±1σ (68.2% probability) 360 BCE (42.9%) 280 BCE and 240 BCE (25.3%) 180 BCE, and for ±2σ (95.4% probability) 370–160 BCE. These ranges include the second half of the fourth century to the first half of the second century. Although there is a very small standard deviation from the average (±32 y), the calibrated range is still about 200 years. This is an effect due to the presence of a small plateau in the calibration curve between 350 and 200 BC. The probability distribution of the average is shown in Fig. 3.7.

It appears that the other samples – two from within the floors and two from the waste pit, represent the occupation of the site prior to the destruction.

Conclusion Radiocarbon dating of five short-lived samples from Kh. erRasm puts them in the last 400 years BCE. Although the site is defined as single occupation, the dates have a distribution, possibly representing the site’s occupation period. There is a possibility of seed redeposition, but as the dates are very close and there is insufficient evidence for very long occupation, its effect may be negligible in this case. Based on the radiocarbon dates, the final destruction can be dated

The contexts related to the other three samples (RTT 5252, 5358 and 5359) are floors made of loose sediments with restorable ceramic vessels. The sediments were identified in the field as ash sediments. Two of the samples came from the ash layers below the level of the vessels (RTT 5252 and RTT 5358), while the third (RTT 5359) came from the same level as the ceramic vessels.

171

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

3.8 Plan of the site with location of Areas Z1 and Z2

172

Avraham Faust and Adi Erlich

Table 3.10: Layers in Square F49, North Side Depth (cm)

General Description*

Color** (Dry)

Texture

Structure

Boundary

1

0–30

Surface: Anthropogenic stony gray-brown rendzina sediment

10YR6/1-5/2 Gray-grayish Brown

Clay loam

Granular/ crumb to massive

Clear to abrupt

2

30– 110/120

Artificial fill

10YR7/2 Light gray

Stony with clay loamy matrix

Massive collapsed material

Clear

3

110– 120/130

Brown rendzina soil

7.5YR4/4 Brown

Clay

Angular to subangular

Abrupt

4

120+

Nari bedrock

Layer

Note

Fill composed of chalk fragments and fine fill material Buried B horizon of brown rendzina Signs of gray and manganese concretions

*Soil classification after Dan et al. 1972, 1976. **Color according to Munsell 1998.

to the second or first centuries BCE. Because of the ceramic typology (Chapters 2.2, 2.4), the abandonment of the site could have been no later than at the end of the second century.

Pottery sherds embedded in various layers enabled age assessment of those layers. Fraction size distribution of fine material (grains < 2 mm) was measured by hydrometer using the sedimentation method (Klute 1986) at the Field Service Laboratory at the regional agriculture research center, Zemach, Jordan Valley.

3.10 Geomorphological/Sedimentological Structure

Occupational / Destruction Area (Square F49, North Side)

Oren Ackermann

Four primary layers were identified in this section. They are as follows, from top to bottom: 1. Current surface; 2. Wall fall layer (destruction); 3. Brown rendzina (Lithic Haploxeroll) soil; 4. Nari bedrock (hard calcrete crust). Data from this section are summarized in Table 3.10.

Aims The aims of this part of the research are as follows: • Reveal the paleo surface previous to site establishment • Study the sediments of the occupational level (wall fall above destruction layer) of the site • Examine the post-site layers • Characterize the site terraces

Layer 1, measuring from surface to depth 30 cm, is composed of anthropogenic stony gray-brown rendzina sediment/soil (henceforth sediment), with a clay loam structure. Layer 2, with a depth of approximately 80 cm, is composed of architectural remains and fine materials. Layer 3, a relatively thin layer of just 20 cm, is composed of brown rendzina soil. This is probably a buried B horizon, perhaps a relic of the paleo-surface prior to human settlement at the site. Layer 4 is nari bedrock that covers chalky bedrock.

Methodology Geomorphological / sedimentological study and characterization were conducted for representative sections of an architectural area (wall fall / destruction debris) within the site (Square F49) and in two terraces (Z1 and Z2). Z1 contained fill sediment of 60 cm in depth; Z2 contained fill sediment of more than 100 cm in depth. In each terrace, probe sections were dug in the upper and lower sides of the retaining walls (Fig. 3.8). In Z1, four sections were dug; two on the eastern side (Z1E; upper and lower sections), and two on the western side (Z1W; upper and lower sections). In Z2, two sections were dug; one on the upper side (Z2U), and one on the lower side (Z2L).

The site is located at the top of a hill, making it unlikely that slope or alluvial sedimentation contributed to the fill in Layer 1. This raises questions regarding the origin of the surface fill. In Israel, additional dust in the soil is a wellknown phenomenon, caused by aeolian dust transportation from the desert in the south (Dan 1990). The source of fill in Layer 1, therefore, is probably dust accumulation and sedimentation over time. It seems that the archaeological remains acted as dust traps. This assessment is strengthened by the fact that the surface soil has a clay loamy structure with relatively high silt content ranging from 33% to 54%.

Sediment and soil stratigraphy were characterized in the profile sections according to structure, texture, and color.

173

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

3.9 Section in Z1

3.10 Section in Z2

3.10). Terraces are generally constructed as consecutive steps along a slope. They are composed of artificial fill that was shifted from the vicinity and deposited on outcrop or soil pockets. In general, the lower part of the fill contains rock fragments and the upper part contains soil mixed with organic matter. Terrace fill is retained by a dry stone wall (built without mortar) founded on bedrock. Gaps between the stones allow water surplus during heavy rain events to flow downwards, thereby preventing the fill from becoming waterlogged. There are several kinds of terraces. Those built on slopes, known as “contour terraces,” are the type found at the site.

It is important to mention that after the site was abandoned, its surface was used for agricultural purposes for hundreds of years. The surface therefore underwent cycles of ploughing and seeding that may have caused pedoturbation and artificial soil shifting. More detailed study is needed to verify this possibility.

Terraces: General Introduction Agricultural terraces and their functions are defined and described in detail by Gibson (2001). Their primary purpose is to extend agricultural areas by providing artificially flattened fields in areas with sloped surfaces (Figs. 3.9,

174

Avraham Faust and Adi Erlich

3.11 Eastern section in Z1 facing south

Terraces in the Site

and the nari bedrock was exposed at a depth of 5–10 cm (Fig. 3.11). In Z1E lower the nari was exposed at a depth of about 40 cm and in Z1W lower, it was exposed at a depth of 50 cm (Figs. 3.12, 3.13).

Two primary terrace types were identified at the site: 1. Shallow terraces composed of sediment fill with a depth of 60 cm and less. 2. Deep terraces composed of sediment fill with a depth exceeding 100 cm (Figs. 3.9, 3.10). Shallow terraces allow high water retention — even saturation — during heavy rainfall events. The water evaporation ratio in these terraces is high, and dry soil conditions often exist between rain events and during the dry summer season (Kutiel 2000). Deep terraces also enable high water retention during heavy rainfall events; however they may continue to hold water for long periods of time, thereby making water available during the dry summer season (Bruins 1986; Gibson 2001; Frederick and Krahtopoulou 2000). The presence of two types of terraces invites the question of whether each type had a different function, and whether shallow terraces could have served as agricultural fields.

In the upper probe section of Z1W (Fig. 3.14), three primary layers were identified. They are as follows, from top to bottom: Layer 1, from surface to depth 40/50 cm, is composed of anthropogenic stony gray-brown rendzina sediment, with a clay loamy structure. Layer 2, with a depth of approximately 10 cm, is composed of brown rendzina soil. It is probably a buried B horizon, and may be a relic of the paleo surface soil prior to the establishment of the terrace. Layer 3 is composed of nari bedrock (at the depth of about 60 cm). In the lower probe sections, two primary layers were identified. These were similar to Layers 1 and 3 in the upper probe section. Data from two representative probe sections, Z1W upper and Z1E lower, are summarized in Tables 3.11 and 3.12.

Probe Sections in Artificial Terraces Terrace Z1 (W and E) (W.37)2

any indicative pottery (one Hellenistic sherd, from what should be regarded as topsoil, near the entrance to the building; Z1E upper). Excavations in Z1W upper, which was much deeper, did produce a number of sherds, including three Hellenistic sherds from the upper and mid-section baskets, and one Late Iron Age or Persian period sherd (for the pottery, see Appendix 2.6 at the back of the volume). See more below.

In the Z1E upper probe section, the soil is very shallow The ceramic finds from within this terrace were quite meager. Excavations in Z1E, naturally (as it was very shallow) contained hardly 2

175

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

3.12 Eastern section in Z1 facing west

3.13 Western section in Z1 facing south

3.15 Upper probe at Z2 facing west

3.14 Western section in Z1 facing north

176

Avraham Faust and Adi Erlich

Table 3.11: Section in Terrace Z1W, Upper Probe Depth (cm)

General Description*

Color** (Dry)

0–40/50

Surface: Anthropogenic stony gray-brown rendzina sediment

10YR4/2-3 Dark grayish brown to brown

2

50–60

Brown rendzina soil

7.5YR4/4 Brown

3

60

Nari bedrock

Layer

1

Texture

Structure

Boundary

Clay loam

Granular/ crumb to massive

Abrupt

Clay loam

Granular/ crumb to sub angular blocky

Abrupt

Notes Angular rock fragments 3–5 cm, gravel up to 10 cm; signs of secondary calcium carbonate sediment Paleo surface comprised of 5–8 cm-long subangular rock fragments

*Soil classification after Dan et al. 1972, 1976. **Color according to Munsell 1998. Table 3.12: Section in Terrace Z1E, Lower Probe Layer

1 2

Depth (cm)

General Description*

Color** (Dry)

0–50

Anthropogenic stony gray-brown rendzina sediment

10YR4/2-3 Dark grayish brown to brown

50

Texture

Structure

Clay loam

Granular/ crumb to subangular massive

Boundary

Notes

Abrupt

Angular rock fragments 5 to 10 cm; signs of secondary calcium carbonate sediment

Nari bedrock

*Soil classification after Dan et al. 1972, 1976. **Color according to Munsell 1998. Table 3.13: Section in Terrace Z2U, Upper Probe Layer

Depth (cm)

1

0–40/50

2

40–80

3

80+

General Description*

Color** (Dry)

Texture

Structure

Boundary

Anthropogenic stony gray-brown rendzina sediment Rock fragments mixed with fine material

10YR5/2 Greyish brown

Clay loam

Granular/ crumb to massive

Abrupt

10YR7/2 Light gray

Clay loam

Massive

Clear to abrupt

Clay

Subangular blocky

Brown rendzina paleosol

7.5YR4/4 Brown

Notes Signs of secondary calcium carbonate sediment Chalk rock fragments up to 10 cm in length with a few chiseled stones Signs of secondary calcium carbonate sediment; bedrock was not exposed

*Soil classification after Dan et al. 1972, 1976. **Color according to Munsell 1998.

Table 3.14: Section in Terrace Z2L, Lower Probe Depth (cm)

General Description*

Color**

1

0–15/30

Surface: Anthropogenic stony gray-brown rendzina sediment

10YR5/3-2 Dark grayish brown –brown

2

15/30+

Nari bedrock

Layer

Texture

Structure

Boundary

Notes

Granular/ crumb to massive

Abrupt

Shallow soil with nari and chalk fragments of 5–10 cm in length Nari bedrock with soil pockets of rendzina

*Soil classification after Dan et al. 1972, 1976. **Color according to Munsell 1998.

177

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

Terrace Z23

capability, and that shallow terraces are sensitive to high water evaporation. This puts into question the likelihood that the shallow terraces in this site — those with a depth of less than 50 cm — were able to sustain agriculture. Examining this question in light of the architectural archaeological finds at the site suggests that the retaining wall in Terrace Z1 was constructed as a retaining structure for the ramp that passes to the site.4

The upper probe section in this terrace reached a depth of about 125 cm. Nari bedrock was not reached. Three primary layers were identified in this site (Fig. 3.15). They are as follows, from top to bottom: Layer 1, from surface to depth 40/50 cm, is composed of anthropogenic stony gray-brown rendzina sediment soil, with a clay loamy texture. Layer 2, with a depth of approximately 40 cm, is composed of chalky rock fragments mixed with fine material. Layer 3, composed of brown rendzina soil, is probably a buried B horizon. It may be a relic of the paleo surface soil prior to the establishment of the terrace. Nari bedrock is probably located a few dozen cm below this layer. The sediment in the lower probe section is significantly shallower than in the upper probe section, with nari bedrock exposed at a depth of 18 cm. Data from the two probe sections is summarized in Tables 3.13 and 3.14.

It seems, therefore, that the two terrace types identified in the research area served two distinct functions: deep terraces probably served as agricultural terraces and shallow ones as ramp terraces. Future research may look for further evidence of these functions, and investigate other uses for dry stone wall terraces.

3.11 Ground Penetrating Radar and GroundTruth Results

Summary and Conclusions

Jessie A. Pincus

Paleo Surface

During the summer of 2003 Ground Penetrating Radar (GPR) data was acquired at Khirbet er-Rasm. The expedition first enlisted the support of Mnemotrix Systems, Inc. in 2003 for GPR (Ground Penetrating Radar) studies to be done to augment the existing archaeological record, and to assist in decision-making for the coming season. Surveys and excavations have been continuing at the site since 1997. In this paper we will focus on the GPR results and those of the ground-truth excavation that occurred soon after.

The site is located at the top of a hill. Its original surface was composed of soil pockets of brown rendzina alternating with nari outcrops. The nari was apparently the basis for many architectural structures in the site, including ancient buildings and terraces.

Current Surface The current surface is composed of anthropogenic stony gray-brown rendzina sediment with a depth of about 30 cm. This layer was apparently deposited over time by aeolian dust transportation and artificial soil shifting.

How GPR Works Ground Penetrating Radar is a geophysical technique that is used to “see” into the sub-surface. It has been used successfully for archaeogeophysical purposes throughout the world and is a technique that has been used to improve efficiency at archaeological sites for many years (Gaffney and Gater 2003).

Terraces: Preliminary Conclusion The sequence and characteristics of sediment layers in Terrace Z2 enabled high water availability and efficiency. The deep (40/50 cm) first (top) layer allows high water infiltration during heavy rain events. In the second layer, rock fragments allow water percolation downwards along the rock edges. This created high aeration conditions in this layer, which prevent water logging that can cause roots to rot. The third layer is clayey brown rendzina soil; it has a high water retention capability and allows water availability below the aerated layer.

This geophysical technique functions by transmitting radar (electromagnetic energy) waves into the ground from an antenna at the surface level. A portion of these waves are reflected back to the surface when the wave rebounds from an interface between two different materials or sub-surface features. These reflected waves are then detected on the ground surface by a receiving antenna (Gaffney and Gater 2003: 47). Interfaces are detected in the sub-surface by the electrical property changes of the sediment or soil, changes in water content of the material, changes in the bulk density of the geological or archaeological stratigraphy, or even void spaces present due to burials, tunnels, or caches of artifacts (Conyers and Goodman 1997). When the radar wave passes through a material that is significantly different from the previous material (an interface), it creates what is

The combination of sediment structures in Terrace Z2 creates advantageous agricultural conditions. The upper layer enables high water infiltration, the second layer allows high areolation conditions, and the third layer permits high water absorption and retention, enhancing water conditions. Terrace Z1 (W.37) is significantly shallow; its eastern side reaches a depth of only 10 cm, and on its western side reaches a depth of about 50 cm. Previous studies show that terrace fill depth is an important factor in water retention

Interestingly, the excavators (sections 4 and 5) date the initial construction of the main building to the Iron Age. The single early sherds unearthed in the lowest part of Z1W (lower) are insufficient to prove this date of course, but can perhaps serve as another indication for this date (and if this is the case, also for the connection between the terrace [W37] and the main building itself). 4

No indicative pottery (only body sherds) was found in the excavations of Z2 (both upper and lower). 3

178

Avraham Faust and Adi Erlich

called a “reflection.” The greater the difference between materials the higher the amplitude of the reflection, which in turn is evaluated in post-processing as an anomaly of significance.

During post-processing, the GPR signal data collected for Area 1 was analyzed closely as to depth (vertical axis) and movement along the horizontal axis in a 3D cube. A velocity analysis showed that the estimated signal depth of useful data was approximately 2.25 meters, traveling through ground made up primarily of limestone. An overview of the area and the features/anomalies within can be seen in a 3D cube in Fig. 3.18.

As the radar waves are propagated into the sub-surface at a particular frequency, the elapsed time from the initial transmission to the receipt of the wave at the surface is recorded. The reflected waves return at a different velocity based on the electrical properties of the materials through which they were just traveling. The time it takes for the radiowave reflections from the transmitting antenna to rebound off of the layer in the sub-surface and to be received by the antenna at the surface gives us the total travel time for each reflected wave. The combination of this two-way travel time (TWT) and the velocity of the reflected wave is used to compute the depth at which the different reflected interfaces and any objects present in the ground are located. In so doing, a map of sub-surface features within the gridded area can be created. (Conyers and Goodman 1997: 23-28; Gaffney and Gater 2003; Reynolds 1997)

An in-depth animated visualization of the anomalies of Area 1 is available at: http://www.mnemotrix.com/geo/sh_a1.gif. In this animation it is possible to see that as the vertical depth slices proceed from the surface down to a depth of approximately 2.25 meters, the presence of long rectangular anomalies can be traced. These could be interpreted as walls, or features of similar physicality. An intentionally shallow profile was used for the collection of the data in Study Area 1 since the major focus of interest, walls, were known to exist only a few meters down. As is the case with much GPR, in order to be sure of the true identity and location of the features seen in the sub-surface results, ground-truth excavation is the key. GPR does not create an exact picture of features in the sub-surface with exact dimensions. Very often are GPR features found in reality located at a shallower or deeper depth, or thicker or thinner depending on the situation. The scans show areas of significant change and there are most certainly “grey zones” in the dimensions. What it is particularly efficient in is demarcating the general location of objects that are worth excavating in the sub-surface separate from a larger area of lesser or no archaeological interest. If the exact placement is slightly “off”, it is expected and not a surprise.

GPR Results Work focused in three main areas of the site, of which the results have been fully discussed in the original 2003 report, which can be seen on the Mnemotrix website at this address: http://www.mnemotrix.com/geo/shepela.doc. For the purposes of this paper we will address briefly the results of the area that was fully ground-truthed through excavation. Fig. 3.16 shows the location of each original grid on the excavation map used in 2003. GPR Study Area 1 was a long section near the modern entryway to the site from the road. The exact location can be found in Fig. 3.16 below. Some walls had been located nearby, so it was thought that the area would show where a continuation of these walls might be. Therefore a 22 x 2.5 meter grid was marked out, in which tight, overlapping survey lines in perpendicular directions could be acquired.

It is important to recognize that many times archaeogeophysical results are not able to be directly ground-truthed through excavation. Therefore, documentation of the exact offset of features within the sub-surface is hard to find in the literature concerning archaeogeophysics. Thus we paid close attention to the archaeological results of the site in this area and took careful in-the-field measurements. We now turn our attention to the excavation results that occurred soon after our initial survey.

A 400 MHz antenna (with a 100 nanosecond pulse range) was chosen for this task, which can be seen with the survey wheel in Fig. 3.17. This setup was chosen in order to have very high resolution at a relatively shallow depth of 1-3 meters below the surface. The Study Area was chosen for its ease of accessibility in addition to the understanding that known walls ran perpendicular to the area in a north to south direction. Thus a survey that would transect these features in a west to east direction was believed to be most beneficial in determining their continuation in the subsurface.

Archaeological Results Several rooms were discovered through excavation by the presence of walls (see Fig. 3.19). We studied the area where the actual GPR survey had been taken. This can be seen in Figs. 3.19 and 3.20. In the excavated section there is a well-formed wall that runs from the east to the west. This wall was measured to be about 1.0 meter wide (see Fig. 3.20). In addition, there are several smaller walls that run in a north-south direction from the east to the west. These were measured in the field to be about a half-meter wide. The walls were made of large chunks of limestone, about 5 or 6 courses in height, with smaller stones placed at the bottom of each course to

Within the grid, which was marked out in the dimensions of 22 meters (East-West) by 2.5 meters (North-South), GPR signal data was acquired in a tight set of 12 overlapping parallel lines taken first East to West, and then transected by 36 lines in a perpendicular direction North to South. These sets of data were put together into a 3D cube for follow-up study in a post-processing environment.

179

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

3.16 Map of the GPR study areas

180

Avraham Faust and Adi Erlich

3.17 GPR survey study area 1 (Area E before excavation), facing west

3.18 Overview of the features seen in the GPR data from study area 1

181

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

fill in gaps (see Fig. 3.23). During excavation we were told that the now-empty spaces between each section had been filled with stone and soil fill mixed together (see chapter 1.1). Discrepancies became apparent when we found that the excavated walls are located about 2 meters south of their location in the GPR scans.

Discrepancies Between the Archaeogeophysical and Excavated Data Archaeological ground-truthing of the GPR data has shown that the east-west running wall (w31, w42) is located about 1.5 to 2.0 meters south of its location in the GPR data, outside of the GPR surveyed area. North-south walls were visible at different depths in the GPR data. Their placement in the excavations seems to be in the generally correct position (according to the GPR data), within a half-meter buffer zone. All things considered, these are very good results.

Explanation of the Discrepancies As with all types of radar data, various degrees of scatter and attenuation of the signal are present in all scans. Depending on the actual physicality of the sub-surface features of the area, the degree to which this scatter affects the survey results varies. 3.19 Area E (GPR study area 1 after excavation, facing west)

As radar energy is propagated into the sub-surface it spreads in the shape of a cone. It is most narrow when it just exits from the antenna and spreads out wider with depth. Based on the wavelength of the antenna frequency and the sediment type of the sub-surface, the width of this cone varies. This is based on an equation that is explained in Fig. 3.21. This creates what is called a “footprint” of horizontal space that the antenna is able to image at different depths. This means that as the antenna is being dragged on the surface, it is seeing past the boundaries of what is directly below it. It is seeing to the sides, the front, and back. Further, this means that when the antenna is sitting on the edge of a grid, it is actually seeing past those boundaries and into the surrounding area. This is one reason why high-resolution GPR surveying creates a higher accuracy allowing us to lessen the chances for gaps in our data. Calculations to find the range of the footprint present in this survey were completed. The approximate long dimension radius of the footprint (Fig. 3.21: A x 2) was 0.60 - 2.16 meters long in the depth range of 0.25 – 2.0 meters. This is within the distance range of the discrepancies and thus is part of the explanation for the phenomenon in this survey. Additionally speaking, all radar data is affected by focusing and scattering of the signal once it leaves the transmitting antenna. This has been studied and models have been created to understand the effects better. One such example is shown in Fig. 3.22 with various convex and concave surfaces occurring in the sub-surface.

3.20 Area E (GPR study area 1 after excavation, facing west)

182

Avraham Faust and Adi Erlich

3.21 Cone geometry and the “footprint” radius at different depths. GPR waves propagate into the sub-surface in a cone shape (after Conyers and Goodman 1997: 36; courtesy of AltaMira Press)

When the antenna is located above a gradual convex surface (A) little scattering of the signal occurs and there is a low amount of focus back to the receiving antenna. When the antenna is located above a strongly concave surface (B) (e.g. a moat or ditch), there is a high degree of scatter because the signal is reflected multiple times while in the ditch before it makes its way back to the surface. There is an even lower amount of focus as many reflections never make it back to the receiving antenna as they are sent in a completely different direction from where they started. Finally, when the antenna is located above a gradually concave surface (C) what happens is a high degree of focus back to the receiving antenna as they are sent out originally as a cone (discussed earlier). (Conyers and Goodman 1997: 54).

After much study, it seems that it is the combination of these effects (i.e. the cone-shape footprint allowing us to see past the borders of the grid and the amount of focus back to the antenna in a buried concave surface) that has caused these discrepancies in location of the sub-surface features at the site. There is an increased amount of focus back to the antenna because essentially we are dealing with gradually concave edges of the walls. Combined with the high-resolution GPR data acquisition at the site and the footprint that extends past the boundaries of the grid, we can explain why the walls are so distinct in the GPR data (the amount of clarity is rare), and the amount of focus that is present because the walls were indeed discovered at very shallow depths.

Final Comments

When we look at the surface of these walls, seen in Fig.  3.23, it can be argued that we are dealing with a concave surface. This is because the ridges on the sides due to the way the rocks were placed to make the wall create a small concave surface in the middle. In addition, the pockets between each top rock surface are arguably concave (with a high degree of focus), rather than convex (higher amount of signal scattering). These observations thus create the more focused effect that seems prevalent in the survey data.

Ground penetrating radar was used at this site in hopes of increasing efficiency at the location. What we were able to do was to survey three main areas at Khirbet er-Rasm and provide useful data to the archaeologists involved that told them what location would yield the best archaeological results. Because of the particular conditions at the site we were additionally able to document for ourselves the phenomenon that caused the discrepancies and study it in further detail. The more archaeologists are able to

183

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

3.22 Cross-section diagram of a buried surface showing the scattering and focusing of signal data at different antenna locations (after Conyers and Goodman 1997: 54; Courtesy of AltaMira Press)

184

Avraham Faust and Adi Erlich

do not reveal drastic changes in the nature of the area being discussed here.

Source of the Name The term rasm in Arabic indicates signs of destroyed structures (a horvah in Hebrew) and destroyed terraces. Vilnai points out the similarity between the word “roshem” (which means impression/trace) in Hebrew, indicating that at the site the only remains were impressions or traces of the destroyed structure and the site that was inside (Vilnai 1981, Rasm). The Arabic name in Latin transcription appears on topographic maps up until 1956. In later printings of that year the toponym no longer appears, despite the fact that adjacent sites, such as Kh. Shiqlon and Horvat Jenaba a-Tahta, are still indicated on the maps. It should be noted that during this period changes were made in the cartographic procedures in the State of Israel. For one thing, this period was marked by the extensive process of Hebraization of the site names and the removal of other names. Hence from that point on the name of Kh. er-Rasm was removed from the maps until it was almost completely forgotten except by geographers.

3.23 GPR area study 1 – Square I48

ground-truth their archaeogeophysical data, the more accurate archaeogeophysicists will be able to pinpoint the locations of the sub-surface features found in the data. This site has been an important step towards the goal of increased efficiency and research of geophysical methods at archaeological sites.

Roads and Paths

3.12 Khirbet er-Rasm in the Ottoman and British Mandate Periods

Only the main road, which ascends from Beit Jibrin (Eleutheropolis; Beth Guvrin) northward toward Sha‘ar Hagai (Wadi Ali), and that passes east of Kh. er-Rasm, is marked on the maps of the British Palestine Exploration Survey in the 19th century CE. There are no other major roads, or remains of roads from earlier periods, which were considered worthy of being indicated on the maps. A branch of the above road connected Beit Jibrin and the village of Agur to their surroundings.

Avi Sasson5 Kh. er-Rasm is located in the midst of the British Park of the Jewish National Fund (JNF – Keren Kayemet LeIsrael). During the Late Ottoman and the British Mandate periods it was located on the eastern edge of the agricultural lands of the modern village of Moshav Agur (Ajjur). Despite its proximity to occupied villages at the time we have no references to it during the Ottoman and British Mandate periods (Ben-Arieh 1987:159; Hartmann 1883; Mills 1931:37; Socin 1879). Therefore, our discussion will deal with the nature, development and behavior of the region where Kh. er-Rasm is situated.

Several paths traverse the area of Kh. er-Rasm; however, none of them actually reach the ruins, a fact that reflects the geographical status and function of the site. Some 250 meters north of the remains, on the saddle, there is a path that connects the village of Agur with Nahal Ha’Elah (Wadi es Sumt) and the area of the Elah Valley. This path, known as “Shi’b eth Thughra” (the beard of seal) in Arabic, is marked as a footpath from the Elah Valley, at the foot of the ruins, into the farmlands of Agur, where it became a country road leading to the village.

Sources and Limitations of the Research The absence of a settlement at Kh. er-Rasm in the period under discussion significantly affects the nature of the historical sources that are of use for the present study. The written historical sources that are accessible to the historian and archaeologist, such as travel journals and the writings of pilgrims, are not relevant to the study of the site in modern times. What is useful to this study are the topographic sources, particularly the analysis of historical maps. The primary maps we used are the British Mandate maps with a scale of 1:20,000 (series 14/12). These maps, whose place names have been changed during the course of the years (Bureij, Beit Netef, Beit Netif, Beit Shemesh),

A similar path passed some 250 m south of Kh. er-Rasm. This path also passes along the saddle between Kh. er-Rasm and Horvat Shiqlon. The segment of the path between the Elah Valley and the plantation/cultivated lands of Agur was a footpath, and from the foot of the ruins to the village it became a wider dirt road. Another footpath reached the area of Agur from the north, from the area of Bureij and Zakariya. This path follows a small ravine known as Khallat el Masri (‘the Egyptian Ravine’) that was used as the northern border of the village

This study was conducted with the assistance of the Research Fund of the Ashkelon Academic College. I would like to thank them. 5

185

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

end of the Ottoman period (19th century CE) if not earlier. Based on the widths of the olive trees that grow over the site, and on historical descriptions, it seems that these lands were planted already by the end of the Ottoman period. Victor Guérin, one of the few who surveyed the area of Agur (on his way from Tell es-Safi to Horvat Dichrin, May 21 1863), describes the botanical nature of the area and the dimensions of the olive orchards:

lands. Subsequent to the establishment of the State of Israel, when the road from Kfar Menahem to the Elah Valley was paved, the path that emerges from this road to the south in the direction of Kh. er-Rasm became a dirt road. From then on the road passes some 200 meters east of Kh. er-Rasm. It appears that the paving of the road is connected to the planting of a forest by the JNF in the area south of Kh. er-Rasm (in the area of Kh. Shiqlon and the settlement of Li-on). Over time this path was paved with asphalt and became a major route in the British Park (Figs. 3.24, 3.25).

“… je me remets en marche dans la direction de l’est-sud-est, et je franchis successivement plusieurs collines incultes, où ne croissent que des lentisques… nous parvenons sur un plateau planté d’oliviers…” (Guérin 1869:103).

During the Ottoman Period During the Ottoman period the area of Kh. er-Rasm belonged to the village of Agur that extended along the western foothill of Kh. er-Rasm. In our opinion, some of the large stone walls that surround the site and are marked on the British Mandate topographical maps were built by the

The absence of any reference to vineyards in the area indicates the preference of the residents for cultivating olive trees, as can be seen even later during the British Mandate period.

3.24 Map of the area 1945 (1:20,000) (2007 © All Rights Reserved by the Survey of Israel; Survey of Israel Persmission)

186

Avraham Faust and Adi Erlich

3.25 Map of the area 1950 (1:20,000) (2007 © All Rights Reserved by the Survey of Israel; Survey of Israel Persmission)

3.26 Map of the area 1959 (1:20,000) (2007 © All Rights Reserved by the Survey of Israel; Survey of Israel Persmission)

187

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

Table 3.15: The Demographic Development of the Village of Agur Source

Year

V. Guérin A. Socin M. Hartmann C. Schick British Census British Census

1863 1870 1871 1896 1922 1931

Population in Numbers 800 508 (254 males) 1767 2073 1492

Number of Patriarchal Units 86 120

506

On both sides of the entrance to the columbarium found at the site (Cave F, Chapter 1.3.2), at the level of the present day floor, are the remains of walls built of field stones, without cement, one meter wide and at a height 0.5–1.2 meters. We assume that these walls are the remains of modern shepherding and farming activities.

Reference Guérin 1869 Socin 1879 Hartmann 1883 Schick 1896 Baron 1923 Mills 1931

The cultivation of citrus fruits, as opposed to olives and field crops, requires a large amount of water, and continuous watering throughout the year. So it is likely that some of the water cisterns at the foot of the ruins were used for watering the citrus orchard, as well. To the south of the ruins two more stone walls were built that enclose open areas. It should be noted that the plots around Kh. er-Rasm, in comparison to plots in other areas around Agur, are not divided according to topography, but rather linearly, as some of the fences are located on earlier terraces. Therefore it should be assumed that the majority of the open area of the ruins functioned during the British Mandate period as grazing fields for flocks (Fig. 3.26).

Under the British Mandate During the British Mandate period the area continued to be part of the agricultural lands of Agur. At this time the site made up the eastern border of the agricultural lands that were surrounded by the Wadi e-Sayil ravine (Nahal Hachlil), from the streams leading to Wadi es-Sunt (Nahal Ha’Elah) that passes some 450 meters east of the hilltop. Two large stone walls that border the site are marked on topographical maps of the period as stone fences belonging to one of the olive orchards of Agur.

From the Establishment of the State of Israel Following the establishment of the State of Israel, the road from Kfar Menahem to the Elah Valley was paved. Some 1,500 m west of the Elah Junction, a road to the south – towards Kh. er-Rasm – was paved over the earlier path. The road passes 200 m east of Kh. er-Rasm. It was paved as part of the forestation carried out by the JNF in the area south of Kh. er-Rasm (the area of Horvat Shiqlon and the settlement of Li-on). In time, this road was covered with asphalt and became a major route in the British Park. In the 1965 map the forests are already marked in the area of Horvat Shiqlon and southwest of Li-on. According to the topographical maps from the first years of the State of Israel up until 1967, indicating the remains of orchards, it seems that the character of the cultivated lands did not change, even though it is unclear whether they were still being cultivated. In any case, Arnon Golan’s research indicates that during the first years after the establishment of the State of Israel the new settlers who occupied the deserted Arab villages, such as the residents of Moshav Agur, cultivated the deserted orchards (Golan 1991:243–5) (see also Figs. 3.27, 3.28).

Uses of the Land of Kh. er-Rasm during the British Mandate There is no doubt that the growth of the population of Agur influenced the status and nature of the land on which Kh. er-Rasm is located, as one of the economic resources of the villagers, but most of the available data pertain to the British Mandate period. On the slopes of the hill (excluding the eastern side) there were several plots of land set off by stone fences. To the north of the site three plots, in an area totaling 8,000 square meters (or app. 2 acres), show no signs of any type of cultivation. It appears that these areas were used to grow field crops, mostly grains: wheat, barley, sesame, as is characteristic of the 19th century. West of the site three plots can be identified, two of which have olive orchards on an area of 6,000 square meters (or about 1 1/2 acres) and the third has a citrus orchard on an area of 2,000 square meters (or about half an acre). The easternmost plot covers part of the ruins.

188

Avraham Faust and Adi Erlich

3.27 Area of Kh. er-Rasm aerial photo from 1956 (2007 © All Rights Reserved by the Survey of Israel; Survey of Israel Persmission)

3.28 View from Kh. er-Rasm toward Moshav Agur

189

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

Maps Consulted and Used Map 1878 Palestine Exploration Fund Map, 1878, sheet xx, London. Map 1944 BUREIJ 1:20,000, 14–12, Survey 1931–33, Revised 1942, Printed 1944. Map 1947 BUREIJ 1:20,000, 14–12, Survey 1933–34, Revised 1945, Printed 1947. Map 1948 BUREIJ 1:20,000, 14–12, Survey 1933–34, Revised 1945, Printed 1948. Map 1950 BEIT NETEF 1:20,000, 14–12, Printed 1950. Map 1952 BEIT NETEF 1:20,000, 14–12, Revised 1950, Printed 1952. Map 1955 BEIT SHEMESH 1:20,000, 14–12, Revised 1954, Printed 1955. Map 1956 BEIT SHEMESH 1:20,000, 14–12, Revised 1956, Printed 1956. Map 1959 BEIT SHEMESH 1:20,000, 14–12, Revised 1959, Printed 1959. Map 1961 BEIT SHEMESH 1:20,000, 14–12, Revised 1959, Printed 1961. Map 1965 BEIT SHEMESH 1:20,000, 14–12, Revised 1961, Printed 1965. Map 1967 BEIT SHEMESH 1:20,000, 14–12, Revised 1963, Printed 1967.

190

Part IV: Synthesis

Avraham Faust and Adi Erlich

Khirbet er-Rasm throughout History

1.1). Since nothing of the period was found in-situ, we cannot say much about the history of the site, but it is likely that, just like practically every site in the Shephelah, it was destroyed in the late Iron Age (or at least was abandoned then).

4.0 Introduction The present section aims at analyzing and integrating the data presented in the first three sections of this report, in order to unearth the history of the settlements that existed at Kh. er-Rasm, and the nature of human activity in each of the various stages. Whenever possible, we will attempt to understand the site in relation to its surroundings, both natural and cultural. Since most of the available information is about the settlement at Kh. er-Rasm during the late second century BCE, the analysis will be divided into three chapters. The first (Chapter 4.1) will discuss the nature of settlements that existed at the site prior to the late second century. The second (Chapter 4.2) will examine the history of the site, the nature of the settlement, and the changes that took place during the second century, while the third (Chapter 4.3) will discuss the processes that took place after the destruction in the late second century BCE, and the nature of subsequent human activity at Kh. er-Rasm.

It is clear that there was some human activity at the site during the Persian period, and it is possible that the reoccupation of the site started as early as in the Persian period (for the Early Hellenistic period, see below).1 We shall return to the pre-Hellenistic history of the site below, after more data are presented. 4.1.2 The Hellenistic Period Most of the finds at the site are dated to the Hellenistic period. We have seen above (Sandhaus, Erlich and Faust, Chapter 2.4) that the main assemblage is dated to the late second century, when the Hellenistic settlement was destroyed, and this is, naturally, the period that will receive most attention below. But when was the Hellenistic settlement established, and how long was it occupied?

4.1 Nature of Settlements at Khirbet erRasm Prior to the Later Part of the Second Century BCE

The Hellenistic Phase(s) of Occupation Theoretical Background

4.1.1 Pre-Hellenistic Finds

The date of the assemblage of complete or nearly complete vessels (as discussed above in Chapter 2.4, and see more below) gives us only the date of the destruction, since what is found in it is the de facto refuse, i.e., tools (or vessels, etc.) that became part of the archaeological record although they were still usable (e.g., Schiffer 1987:89; Stevenson 1982:237. The issue will be discussed at length in Chapter 4.2.2 below), and this happens mainly in periods of destruction or abandonment (ibid.).

On the basis of the available evidence, it appears that the Kh. er-Rasm hill was first settled during the Chalcolithic period. This is based on the flints found at the site, as well as of the cornets found during the survey by Yehuda Dagan (Marder, Chapter 3.1). The finds from this period were few and scattered, and the nature of the small settlement that existed in this period is not clear. There seems to be a gap in human activity at the hill for over two millennia, until the Iron Age.

In order to determine the overall period of occupation we have to rely on smaller fragments of vessels and tools that were discarded (or lost) during the site’s occupation. Notably, “discard accounts for the presence of most broken, exhausted, and waste materials in archaeological sites” (e.g., Stevenson 1982:237), and it is the tiny scattered and broken sherds that give us the dates of occupation at the settlement. There are several processes through which such discarded objects (mainly sherds) can reach the archaeological record. For example, pottery has a limited shelf life, and is usually broken after a few years (Rice 1987:296–299; Orton, Tyres and Vince 1993:207–209; Shott 1996, and many references; 20 years is regarded as a very long period). As is well known, pottery is not only breakable, but also cheap, and therefore in most cases broken pots are not repaired

Pottery dated to the Iron Age was found in many parts of the site, including Areas A, B, C, and E. All in all, some 40 pre-Hellenistic sherds (rims only) are discussed by Shai (Chapter 2.1), though it is clear that the actual figure was somewhat higher. The finds include six Iron Age I sherds (mainly from the later phase of this period), 15 Iron II sherds, and additional 12 sherds that could be dated to either the late Iron Age or the Persian period. Six sherds should be dated to the Persian period. The number and distribution of Iron Age sherds indicate that there was some occupation during the Iron Age at the site. Since, however, no “clean” Iron Age loci were discovered, the nature of the activity is not certain. We believe, however, that in addition to the significant quantity of Iron Age sherds that was discovered, there is a possible additional hint that might help us in deciphering the nature of the Iron Age II occupation at the site: architectural features. We shall develop this issue below, and it is sufficient to say that it is possible that the main building was initially built during the Iron Age (architectural phase 1, including 1a, cf., Chapter

Given the sherds, it is not impossible that the Iron Age settlement lingered on to the Persian period, but given the overall settlement history of the Shephelah (which was devastated already in 701; Dagan 2000; Faust 2008, and references) in the late Iron Age this is highly unlikely. Additional evidence for this reconstruction will be discussed below. 1

193

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

but discarded.2 In most cases the sherds are collected, and thrown into a trash pit or the like (secondary refuse). Often these pits are not discovered, and in many instances they are located outside of the settlement (and are therefore usually outside the area of the dig itself). Still, in many cases there are also small pieces that are missed during the “cleanup” and are therefore left over (these sherds are, therefore, primary refuse). Since we are discussing earthen floors, many such sherds are embedded in the floors and are not found until the excavations. These sherds, when analyzed, can give us the period of occupation at a site before its final phase. The latter, in many instances, including at Kh. er-Rasm, is characterized by many finds and a large amount of de facto refuse, and is therefore easily identified.

of pottery by Sandhaus in Chapter 2.2), but this does not mean that a site existed for a long period: After all, even a site which existed for ten years will have pottery whose “life” covers a much longer period – if each pottery form has an average life (for the sake of argument) of 200 years, the possible dates for the pottery unearthed in this theoretical site will cover more than 300 years (since not all the forms will have the same 200-year span). Moreover, while most pottery forms have identifiable peaks of usage during this time span (the battleship curve, see for example Deetz 1967:26–33; Sinopoli 1991:74–75; Orton, Tyres and Vince 1993:190), this says nothing about the actual time in which the pottery was used. Again, a site which existed for ten years will probably contain a few forms whose peak of usage had already passed (or had not yet arrived). The fact that much of the Hellenistic pottery revealed at Kh. er-Rasm covers most of the Hellenistic period does not necessarily mean, therefore, that the site was established before the mid-second century BCE.

It should be noted, however, that due to various processes, both cultural (earth moving during various phases) and natural (rodent and worm burrowing, penetrating roots, etc.), small sherds and artifacts are moved around the site, as well as vertically through different layers (e.g., Schiffer 1985:30; 1987:207–212, 300; see also pp. 121–140; Wood and Johnson 1978; Orton, Tyres and Vince 1993:32; for coins, see Casey 1986:74–79). The finding of “orphan” sherds, i.e., sherds that are not part of complete vessels, is therefore expected (Schiffer 1987:298–302). At any event, the finding of many sherds from an earlier period can be an important tool in dating earlier phases of occupation in sites. In the present case, if there are many sherds (or other items) from earlier phases of the Hellenistic period, we may assume that the site already existed by that time.

However, while most Hellenistic pottery forms cover all or most of this period, there are a few early forms that were no longer in use in the second century BCE. Therefore, though it is very difficult to ascertain earlier Hellenistic phases of occupation on the basis of sherds (in contrast to assemblages – see discussion in Chapter 2.4), this is not impossible. A few forms which belong only to the earlier phases of the period were discovered at Kh. er-Rasm, and those allow us to identify earlier phases of occupation. Those forms include MT1 (three sherds) which is dated to the fifth–third centuries BCE, and CP1 (five sherds) which is dated to the transition from the Persian period to the Hellenistic period and to the Early Hellenistic period.

Another related process is loss of items (“loss refuse”, Schiffer 1987:76–79, and references), for example coins (for factors influencing the loss of coins, see Casey 1986:74–80). Naturally, coins are “usable” and are therefore not typically discarded on purpose. The coins found in the archaeological record are usually a result of loss (with the exception of hoards, which are part of a different process, but this is irrelevant for Kh. er-Rasm, as no hoard was found). Just like discarded objects (and unlike the ceramic assemblages discussed in Chapter 2.4), lost artifacts like coins, present us with evidence regarding the site’s periods of occupation, and not only its last phase.

Those sherds, although few in number, appear to represent the earlier phase of the Hellenistic occupation at the site, and it is likely that many fragments of the other forms (that can be dated to both the third and second centuries) probably represent pots broken already in the third or early second century BCE. The pottery, therefore, indicates human activity at the site already as early as in the first part of the Hellenistic period.3 The earlier date of Hellenistic occupation is even more clearly evident in other types of finds.

The Finds at Kh. er-Rasm In light of the above, what can we say about the length of occupation at Kh. er-Rasm during the Hellenistic period? We shall now examine the finds.

Oil Lamps: Dating the occupation to earlier phases of the Hellenistic period is supported very clearly by the oil lamps uncovered at the site. Oil lamps in general have received a great deal of scholarly attention, and thus allow very precise dating in many periods (e.g., Sussman 2007). The Hellenistic oil lamps of Israel received a detailed treatment recently from Ambar-Armon (2007; see also Chapter 2.3).

Pottery: When examining the pottery unearthed at Kh. erRasm it is clear that it covers the entire Hellenistic period. Obviously, most Hellenistic pottery forms have a long span, which covers at least a century, and usually about two centuries (see the dates attributed to the various types

As stated above, it is possible that the reoccupation of the site began already in the Persian period, but one should remember that the number of finds which is securely dated to this period is minimal, and includes only a few sherds. The number of finds which are dated to the Early Hellenistic period is much larger (and includes not only pottery), and this might hint that the phase we are discussing now started during the early years of the Hellenistic period (or the last years of the Persian period at the earliest). 3

Holes in sherds are sometimes considered as evidence for an attempt to “mend” broken vessels. This, however, does not necessarily indicate the value of the pots, but rather the time available for “menders” and at any event, such sherds are extremely rare at Kh. er-Rasm (and none of the complete vessels was “mended” this way). 2

194

Avraham Faust and Adi Erlich

The examination of the Kh. er-Rasm collection by AmbarArmon (Chapter 2.3) revealed forms which are relatively early in date, and cannot belong to the late second century.

extended periods of time. Hence, for our purposes, it is better to examine what Waner and Safrai (2001) called “revolt hoards”. And indeed, the average age of the coins in the 57 “revolt hoards” they examined was 44.7 years, and the median was 34 years. It should be noted that about half of the “revolt hoards” (44.7%) had only coins from the last 30 years before hoarding (Waner and Safrai 2001:308). Clearly, most of the coins are “late” (and this is true even in regard to other kinds of hoards; e.g., Waner and Safrai 2001:317–319). Earlier coins, one way or the other, went out of circulation (not only as a result of loss, of course, but also because some were collected for “re-minting”, etc.).

A nozzle of an Attic lamp, for example, should be dated from the second half of the fourth century BCE to the first quarter of the third century BCE (or slightly later). And another lamp (of which sherds of five items were unearthed), which is an imitation of the Attic lamp, should be dated to the end of the fourth to third centuries BCE and, at the latest, to the beginning of the second century BCE.4 As the above clearly demonstrates, the lamps indicate that the site was already inhabited long before the late second century, and given the Attic lamp, probably throughout the Hellenistic period.

It is therefore logical to expect that most lost coins that are found in a dig were lost within a generation or two after they were minted. Given the above data on the Kh. er-Rasm assemblage, it appears that the Hellenistic occupation of the site is likely to have started in the late fourth or early third century, as indicated by the oldest coin (Alexander Philip III, 330–320 BCE). The evidence of the coins, therefore, indicates that the site was inhabited continuously from the first years of the Hellenistic period, or at least from the early third century BCE, until the late second century (the time of John Hyrcanus or slightly later).5

Coins: Dating the occupation of the site to the entire Hellenistic period is also supported by the coins found (Barkay, Chapter 3.2). The Kh. er-Rasm assemblage is very small and includes only 17 coins, most of which were found on the surface, and this caveat should be recalled throughout the discussion (which, at any event, is conducted on a very elementary level, and does not include quantities and the relationship to the presumed quantity of minted coins, etc.).

Carbon 14: Occupation throughout the Hellenistic period is also supported by the range of the 14C dates (Boareto, Chapter 3.8). While not accurate enough, it is clear that many of the dates are earlier than the date of the final destruction, and hence they appear to support the view that the site had a relatively long period of occupation during the Hellenistic period.

Five of the 17 coins that were unearthed come from later activity at the site, and the rest are Hellenistic coins. Notably, these 12 coins “cover” a period of about 200 years. In contrast, the other five coins unearthed cover a period of over a millennium, and, moreover, the earliest of these date to about 150 years after the last coin of the Hellenistic assemblage. They clearly belong to a different phase (or phases). This allows us to treat the 12 Hellenistic coins as an assemblage and to discuss it on its own right – as Casey (1986:78) wrote: “the overall distribution of coins through time gives a good general view of activity…” (on the composition of the coins assemblage, see Chapter 4.2).

Summary: The evidence from pottery, oil lamps, coins, and 14 C clearly indicate that the site was inhabited throughout the Hellenistic period, until its abandonment and destruction in the late second century. Most indicative are the oil lamps and coins, which clearly demonstrate that the site was inhabited already during the late fourth century, or the early third century at the latest.

The finds include a coin of Alexander Philip III (330–320 BCE), as well as four Ptolemaic coins (third century BCE), five Seleucid coins, dated to approximately the first decade of the second century (probably 198–187 BCE), one coin which is dated to the time of Antiochus IV Epiphanes (175–164 BCE), and one coin of John Hyrcanus (127–104 BCE).

The abundance of Early Hellenistic finds, when compared with earlier (Iron Age and Persian period) material, as well as its location (below), indicate that the Early Hellenistic activity should not be viewed as a continuation of that of the Persian period, but rather as a new foundation of the first years of the Hellenistic period, or slightly earlier (see more below).

But can we learn from the above about the site’s occupation? Couldn’t these coins have been lost in the late second century?

Before discussing the nature of settlement at Kh. er-Rasm in the Early Hellenistic period (Chapter 4.1.4), we would like to discuss additional evidence for activity at the site during this time.

Several studies have attempted to calculate the “shelf life” of coins, mainly on the basis of an analysis of the finds in hoards of various types (e.g., Waner and Safrai 2001). In order to learn about the usage of coins at any given moment, it is better to use hoards that were collected in haste, and are thus more representative of the situation at a certain moment, rather than hoards collected over

The Hellenistic coins that were found seem generally to predate the late second century, and we attribute the “scarcity” of mid-late second century coins at Kh. er-Rasm to the fact that the inhabitants probably took their coins with them when they left, and those that were found represent mainly those that were lost over the years. This is why the (limited) coin assemblage at Kh. er-Rasm appears early – it represents only the “lost” coins, cf., Schiffer 1987:76–79; see also Casey 1986:74–79, 80; see more in Chapter 4.2 below). 5

Note that at Tel Dor they were dated to the middle of the second century BCE (see Ambar-Armon, Chapter 2.3). 4

195

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

4.1.3 Additional Finds from the Early Hellenistic Period: Subterranean Complexes

We know that the cave system predates the late second century, because it collapsed during the Hellenistic period, and the collapse changed the character and plan of the main building (Chapter 1.1; see more below). This happened before the late second century (when the structure was destroyed), and we tend to date the collapse to the first half of the second century (although this is based mainly on circumstantial evidence, and it could happened shortly before or even after that date).

Kh. er-Rasm is surrounded with man-made caves (Chapter 1.3), which should be attributed mainly to the Early Hellenistic period. One system of caves runs below the main buildings, and others were hewn to its north, west and south. We did not identify any cave to its east, though this is probably incidental. Quarrying caves by penetrating the narrow nari layer and cutting into the soft white kirton rock, is a phenomenon well documented all-over the Judean Shephelah (Dagan 2006:18*–20*). The local soft chalk (kirton), allowed for easy hewing and while the initial purpose of the hewing was to obtain building stones, it also resulted in stable underground rooms and passages, which were used for various purposes (Kloner 2003:4). The phenomenon seems to have started in the Hellenistic period when the local stonemasons became specialized (Dagan 2000:231; 2006:19*). The foremost representation of this phenomenon is Maresha (Kloner 1993:951–953; Kloner 2001:117–125; fig. 3.5; Kloner 2003:18–21).

The other caves in the site should also be attributed to the Hellenistic period, although in general caves in the Judean Shephelah appear throughout the Hellenistic, Roman, Byzantine and Early Arab periods (Dagan 1982; 2006:18*–20*). Cave F: This cave is also known as the columbarium, since it served as one during part of the time. It went through several stages of usage, and in this respect resembles caves from Maresha (Kloner 2003, passim). The various stages attest to a long period of use. The cave was originally cut as a quarry, most probably in the first part of the Hellenistic period. This quarry joins other quarries (Cave G) in the site. Interestingly, at Kh. erRasm we observe the same abnormality which is prominent in Maresha, regarding the ratio between hewn caves and upper structures. In Maresha, over 50% of the total surface of the city was cut beneath the surface, for various purposes which correspond with the Kh. er-Rasm caves’ function (Kloner 2003:152). The Maresha caves originally served for quarries, and some were quarried further and expanded through the years. It is clear that the volume of the hewn soft limestone is enormous, and this is the case in Kh. er-Rasm, if indeed the caves around and below the main building were quarried at the same time. We have no evidence to attribute this large-scale quarrying to any economic activity, such as selling stones or manufacturing lime for building purposes. It is more likely that the stones were cut for local use in the main building. Perhaps the maintenance of the brittle kirton walls required new building material every few years and as the kirton was available and quarrying specialized and cheap, the caves were cut throughout the whole period.

Cave H: At Kh. er-Rasm, subterranean system H is probably the most similar to the Maresha phenomenon, and although it was not excavated (see Chapter 1.3), some of its spaces were surveyed. It seems to be composed of several connected spaces, and it had more than one entrance. This system seems to cross beneath the main building from northwest to southeast, probably with one entrance in the courtyard (Square G50, L.809) and perhaps two more at each end close to the outer walls of the building (see Fig. 1.89). This layout is most typical of the Maresha caves, for example cave 61 which lies under three or more residences, and was a result of changes and openings between spaces that occurred throughout the years (Kloner and Arbel 1998; Kloner et al. 1998; Kloner 2001: fig. 3.9). Subterranean H relates to the main building in its layout, axis and entrances, which either flank the building or are located in its courtyard. The cave seems to postdate the original building, which we attribute to a stage earlier than the Hellenistic period (below). We suppose that the original nari-built building was dug-under during the Early Hellenistic period, and the rock material quarried out of subterranean system H and other caves around the site supplied the kirton stones for the walls of phase 2b (see Chapter 1.1). The use of rectangular brick-like stones made of kirton is typical of Hellenistic Maresha, as observed already by Bliss and Macalister (1902:53; see also Dagan 2006:20*). Walls made of kirton bricks or walls that combined nari and kirton, as in Kh. er-Rasm, were found in all the upper structures in Maresha, especially in the residences (Kloner et al. 1998; Kloner, Finkielsztejn and Arbel 1998; Kloner, Erlich and Whetstone 2000). The original use of the caves in the site was for quarrying masonry stones (Chapter 1.3), and this was done in order to refurbish the old nari building and to create new divisions and an inner plan.

The second stage of Cave F is a cistern, as is evident from the plastered fissure traced in the cave. The shape of the cave is very similar to Maresha cisterns, which like Cave F are usually bell-shaped, with a spiral staircase supported by a pillar and protected by a parapet (Kloner 2001–2002:465, figs. 1–9; Kloner 2001: figs. 3.9, 3.10). Since kirton is a relatively water-proof rock, cisterns in Maresha during the Hellenistic period were never plastered, except for fissures (Kloner 1996:24; Kloner 2003:52–53), and the plastered fissure in Cave F corresponds with that phenomenon. In the last stage of its use in antiquity, Cave F served as a columbarium, like many similar installations in Hellenistic Maresha (Kloner 2001:122–124; Kloner 2003:42–45). Hewn columbaria are common in the Judean Shephelah

196

Avraham Faust and Adi Erlich

(Dagan 2006:20*). Columbaria started playing a role in the economy of the region and nearby areas as early as the Hellenistic period, as indicated by over 85 columbaria in the Maresha caves (Kloner 2001:122–125) and in a Hellenistic-built columbarium from Ziqim (Zissu and Rokach 1999). Certain columbaria from Maresha were hewn as such and thereafter changed their function several times (for example the columbarium in complex 21, Kloner 2003:42–45). Berlin (1997b:8) attributes the development of raising doves in Maresha to the Ptolemaic reign, and claims that when the Seleucids took over, many columbaria went out of use. This is not the case in the Kh. er-Rasm columbarium, as it was converted into a columbarium at a relatively later stage.

change in the needs of water supply to the site, or the creation of an alternative one. Perhaps cistern J (Chapter 1.3) replaced it in a later period, or perhaps the demand for water declined, due to a decline in the population at the site, possibly because of the changes in the nature of the site at some point in the Hellenistic period, after the collapse of Cave system H. While not certain, we find the following reconstruction to be the most plausible: Cave F served as a quarry in the earliest phases of the Hellenistic period, when the main building was rebuilt (for the earlier phases, see more below), at the same time as Cave H was hewn (as well as other caves, see below). After quarrying ceased, probably because the refurbishing of the house was completed, the cave was turned into a cistern. This change took place at a relatively early stage, most probably in the early part of the third century (cf., the dating of the first stages of the building during the Hellenistic period). Later, probably during the second century, following the collapse of Cave system H and parts of the main building, the use of the cave was changed again and it was turned into a columbarium, which probably supplemented the income of the site’s owners.

Just as in Maresha, where installations changed their designation over a span of 200 years, all three stages of Cave F (a quarry, a cistern and a dovecote) probably took place in the Hellenistic period. If we try to attribute each stage of Cave F to the phases of activity in the main building, then the quarry in Cave F (the initial use of the cave) was probably cut in the same period as subterranean system H, i.e., in the earlier phases of the Hellenistic period, and it also supplied masonry material to the building (phase 2b). The columbarium, the last stage of activity in the cave, might have been related to the main building in its last phase of use – that of the Late Hellenistic period occupation, or slightly earlier (i.e., until phase 3). Its use in the Early Roman period cannot be entirely ruled-out,6 but this is less likely, as the presence in the main building during the Early Roman period is very limited and due to the ephemeral nature of occupation in the site at this period (see more below).

It is most likely that additional caves should be dated to the same period: Cistern J: The opening and inner shape of the cave differ from the typical Hellenistic cisterns in Maresha, which are usually bell-shaped with a spiral staircase. The water troughs near the opening of cistern J could have been in use by shepherds when the site was uninhabited. However, it is more likely that cistern J supplied water to the main building already in the Hellenistic period, or perhaps even earlier, and it was later adapted to additional usages.

Hence we prefer to attribute the columbarium in Cave F to one of the Hellenistic phases, most probably to the last stage of the Hellenistic period, when it provided an additional income to the owner. Accordingly, the quarry and the cistern should therefore be dated to the earlier phases of the building, when it was still used in its original plan (and to the time of Cave complex H).

Cave G: This is yet another quarry just west of the main building. It was probably also quarried during the same stage at the early part of the Hellenistic period when quarrying in the soft kirton became a specialization, although a later date cannot be ruled out. Cave G does not seem to have further stages of usage during the Hellenistic period. The hideout burrow which cuts and connects Cave G with Cave D proves that both caves predate the tunnel. As the tunnel is typical of the Early Roman period (see below), it is logical to attribute both caves to the Hellenistic period, though we cannot comment on its date within this period. It seems, therefore, that some of the caves at Kh. er-Rasm were cut as quarries in the Early Hellenistic period, and supplied the kirton material for the walls of phase 2b of the main building.

The second phase of use of the cave – as a cistern – is also very interesting. We cannot date it with any precision, of course, but it is likely that it served in this capacity during most of the Early Hellenistic period, from the time when the quarrying ceased; this phase is probably overlapping, partially at least, with phase 2c in the main building. Kh. er-Rasm is situated far from a natural source of water: the Nahal Ha’Elah is dry most of the year, and Kh. er-Rasm is quite far from it in any case. During the Hellenistic period the stonecutters took advantage of the relatively waterproof nature of the soft kirton and quarried many water cisterns, which are bell-shaped and have a spiral staircase (Kloner 1991:84; Kloner 2001: fig. 3.10). It is possible that the conversion from a reservoir to a dovecote followed a

4.1.4 The Nature of the Site during the First Stages of the Hellenistic Period It is difficult to reconstruct the nature of the site during most of the Hellenistic period. As we have seen above (Chapter 1.1), the site was originally very symmetrical and well built (phase 1; Fig. 4.1; it is likely that this is also how it was used during phases 2a, 2b; Fig. 4.2) but its usage and

Columbaria continued to be in use, and in even greater numbers, throughout the Early Roman period, when they supplied doves for daily consumption and to the Jewish temple (Tepper 1986). 6

197

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

4.1 Phases 0, 1 and 1a of the main building

plan changed at one time, and when it was destroyed in the late second century it was used following a different plan (phases 2c–3; Fig. 4.3; and see mainly Fig. 1.18). We cannot be certain, but we attribute this change to the consequences of a collapse of a system of caves (see Fig. 1.16). It can be suggested that this was a result of an earth-quake, but there is no supporting evidence for that. This collapse prevented the continued use of the structure and the caves system

as they had been used previously. Since the caves below the structure should be dated to the Hellenistic period, when such caves were in use throughout the Shephelah (above, see also Chapters 1.3, 4.1), and as the building was destroyed in the late second century BCE, it is likely that the change in plan took place sometime in the Hellenistic period, probably during the second century BCE. The function of the site after this change and during the site’s

198

Avraham Faust and Adi Erlich

4.2 Phases 2a and 2b of the main building

Zenon papyri it appears that Zenon and his friend, when traveling in the region in the middle of the third century (259–258 BCE), “occasionally encountered what one might call ‘village strongmen’—that is, well-to-do land-owners, living in areas relatively remote from centers of government authority, who were influential enough locally…” (Schwartz 1993b:306). Tscherikower learned of their standing from a number of papyri, and it is clear that such was the situation in Idumaea, in the vicinity of Maresha, where Tscherikower refers to Zaidelos and Kollochutos as “powerful sheikhs” (Tscherikower 1937:51; see also Schwartz 1993b:306).

final stage (for which we have a wealth of information) will be discussed in detail below – but how was it used during the earlier parts of the Hellenistic period? If the site was indeed used in its original plan, than it was quite an impressive structure (the original plan will receive more attention below). The limited evidence we posses precludes any detailed discussion, but we would like to suggest that perhaps the structure was used as a center of an estate. Schwartz (1993b:305–306), following Tscherikower (1937:48–51), suggested that from the

199

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

4.3 Phases 2c and 3 of the main building

While this is only speculation, one could suggest that perhaps the owner of Kh. er-Rasm at this time was one such influential person, who had a type of “estate” in this part of the Shephelah. This might be further supported by the existence of the pyramid-like stone structure, which most probably served as a status marker, if indeed dated to the same period (see Chapter 1.2; see more below). The stepped structure, visible from a distance, would also mark the status of the structure’s owner (one should note, however, that it is more likely that the structure should be dated to the Early Roman period, see more below). Pottery cannot always be

an indicator of status, but we would like to note, in addition to the above, that the only imported pottery unearthed at the site is the Attic lamp which is dated to this early stage. Nothing in the rich assemblage of the late second century (below) was imported to the site from outside the Land of Israel. As speculative as this evidence is, it adds to the above in suggesting that perhaps the Early Hellenistic structure housed such a powerful local “strongman”. This suggestion, however, will remain speculative until more archaeological data allow us to learn more about the social composition of third century Idumaea.

200

Avraham Faust and Adi Erlich

4.4 Oil press weight in Square H49

Economy and Subsistence

is indicative also of the construction of the roofing in the Early Hellenistic period (for the use of olive trees in the construction of building, see Biger and Liphschitz 1999). It is likely, of course, that olive trees were used also in the later phase, and they were simply the trees available for construction (and were therefore chopped down for such purposes when the trees were old, even if the inhabitants of Kh. er-Rasm did not grow them and only bought the wood). Interestingly, on the basis of the historical sources, the Shephelah in the third century was regarded as Egypt’s olive oil supplier (e.g., Tscherikower 1937, and see below), and hence the finds at Kh. er-Rasm are in accordance with this line of evidence.7 The caves underneath the main building and around it served for quarries, cisterns, columbaria and olive presses, similar to Hellenistic Maresha. It is sensible to attribute most of the usage of the caves (except for modern shepherds) to the Early Hellenistic period, and it appears that agriculture was a major component of the site’s economy. In light of the above reconstruction, it is possible that the site was the center of an estate of some sort.

The finding of an out-of-context olive-press weight (Fig. 4.4) seems to indicate that olive oil production was practiced at this stage. The weight resembles olive-press weights from Maresha (Fig. 4.5; Kloner 2003:53–72), and so is the possible olive press installation unearthed at Cave D (Chapter 1.3; note that its date is uncertain). Oil production was significant in the economy of Maresha (Kloner and Sagiv 1993) and throughout the Shephelah (Dagan 2006:22*–23*). This also seems to have been the case at Kh. er-Rasm during this stage. Most of the processing of the olives probably took place in the caves, which were not excavated. They fell into disuse during the Hellenistic period, most probably in the second century, and must therefore date to the earlier phase. In addition, the fact that 94% of the charcoals examined (Liphschitz, Chapter 3.7) were from olive trees seems to indicate that the region was covered with olive trees. While found in the conflagration of the last phase (late second century), we suspect, since the changes between the two Hellenistic phases were only partial, that some of the roof was built already in the Early Hellenistic phase, and hence

7 The use of olive trees for construction does not mean that the inhabitants were engaged in olive oil production. Rather, it shows that olive trees were extremely dominant amongst the region’s trees, and hence the most available for building. The issue will be further discussed below.

201

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

4.5 Oil press in complex 44 from Maresha with weights in situ (after Kloner and Sagiv 2003:61; courtesy of Prof. Amos Kloner and the Israel Antiquities Authority)

4.1.5 The Foundation Date of the Main Building

western slopes of the Hebron hills-country, all located not far from each other (and from Kh. er-Rasm). Among the structures discussed by Mazar are Kh. Tibnah (11 km from Kh. er-Rasm), Deir Baghl (Fig. 4.7; about 16 km from Kh. er-Rasm), and others (Mazar 1982:105–108; Faust 2005a:200). Although not excavated, a few features are common to all those Iron Age structures:8 all are square structures, measuring approximately 30 x 30 m, with a square yard in the middle.

From the above it is quite clear that the main building (main structure) was in use throughout the Hellenistic period, until it suffered from a partial collapse of the subterranean cave system, probably in the second century BCE. From this time until its final destruction before the end of this century the site functioned in a different way (below). But when was the main structure built?

Notably, Hoglund (1992) had attempted to date the structure unearthed at Kh. Abu et-Twein to the Persian period, on the basis of some Persian period pottery that was found at the site, and in light of an architectural typology he developed for the identification of Persian period forts. While this note cannot be used for a very detailed refutation of Hoglund’s claims, it should be stressed that his conclusions are extremely problematic. Most of the pottery found at Kh. Abu et-Twein is Iron Age pottery (Mazar 1982), and while some forms can be dated to the seventh-fifth centuries time period, only a few cannot predate the Persian period. On this basis Mazar concluded that the fort is an Iron Age fort, and that there was some occupation at the site in the Persian (or Neo-Babylonian) period. Since the vast majority of the pottery that was found in the debris belongs to the Iron Age period (note that the pottery group that can be dated to the seventh–fifth centuries should probably be dated also to the Iron Age), it is more than likely that the site was founded then, and the later pottery only shows that there was some use of the building at a later stage. This is reinforced by the fact that the village below the fort existed only in the Iron Age (Mazar 1982). The village was a “by-product” of the fort (Faust 2005a:199–207), and the latter must, therefore, also be dated to the Iron Age. Furthermore, Hoglund’s architectural dating (which should be conducted cautiously in any event) is also problematic. Almost all the clear examples of Persian period forts he mentioned (and which he used to build his typology) are located outside of Judea. All the Judean examples

On the basis of the ceramic and stratigraphical evidence alone, a foundation date in the Early Hellenistic period (or, theoretically, even in the Persian period) fits the data best. But there are problems with this dating which accompanied us throughout the dig.

8

The plan of the structure, as well as the building techniques, are very reminiscent of earlier buildings, a few of which were unearthed not very far from Kh. er-Rasm. We are referring, first and foremost to the building excavated by Amihai Mazar (1982) at Kh. Abu et-Twein (see Fig. 4.6), and interpreted as an Iron Age fort. This is a relatively large building, measuring some 29.5 x 31 m, which was unearthed on the western slopes of the Hebron hills (west of Gush Etzion), some 15 km as the crow flies east of Kh. er-Rasm. Interestingly, Kh. Abu et-Twein is only one of a number of similar structures identified by Mazar on the

202

Avraham Faust and Adi Erlich

4.6 Plan of structure at Kh. Abu et-Twein (after Mazar 1981: 231; courtesy of Prof. Amihai Mazar and the Israel Exploration Society)

203

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

Clearly the plan of Kh. er-Rasm does not seem to fit the Hellenistic period, as it does the Iron Age. Although architectural similarities should be treated with great caution when it comes to dating, such striking similarities should be taken into consideration, and, moreover, one should also recall the Iron Age pottery unearthed at Kh. er-Rasm. So what can we say about the date of the building of the main building? It is difficult to be certain, since no clean Iron Age loci were discovered, but it is possible that the structure was erected already in the Iron Age, and that the few dozen Iron Age sherds, all out of context, belong to this phase. Interestingly, the ramp which was built near the gateway to the building (W.37) does not only add to the impressiveness of the building, connecting it to the above-mentioned Iron Age fortified buildings, but may also reinforce the Iron Age date. A probe in the ramp, which prevented direct access to the building (it was not an agricultural terrace, see Ackerman, Chapter 3.10), unearthed only one datable sherd at the bottom of the section – and the latter was dated to the Iron Age. While far from secure, as the sherd might have reached the ramp when earth was brought there at a later date, we believe that it is most likely that the sherd indicates the dating of the ramp, and this, along with the architectural similarities, suggests that the entire structure was built during the Iron Age.

4.7 Plan of structure at Deir Baghl (after Kochavi 1972: 50; courtesy of the Israel Antiquities Authority)

The size of the building at Kh. Abu et-Twein, as well as the other sites discussed by Mazar, is almost identical to the one at Kh. er-Rasm. Also similar is the fact that we are discussing a square courtyard surrounded by rooms. The ratio of the yard to the built structure is also relatively similar (app. 27.5% at Kh. er-Rasm and app. 20.5% at Kh. Abu et-Twein). Moreover, in both cases the buildings use rows of monoliths in order to support the roof (although there are more rows at Kh. Abu et-Twein than at Kh. erRasm). Extensive use of monoliths in building technique is very common in the Iron Age (e.g., Netzer 1992), but much less so in the Hellenistic period.

The structure was reused for a long period of time in the Hellenistic period, and during this time it was cleaned in order to enable its use. While by no means a simple and straightforward reconstruction, we feel that the available evidence supports it, and this might be the best solution for the data at hand.9 Interestingly, similar processes, by which new settlers “clean” an old building in order to reoccupy it, are known in the archaeological literature. The site (fort) at Kh. ‘Uza, for example, was abandoned in the late Iron Age, and was resettled in the Hellenistic period. The new settlers, however, appeared to have cleaned most of the rooms, and according to Beit- Arieh and Cresson (1991:132): “(T)‌his is probably why very little [Iron Age, A.F. and A.E.] pottery and practically no other objects were found during the excavation of the fort’s rooms, especially on its western side”. A large dump (heap) below the fort was discovered and probed. The dump contained many Iron Age sherds (not restorable) mixed with ash, and Beit- Arieh (2007:53) summarized: “(T)he discovery of the dump with Iron Age pottery near the fortress explains the phenomenon of the empty rooms excavated, mainly in the western part of the fortress. Apparently, during the Hellenistic period that part of the fortress was cleared out and debris dumped down the slope…” (see also Figs. 2.55, 2.56 in Beit-Arieh 2007:54). This was done “when they [the settlers] cleaned the ruins

The striking similarity between Kh. er-Rasm and the Iron Age structures is disturbing, and the dissatisfaction with a Hellenistic dating for the structure is increased when one looks for parallels for such structures in the Hellenistic period. All the isolated Hellenistic structures, mainly farmsteads and fortified farmsteads, are very different in form. These include, for example, the nearby structure at Aderet (Fig. 4.8; Yogev 1982), Tirat Yehuda (Fig. 4.9; Yeivin and Edelstein 1970:57), Mazor (Fig. 4.10; Amit and Zilberbod 1998: fig. 117) and many others. None of these structures is built as a closed structure with a system of rooms around a courtyard, and none has rows of monoliths. In addition, a look at the collection of plans in Tal’s (2006) new volume also does not provide anything which even resembles Kh. er-Rasm. are problematic and speculative at best. At some of the sites not a single Persian period sherd was found (e.g., Deir Baghl; Hoglund 1992:195–196; cf., Kochavi 1972:41), while others are multi-period sites so the dating of the structure, which was only surveyed, is in doubt (e.g., el Qatt; Hoglund 1992:196–197; cf., Kochavi 1972:78). Notably the most important site in Judea is that of Kh. Abu et-Twein, discussed here, as it is the only one that was excavated (Mazar 1982). We are, therefore, quite positive that Mazar’s dating is correct, and so the structure at Kh. Abu et-Twein should be dated to the Iron Age.

This was suggested to us by Yehuda Dagan during one of his visits to the site. Admittedly, we tended to reject this suggestion, until we examined the find spots of the Iron Age pottery (below), which led us to support it. 9

204

Avraham Faust and Adi Erlich

4.8 Plan of structure at Aderet (after Yogev 1982: 82; courtesy of the Israel Antiquities Authority)

of the burned Iron Age fort for rebuilding purposes” (BeitArieh and Cresson 1991:132).10

In light of the available data, we would like to suggest that it is possible that the main building was initially erected during the Iron Age and moreover (and though not without reservations), this might be the best explanation for the data.

Interestingly, the vast majority of the Iron Age pottery found was not unearthed in good archaeological contexts, and was found (in addition to topsoil) mainly in the yard, in unused parts of the building, or just outside it (the area outside the building was not systematically excavated, of course). While this is similar to the pattern at Kh. ‘Uza, and might be taken to support a similar reconstruction at Kh. er-Rasm, one should also remember that if the building was erected during the Hellenistic period the pattern would be the same (i.e., the Iron Age remains would be thrown away when preparing the erection of the building). Still, as we shall see below, the find spots of the pottery are of importance. 10

4.1.6 Khirbet er-Rasm in the Iron Age Reconsidered If the above suggestion is correct, and if Kh. er-Rasm was similar to Kh. Abu et-Twein, then we can use the parallels in order to speculate about the nature of the site in the Iron Age.

205

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

4.9 Plan of structure at Tirat Yehuda (after Yeivin and Edelstein 1970: 57; courtesy of the Israel Antiquities Authority)

4.10 Plan of structure at Mazor (after Amit and Zilberbod 1998: fig. 117; courtesy of the Israel Antiquities Authority)

206

Avraham Faust and Adi Erlich

4.11 Isometric reconstruction of the main building at Kh. er-Rasm

is phase 1 (including 1a) of the building, as discussed in Chapter 1.1.

It was suggested that Kh. Abu et-Twein was a fort, but we believe it is more likely to have been a royal estate. This is not the place for a lengthy discussion, but it appears that Kh. Abu et-Twein and the other structures mentioned above were used as central buildings located within a large agricultural area that was cultivated for the crown (perhaps, partially, by corvee labor; for extended discussion, see Faust 2005a:199–206).

A few minor additions and changes, built also in hard nari stones (phase 2a), as well as the construction of kirton walls and installations (phase 2b) are probably contemporaneous, and took place when the structure was rebuilt, probably in the Early Hellenistic period (though some human activity took place already in the Persian period, and this might indicate the reoccupation phase). This phase is characterized by the massive usage of kirton blocks not only in the inner walls but also for building of kirton walls on top of nari foundations.11 The parallel process of building with kirton is, of course, hewing caves in order to supply the kirton blocks, and it is therefore clear that the caves below the main building should be dated to this phase (in addition to their “generic” dating to the Hellenistic period; see Dagan 2000:9; Kloner 1993). Since the main building ceased to function in its original plan at some point due to the collapse of the caves system below it, it is clear that phase 2b, which includes the building of kirton walls, is indeed earlier than phase 2c (when the structure’s plan changed, and it was only partially used as a consequence of this collapse). This is, of course, only circumstantial evidence, but it is quite powerful, and to this one can add the fact that kirton blocks, even if in smaller quantities, were found also outside the wings that were in intensive use during phase 2c

Interestingly, while most such sites were built in the eighth century, at Kh. er-Rasm we have evidence for earlier occupation, though there is nothing we can say regarding the nature of this early settlement. The only wall that might have belonged to this “pre-main building” phase is W.21 (above). It is likely, therefore, that the earlier Iron Age settlement was very small. 4.1.7 Summary: The Pre-Late Second Century History of the Main Building The earliest human activity at Kh. er-Rasm took place in the Chalcolithic period and the early Iron Age, but those were probably before the erection of the main building, and nothing can be said about the nature of this activity. The main building was originally built as a symmetrical square building (Fig. 4.11), using only nari stones (and with rows of monoliths), probably in the Iron Age II. This

It is possible that the addition of kirton walls was gradual and took place throughout this phase. 11

207

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

(below). It is quite certain, therefore, that during the earlier phases of the Hellenistic period the structure was still used in the symmetrical plan, and was perhaps used by a local “strongman” as his dwelling.

Hellenistic coins span two centuries, from the late fourth century (Alexander Philip III), until the time of John Hyrcanus (127–104 BCE). This is followed by a gap of about 150–170 years, until two coins from the Early Roman period represent the next phase at Kh. er-Rasm (below). Interestingly, while limited in number, the Hellenistic coins generally predate the late second century. We attribute this phenomenon to the fact that at Kh. er-Rasm the inhabitants probably took their coins with them when they left (for the formation processes, see below), and those that were found represent mainly coins that were lost over the years. This is why the (limited) coin assemblage at Kh. er-Rasm appears early – it represents only the “lost” coins (i.e., “loss refuse”; cf. Schiffer 1987:76–79, and references; see also above). Had the site continued to exist, it is likely that more coins from the second half of the second century would have been lost, and therefore discovered by us. It appears that the coins from the final days of the site are simply underrepresented in the archaeological record because they had less time to be lost.

The cave system most probably collapsed in the second century (though an earlier date is not impossible), and this led to a change in the use of the building (phase 2c). Some parts, especially those above the axis of the cave system, ceased to function, and the nature of the building changed. This phase and its termination (phase 3), will be discussed below. We would like to stress that while somewhat different interpretations are possible for the various phases, the relations between them and their respective dating, the above synchronization between the architectural phasing and the other types of finds is by and large the most plausible.

4.2 Khirbet er-Rasm in the Late Second Century BCE

The total lack of Alexander Jannaeus coins, dated to the early first century BCE, is also very instructive. Such coins are usually extremely abundant in sites that existed during the first century BCE, as these coins were common and many were minted (Meshorer 2001:37). As Casey (1986:87) wrote: “(T)he notion that the absence of coins from an archaeological deposit is as important, or more important, as the presence of coins is one that must be stressed. Archaeology is not just about what is found, it is equally about what is not found”. We cautiously conclude that the absence of Alexander Jannaeus’ coins supports the conclusion that the settlement at Kh. er-Rasm ceased at the time of John Hyracanus, and did not continue to exist afterward.

As we have seen, the site went through some changes during the Hellenistic period, probably in the second century BCE. Toward the end of the second century, however, the site was abandoned and destroyed, and many finds were sealed below the thick wall fall. The phase that preceded the destruction (composed of phases 2c and 3; cf., Chapter 1.1) is therefore the best known at Kh. er-Rasm. Before analyzing the finds and discussing the function of the site, we should say a few additional words regarding the dating of the destruction of this phase (see already Chapter 2.4) – the destruction which supplied us with this wealth of information. 4.2.1 Dating the Destruction of the Site

And finally, the dating of the destruction is also in accordance with the 14C interpretation (Chapter 3.9). The latest date from the destruction hint at destruction in the second or first century. Therefore, given the information from the pottery as well as the oil lamps and coins, we are left with a destruction layer in the late second century BCE as the only viable interpretation of all the data.

We have seen that the available evidence regarding the dating of the destruction indicates very clearly that it occurred during the late second century BCE. The most important tool for dating in archaeology is, of course, pottery. The issue was already dealt with in details in Chapter 2.4, and it is sufficient here to state that the only possible date for the large assemblage of complete and almost complete vessels that was found below the destruction is the late second century BCE. Some of the vessels types were not in use after the end of this century, while others did not appear before the late second century BCE, leading us to the above dating.

The Background of the Destruction The exact historical circumstances will be discussed in further detail in Chapter 5.1, but it appears that the site was destroyed in the Hasmonean campaign of the late second century BCE. At this time the Hasmoneans conquered Idumaea and destroyed Maresha and other sites, including Tel Beersheba, Kh. ‘Uza, Arad, and others (see extensive discussion and references in chapter 5.1, below).

The same is true regarding the oil lamps. Those found intact could not have been in use later than the end of the second century BCE.

While there is a debate today regarding the fate of the rural sector in Idumaea, as well as in other regions, following the Hasmonean conquest (e.g., Rappaport 1967; 1993; Kasher 1988; 2002; Shatzman 2003; 2005; Schwartz 1991; 1993a; see also Faust and Erlich 2008, and Chapter 5.1 below), we believe that Kh. er-Rasm (see below for the nature of the

The coins unearthed at the site (mainly on the surface) represent a long time span, from the fourth century to the Medieval period, but the vast majority of them are dated to the Hellenistic period (Barkay, Chapter 3.2). The twelve

208

Avraham Faust and Adi Erlich

site at the time) was abandoned by its inhabitants due to the impending campaign, and was subsequently destroyed, probably by the Hasmoneans. We would now like to discuss the processes of abandonment, destruction and collapse in detail. 4.2.2. Abandonment, Destruction and Collapse: Historical Events and Formation Processes at Khirbet er-Rasm The late second century phase at Kh. er-Rasm is the best preserved period of occupation at the site, and the present chapter aims at discussing the formation processes of this phase, and attempts to reconstruct the actual processes of the site’s abandonment and destruction and its subsequent collapse. Khirbet er-Rasm in the Late Second Century BCE The finds within the last Hellenistic occupational phase at Kh. er-Rasm included dozens of complete vessels that were found sealed under a massive layer of stones, probably originating from a wall fall (Figs. 4.12, 4.13, 4.14). In most places, the pottery was found within an ash layer. Most of the rooms in which the pottery was unearthed were purposely blocked prior to the destruction (Figs. 4.15, 4.16; see also Figs. 1.56, 1.57).

4.13 Storage jar in situ

It appears as if the inhabitants of this phase left the site, leaving most of their daily utensils behind, and the site was subsequently destroyed. But why was the pottery left behind? When did the walls collapse? What can we learn from the location of the pots? Why were the rooms blocked? Those are but a few questions that need to be answered in order to enable us to decipher the processes of abandonment, destruction and collapse at Kh. er-Rasm. Before attempting to answer these questions, we would like to introduce a few concepts that will help us in the discussion.

4.14 Pottery in situ

4.12 Lamp in situ

209

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

4.15 Blockage of entrance to room 4

4.16 Blockage of passage between rooms 13 and 14

210

Avraham Faust and Adi Erlich

The Formation Processes of Abandoned Sites: A Brief Introduction

is missing. The two processes are, of course, interrelated, and both are influenced by many other factors.

Processes of abandonment and collapse of structures, settlements and even regions, have been the subject of a great deal of scholarly research (e.g., Schiffer 1987; see also various papers in Cameron and Tomka 1993; Inomata and Webb 2003; and also Stevenson 1982; Cameron 1990; Ziadeh-Seely 1999; Hegmon, Nelson and Ruth 1998; Nelson and Schachner 2002, and additional references). While it is clear that abandonment is an important component in the processes that form the archaeological record, it does not always receive adequate and explicit attention, with many archaeologists working implicitly under the “Pompeii premise”, or the “disaster movie” “mind set” (see also Binford 1981; Schiffer 1985; CiolekTorrello 1989; for the latter term, see Cameron 1993:3). The significance of understanding abandonment processes in archaeology can be illustrated by the following questions, or assumptions, concerning the distribution of artifacts, as posed by Cameron (1993:3): “Should we assume that artifacts found on room floors were left exactly where they were used? Were they dumped there days or hours before abandonment when normal clean-up processes were relaxed? Were they, instead, cached for later use during an anticipated return? Do they represent trash tossed into an abandoned room years before the settlement was abandoned?” We believe that the detailed archaeological information regarding the abandonment of Kh. er-Rasm during the late second century BCE allow us to decipher the processes which accompanied the destruction, but first we should introduce a few additional concepts that will enable a better discussion.

According to Schiffer (1987:90) there are a number of variables that influence de facto refuse, mainly in case of settlement abandonment. The relevant determinants are: the rate of abandonment (e.g., rapid and unplanned versus slow and planned), means of available transport, season of abandonment, distance to the next settlement, principal activities in the next settlement, size of emigrating population, and whether or not return is anticipated. In addition, variables pertaining to the artifacts themselves, including artifact size and weight, replacement and cost, remnant uselife, and function(s), condition curate probabilities and thus influence de facto refuse deposition (Schiffer 1987:90–91). The “interaction” between the above variables determines the nature of the de facto refuse unearthed at a site (as well as the “curate behavior”, as the same variables influence both). Schiffer (1987:91) adds that “rate of abandonment may be the most important variable, for it influences the magnitude of effects produced by the others”. In his important study, Stevenson (1982) concluded that two major variables of abandonment behavior contribute to the formation of the archaeological record of abandoned sites: the manner in which sites are abandoned (i.e., planned or unplanned) and whether or not return is anticipated (e.g., Stevenson 1982:237). Those two variables are, according to Stevenson (1982:240–241), “expected to be the major factors producing variation in (a) the kinds, quantities, and distribution of de facto refuse left on these sites, and (b) refuse disposal behavior within enclosed living areas”.

De Facto Refuse and Curate Behavior Generally speaking, much of the archaeological record uncovered at abandoned archaeological sites is a result of what is called de facto refuse and curate behavior (e.g., Schiffer 1987:89–98).12 According to Schiffer (1987:89): “(D)e facto refuse consists of the tools, facilities, structures, and other cultural materials that, although still usable (or reusable), are left behind when an activity area is abandoned”. This is the process by which “most usable tools, facilities and other elements… enter the archaeological record” (Stevenson 1982:237). The second variable, curate behavior, “designates the process of removing and transporting still-usable or repairable items from the abandoned activity area for continued use elsewhere”. Curate behavior has an impact on the formation processes of the archaeological record both at the abandoned sites and the sites into which the people immigrate (ibid., 90). It is mainly the former (de facto refuse) which is of relevance for the present discussion, of course, as it refers to the usable artifacts found at the site, while the latter (i.e., curate behavior) is responsible for what

We have already seen that in addition to these major variables (rate and nature of abandonment, and whether return is anticipated or not), there are many other variables (seasons of abandonment, distance to the new site, available transportation, size of population, etc.) that contribute to the formation of the archaeological record (above; see e.g., Schiffer 1987:90–91; Stevenson 1982:241).13 The precise importance of each of the variables changes according to the exact context and the specific circumstances of each case. Since there are various possibilities that influence every variable, the number of possible combinations is almost endless. For example, much discussion was devoted to understanding gradual processes of abandonment, in which some parts of a site are abandoned, while others are still being used – either temporarily until they were abandoned too, or permanently (e.g., Schiffer 1987:89–98; Ziadeh-Seely 1999). This, of course, has an effect not only on the social reality on the site as a whole, which differs greatly from a site whose all

For other aspects of formation processes, e.g., “discard” (which “accounts for the presence of most broken, exhausted, and waste materials in archaeological sites”; e.g., Stevenson 1982:237) and loss, see Chapter 4.1.2. 12

Stevenson believes that for the sites he studied, considerations of the season of movement, means of transport, and distance to next site were similar, and were therefore not of importance for the discussion. 13

211

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

areas were abandoned at the same rate (whether rapidly or not, which is another variable), but also for the use of the abandoned structures, rooms or spaces. If the site was abandoned gradually at an uneven rate (i.e., some parts were abandoned while others were not), abandoned rooms can be used for various purposes, even for trash disposal, by the people who remained at the site, before the settlement’s final abandonment. To make things even more complex, the exact usage of the abandoned spaces is dependent also on whether or not the former occupants or owners (or their close kin) are among those who totally left the site (e.g., Schiffer 1987; Ziadeh-Seely 1999). This is but a brief discussion of one variable, of course, but it makes it clear that a full discussion of all possibilities involves a long and detailed study.

from the outside, i.e., they were abandoned. Blocking of rooms is characteristic of a stage of abandonment, when the inhabitants intend to return to the structure. This is very typical, for example, in cyclic abandonment, when the inhabitants leave the site for a season. They intend to come back, and by blocking the house’s doorways they indicate that it is not abandoned permanently. In reporting the finds from the Early Bronze Age settlement of Numeira, Coogan (1984:80–81), for example, wrote that “(M)ost identifiable doorways from the latest phase of occupation had been deliberately blocked”, and he adds “(A) possible explanation is provided by a modern parallel in the same region; when inhabitants of a house leave for some time, they frequently block their windows and occasionally their doorways with cinderblocks or wood”. Hart (1988:98), after reporting of a similar blocking at Iron Age Ghareh, writes in a similar fashion: “This practice is still observed today when a householder wishes to leave his building for an extended period of time. The doorways will be unblocked on his return”. We suspect that this is the case here too. The inhabitants left the dwelling, but intended to come back (cf., Fig. 4.17).

Luckily, however, not all the information and the possible interplay among the various variables is relevant for every case, and since it seems that the situation at Kh. er-Rasm is quite straightforward, there is no point in discussing all the possible scenarios. We will, therefore, refrain from further discussion of theoretical options, and will now go back to late second century BCE Kh. er-Rasm, in order to examine the process which accompanied its abandonment, destruction and subsequent collapse.

But if it was abandoned intentionally, why were so many artifacts left behind? Some of the rooms were almost filled with pottery, and it is likely that the pottery was left in its original place of use, including on shelves on the walls (below). Usually, it is “expected” that abandoned site will be void of many finds (e.g., Segal, Kletter and Ziffer 2006:4*; Beit-Arieh 1992:111; Riklin 1997:12; Dadon 1997:74; but see Coogan 1984:81).

The Abandonment of Khirbet er-Rasm Schiffer (1987:90) wrote: “…processes of de facto refuse production are still poorly understood”, and we therefore believe that not only will the above discussion permit us a better understanding of the formation processes at Kh. er-Rasm, but the good preservation and detailed information from the latter will allow us to contribute something to the more “general” discussion of formation processes in the archaeological record.

We believe that the nature of the historical circumstances surrounding the abandonment of the site can explain this apparent contradiction. During the Hellenistic period the site was located close to the border between Idumaea and Judea, apparently, on the Idumaean “side” (e.g., Kokkinos 1998:60; Dagan 2000:229, 233; more below, Chapter 5.1). It is probable that the site belonged to Idumea and was destroyed in circumstances similar to those in which Maresha was destroyed. Both were probably destroyed by the Hasmonean king John Hyrcanus in his 111/112 BCE campaign (e.g., Berlin 1997b:28; Kloner 2003:5–6; see also Finkielsztejn 1998; see detailed discussion below, Chapter 5.1). Kh. er-Rasm, however, was a small site, and the inhabitants knew they could not survive the campaign on their own. They most probably left the site, taking only the most valuable objects with them (curate behavior), leaving most of their daily vessels behind (and these became de facto refuse). They blocked the house, hoping to return, and left for a larger settlement, perhaps even Maresha, but more likely for a nearer settlement, perhaps Tel ‘Azekah (for the latter, see Dagan 2000:232–233).

When the main building at Kh. er-Rasm was unearthed, many of the building’s hallways were found blocked (W.24; W.46; W.47; W.48).14 Initially, these blockages were thought to have resulted from changes in the use and the building plan – changes that involved blocking some doorways and opening others. In the course of the excavations, however, we found out that once blocked, the rooms were simply sealed. No alternative entrances were created. A large collection of intact and restorable pottery vessels, along with other finds, was unearthed in the various rooms of the building, sealed under a heavy layer of stones, and in many parts (mainly in the north) also within a layer of ash (Chapter 1.1). Why were the rooms sealed prior to the site’s destruction? Clearly, the rooms could not have been used any longer. The rooms under discussion were, of course, sealed

This scenario is supported by the coins unearthed (Barkay, Chapter 3.2). While limited in number, the coins that were found seem to be generally earlier than the late second century (above). We already attributed this phenomenon to the fact that at Kh. er-Rasm the inhabitants probably took their coins with them when they left (similar to curating),

We should note that those blockings are different from blocking W.49 (in W.15). In this case the blocking is well built and seems to have been built only to the level of the nearby installation, and probably in conjunction with it (note that an oil lamp was found, in situ, on top of this blocking). It also seems to have been part of gradual changes in the use of the building. The other blockings seems to have been done in haste and moreover, completely prevent any option of using the room. 14

212

Avraham Faust and Adi Erlich

4.17 Blocked house from Ethiopia

and those that were found represent mainly coins that were lost over the years. This is why the (limited) coin assemblage at Kh. er-Rasm appears early – it represents only the “lost” coins (i.e., “loss refuse”; cf. Schiffer 1987:76–79, and references), and is representative of the entire Hellenistic occupation at Kh. er-Rasm (the lost coins are similar in date to the “discarded” pottery sherds discussed above).

the inhabitants’ hope of return (which did not necessitate taking all the artifacts). Clearly, the occupants took all their valuables with them, and this explains the scarcity of coins at the site, but they left all the cheap items (e.g., pottery15) in place. As we will see below, this observation is also very important for the analysis of the use of space within the structure. Interestingly, Stevenson’s (1982) observation that the rate and nature of the abandonment as well as the intention of the occupants to come back (or not) appear to have been the major factors that influence the formation process of the archaeological record of abandoned sites finds support at Kh. er-Rasm. Those variables were indeed responsible for the formation processes of Kh. er-Rasm in the late second century. Other factors, such as the season of abandonment, the principal activities in the next settlement (where the occupants seek refuge), were not of any importance in our case. Moreover, the importance of the above major factors even made variables such as the distance to the next settlement and the means of available transport (which could greatly influence what is taken away and what is not)

Clearly, in the face of an approaching army, the abandonment of the site was relatively rapid, and was carried out throughout the site at the same rate. The inhabitants intended to come back, and therefore blocked the doorways. This is quite similar to the abandonment of Bullion Creek, as discussed by Stevenson (1982); the abandonment was rapid, with people intending to return. The blocking with nari stones, rough and unworked, is an indication of the hastiness of the enterprise. The same can be learned from the fact that few of the rooms were not blocked (although it is possible that the blockings of those rooms were simply broken into). As far as the de facto refuse is concerned, we are facing an interesting situation. Most of the vessels were left behind, probably both because of the rapid abandonment (which did not allow time to take all the items) and because of

It should be noted that pottery, probably of all kinds, was probably a very cheap commodity in antiquity (e.g., Vickers and Gill 1994; see also Faust 2005a:131–141, and references).

15

213

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

from the north (where the road is located, see Chapter 4.3 below). It is possible that Kh. er-Rasm was burned down after the conquest of Tel ‘Azekah, when the conquering army marched southward. The structure was destroyed, and most of the pottery that was left behind was broken and smashed during the fire and the subsequent collapse of the roof and of some of the walls (though the collapse continued later, see below). The fact that not a single late sherd was unearthed below the wall fall, in what was defined as occupational debris,16 indicates that the destruction and conflagration took place not too long after the abandonment, and we believe it is most likely that the time gap was relatively brief (we shall return to this issue below). That the major part of the collapse took place during a short period of time is also supported by the fact that while some of the vessels were found intact (or complete) within the ash layer below the wall fall, some were reconstructed from sherds that were found both in the ash layer and within the wall fall (e.g., Fig. 4.19).17 We attribute those to vessels that were probably left on shelves along the walls and broke during the collapse. Had the collapse been a long process, covering decades, it is unlikely that many such vessels would have been reconstructed within the rooms. This further supports our conclusion that the artifacts were left at their original place of use (see more below). 4.18 Ash layer in Area E (Squares E49, F49)

Naturally, the destruction of the site could be attributed to an accident, but this is less likely due to the blocking of the rooms and the finding of the arrowhead and balistra stones, as well as the relatively short time span between the abandonment and the destruction. While the scenario above may seem overly dramatic, we feel it is also the most likely one.18

relatively less important to the present context (cf., Schiffer 1987:90–91). The Destruction of Khirbet er-Rasm Since a large part of the building was burned down (Fig. 4.18, see also Fig. 1.38), and an arrowhead (NennerSoriano, Chapter 3.4.1) along with a few balistra stones (Peshin, Chapter 3.3) were uncovered within the destruction layer, it appears that following the abandonment, either immediately or shortly afterwards, the site was destroyed.

Final Collapse After a site is abandoned, there are various “forces” which are set into motion and continue to change it (e.g., Schiffer 1987:143–156, 163–180, 199–234).

As stated above, it is more than likely that the site was abandoned due to the impending Hasmonean threat, and the inhabitants left the site for a more secure place – perhaps the settlement or fort at Tel ‘Azekah. The destruction of the site can therefore be attributed to the Hasmoneans. It is possible that a group of warriors came across the building and simply burned it down on their way, perhaps without even entering all its rooms. This suggestion is further strengthened when one remembers that while the northern wing was totally burned down, this is not the case in the southwestern section of the building, where the fire was partial (see more below). This might suggest that the troops did not stay at the place for any period of time, otherwise, we would expect both wings to be burned with the same intensity. Interestingly, the only John Hyrcanus coin unearthed at the site was found just outside the building, on its northwestern corner. We may speculate that perhaps one of the attackers lost it while approaching the building

As far as Kh. er-Rasm is concerned, it is likely that the roof collapsed already during the fire, at least in the northern wing where the fire seems to have been very extensive. Notably, a few sherds can always appear out of context, due to various depositional processes (e.g., Schiffer 1987:207–212, 300, see also pp. 121–140). That this is not the case is probably a result of the massive wall fall (e.g., Figs. 1.31, 1.38) which made such penetration somewhat more difficult and probably reduced the number of later sherds in the rooms of the main building. 17 For example, a cooking pot in Square E49 (room 16) was composed of sherds unearthed in L.815, L.822, and L.825, covering the kirton wall fall and debris below it. An additional example is a cooking pot whose sherds were found in L.818 (below the wall fall) and L.815 (in the kirton wall fall) – in this case, there was also a distance of some 8 m between the sherds which were also found in two adjacent rooms. The latter example raises more questions regarding the formation processes, but one has to remember that most vessels were found complete, within one locus (see also Chapter 2.4). 18 We should note that a similar situation, in which the ash layer was identified in many places but not everywhere, was observed in areas 61 and 930 at Maresha (Kloner et al. 1998; Kloner, Erlich, and Whetstone 2000). 16

214

Avraham Faust and Adi Erlich

the use of space within the structure. Analysis of activity areas and the use of space is a common archaeological procedure (e.g., Kent 1990). A common criticism of such analysis results from the fact that many archaeologists are implicitly assuming that the artifacts found represent the activity that was carried out within the same space (see the debate over the “Pompeii premise”, Binford 1981; Schiffer 1985; Ciolek-Torrello 1989). The finds, however, could have reached the places in which they were discovered in various ways, not necessarily reflecting the activity which took place in them. For example, small sherds might be found as a result of a long process of raising floors throughout generations, and finding them embedded within a floor does not indicate the usage of the space, since this could have changed many times during the structure’s existence – and the finds in the floors reflect all the various stages. Moreover, at times, pottery, even complete vessels, could have been thrown into a room after it was abandoned and it turned into an area of trash disposal (secondary refuse). In addition, small sherds can also arrive at their place through post-depositional processes (e.g., Schiffer 1987), and hence be misleading when we attempt to understand the nature of the room or structure. To sum up, various processes can lead to the finding of pottery which pre-dates and post-dates the time of the floor under discussion, and hence can be misleading when we attempt to reconstruct the use and function of the discussed rooms. One should therefore not assume that the finds can be used to reconstruct the activity that originally took place in the spaces where they were discovered within the structures.

4.19 Storage jar in Square H49

However, in some instances it is possible to distinguish a clear, short-lived phase of occupation, in which we can assume that the finds were left in their original place of use (below). In such cases, the distribution of the finds that can safely be attributed to this phase can indeed indicate the use of the space at this stage. While caution is in order, we believe that the data from late second century BCE Kh. er-Rasm allow us to reconstruct the use of spaces within the site on the eve of its destruction and to rule out the above “objections”.

Here, the ash layer below the wall fall (and mixed with its lower part, though large pieces of charcoal were sometimes found throughout the wall fall debris) covered the entire area, and a large amount of charcoal was collected (e.g., in room 13 [L.132, L.160, L.162), room 14 [L.814, L.821, L.823] and many others). Interestingly, an analysis of the sediments (using Fourier Transform Infrared Analysis – FTIR) from such a locus (L.824) provided a spectrum indicating a significant presence of charcoals (Boaretto, Chapter 3.8; note that such remains were not found in other places, e.g., in topsoil and below the wall fall in the western wing). It appears that the southern and western wings were burned less systematically, and it is possible that the collapse in this area was a slightly longer process.

Much of the finds at Kh. er-Rasm is composed of complete and intact vessels which were left when the site was abandoned (de facto refuse), some of these were probably even left on shelves along the walls. The site was then destroyed, and the vessels were sealed below a heavy wall fall. Admittedly, the finds include also sherds, which couldn’t have been restored. We cannot know whether this is a result of the fact that no baulks were removed, and hence the rest of the pot was not excavated, or that we are discussing orphan sherds, which arrived at the loci through various process (e.g., sherds that remained within the floors and represent past activities, or sherds that reached the corpus through other, post-depositional processes, e.g., worm burrows, etc.; see above). Below, however, we will not analyze sherds – only complete or almost complete

As far as the walls are concerned, it is likely that some of them collapsed during the fire, but others deteriorated and fell gradually, following a process of decay, and it is even possible that the process continued during the Early Roman period (the issue will be discussed further below). Khirbet er-Rasm and the “Pompeii Premise” or Can we Study Activity Areas? The above observation is very important for the analysis of

215

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

vessels,19 hence ignoring the orphan sherds of whatever origins. The complete vessels were probably left at their original place of use when the site was abandoned in haste, and hence can be used to infer activity areas. As far as post-abandonment trash is concerned, due to the destruction of the site immediately after the abandonment, the remains were sealed below a massive wall fall, hence preventing any trash from being mixed with the “clean” late second century assemblage. It is also important to note that most of the rooms or spaces which functioned in the late second century were excavated, at least partially, thereby strengthening the plausibility of the analysis (the unexcavated areas in the main building were mostly not used as rooms at this time).

stones, and prevented the addition of later artifacts, and even sherds, into these loci. Other parts of the wall had collapsed later and the overall process seems to have lasted until the Early Roman period (Chapter 4.3). The study of Kh. er-Rasm adds to our understanding of the processes of abandonment. Not only is the site an additional example of the consequences of abandonment and destruction, but it can exemplify the circumstances which accompany abandonment. As can be seen at Kh. er-Rasm, the fact that the abandonment was rapid does not necessarily make it unplanned, and the inhabitants had enough time to block the doorways. The analysis at Kh. erRasm also supports Stevenson’s suggestion that the rate and nature of the abandonment and the question of whether the occupants plan to return are the most important variables in determining the formation process of the archaeological record of an abandoned site.

Clearly, the situation at Kh. er-Rasm makes it legitimate to use the finds of complete vessels in order to learn about activity areas within the site. We can learn about the use of space within the various rooms by examining the finds within them, in combination with the architectural features that characterize them (e.g., various papers in Kent 1990; see also Hillier and Hanson 1984). A broad examination of the nature of the various rooms and of the structure as a whole can therefore teach us how and for what purposes it might have been used (below).

It should also be noted that the above discussion also makes it clear that as far as Kh. er-Rasm is concerned, one can use the finding of complete or almost complete vessels in order to learn about activity areas and the use of space within the main building during its last phase of usage in the late second century BCE, and this is the next issue we would like to discuss.

Summary

4.2.3 The Use of Space at Khirbet er-Rasm: The Function of the Main Building during the Late Second Century BCE

The settlers at Kh. er-Rasm in the late second century BCE abandoned the site in relative haste, prior to the arrival of the Hasmonean forces (phase 3, Chapter 1.1). Due to the rapid and unplanned abandonment they took their valuables with them, but left practically everything else behind. The low value of the vessels might have prevented their being carried away from the site even in other circumstances, but appear to have made such removal out of the question under the circumstances in which the site was abandoned. Despite the haste, the inhabitants, who hoped to return to the site, found the time to block most of the doorways, and by doing so marked the site as not being deserted and ownerless, and indicated that the abandonment was temporary and that the owners intended to come back. The blocking also prevented stray animals from entering the building. The temporary nature of the abandonment probably made it easier for the site’s occupants to leave behind most of their daily possessions.

The main building, it is quite clear as we have seen, was used in the second century BCE. The second century BCE level at Kh. er-Rasm is the best preserved period in the history of the site, and this preservation (above) presents us with a unique opportunity to analyze life at the time. At this time the structure was not used in its original form,20 and while it was still surrounded by walls from all sides, it is quite clear that only the northern, western and part of the southern wings were used for habitation. The eastern and southeastern part of the building as well as part of the western wing, though still standing (at least partially), were used for other purposes, and should, for all practical purposes, be viewed as part of the yard (Fig. 1.18). Khirbet er-Rasm in the Late Second Century BCE

As happens in many cases, however, the occupants never came back, and the vessels turned into de facto refuse.

What kind of a site was Kh. er-Rasm, before it was destroyed? How did it function?

Soon after the abandonment of the site, the structure was burned down, probably by Hasmonean forces. The destruction of the site involved burning it, which caused a total conflagration in the northern wing, and the collapse of the roof and much of the walls – a collapse that sealed the vessels that were abandoned under a massive layer of

The building, an isolated structure built in the countryside, could have been a farmstead, an estate, a fort, an inn, or even a combination of some of these (see more below). How can we identify the use of the building? The fact that the finds from the last phase at Kh. er-Rasm were found in a destruction layer and included many complete and restorable vessels, enables us, as we have

Because no baulk was removed, and no room was excavated in its entirety, we use all complete or nearly complete vessels in the analysis (see below for a more detailed explanation; see also Chapter 2.4), since it is likely that the missing parts were also left in the room, but were not excavated. 19

20

216

This corresponds with phase 2c (Chapter 1.1).

Avraham Faust and Adi Erlich

seen, to analyze the use of space at the site during this phase. By examining the use of space within the structure we may get a good idea of the function of each of the rooms immediately prior to the abandonment. Still, one must practice caution when attempting to move beyond an analysis of the use of space within the rooms, and to discuss the function of the structure as a whole – a farmstead, an estate, a fort or an inn. After all, in some cases it could be very difficult to differentiate between the types of structures mentioned above, and only a comparative study can put any interpretation in context (cf., Faust 2005b:116). Furthermore, any interpretation of the structure’s function should take into account the historical and social background of the time. For example, were there estates in the society under discussion? As of now, however, no attempts have been made to study the local society and, moreover, no other buildings were published in a manner that enables even an analysis of the use of space within them. This is the situation not only regarding Idumaea or nearby regions, but for most of the country. It is a major setback that must be acknowledged. Our enterprise is therefore only a preliminary attempt to learn about the structure’s function and through this, perhaps to begin to study the local society.

Several rooms in the southwestern part of the building were also excavated almost in their entirety. In room 4, most of which was excavated, one of each of the following was found: an oil lamp, a jug, a bottle, along with four bowls. Similarly, in room 5, a bowl and a plate were unearthed, along with a bottle. In the small corridor leading to this room (space 6), one lamp, one juglet, and one storage jar were found, along with two bowls and a plate. Just north of the corridor (room [or rather, space] 9), inside a small installation, two cooking pots and a bottle were found; a bowl was found just outside this installation. Other rooms were excavated to a limited extent only, hence not allowing any significant patterns to be identified.23 Generally speaking, however, it appears as if a bottle, an oil lamp and at least one bowl/plate were found in almost every room. In the western/southern wing, most spaces had an assemblage which is similar to this (though with more than one bowl), while in the northern wing the assemblage was much larger (though it usually included these vessels too). The majority of cooking pots, which are generally less common in the assemblage, were concentrated in one room in the northern wing (room 15). In another room (14) in this wing a large concentration of bowls/plates and bottles was found. While almost every room in the building had at least one or two bowls/plates (above), the situation in the northern wing clearly differs. In room 15 (in which many cooking pots were found) no bowls/plates were found, while in room 14, four bowls / plates (and a krater) were unearthed. It is worthwhile mentioning that room 15, which contains many cooking pots, is not adjacent to room 13, which contains many tabun fragments,24 and the two are separated by room 14.

Due to the good preservation at Kh. er-Rasm we believe that such an attempt is worthwhile. Analysis of the Finds An examination of the findings clearly indicates that they are not scattered evenly across the building, and significant patterns in the distribution of the various artifacts and materials can be discerned.

It is also worth mentioning that the only two cooking pots to have been discovered in the western wing, were found within an installation, and with no evidence for a tabun or anything else that can be connected with cooking. A tabun was unearthed on the southeastern room of the main building, with hardly any associated finds. The limited number of storage jars unearthed may indicate that storage was limited at the site (cf., Sandhaus, Chapter 2.2), but it is also possible that the storage room has not been discovered yet.

Pottery (Fig. 4.20): The following is based on an analysis of all the complete vessels found, along with those restored to a great extent, having either full profile or full rim. In rare cases, when most of a pot was found, we referred to it even if the rim was missing.21 This is because no rooms were excavated in their entirety, and no baulks were removed; hence finding most of a vessel should be sufficient to prove its existence at the time discussed, as we cannot always expect to find all parts of it (see also Chapter 2.4). It appears that most of the pottery unearthed was found in the northern wing of the building. In one of the rooms, about half of which was excavated (room 15), six cooking vessels (five cooking pots and one casserole) were found, along with a storage-jar, two oil lamps,22 one juglet and one bottle (unguentarium). In an adjacent room (room 14), excavated almost in its entirety, three bowls, one plate and one krater were found, along with three bottles, one cooking pot and one table amphora. Still, in an inner room in this section (room 13), only a large decanter was found, along with many tabun fragments (not in situ).

Coins and Metal Objects: All the coins found in the excavations (only about ten coins), i.e., which were not surface finds, were found in the northern part of the building (including the northern part of the eastern wing – the latter functioning as part of the courtyard). In addition, with the exception of two cosmetic (kohl) sticks (to be discussed below), all the metal objects were also found in the northern wing. In some cases pots were found in other spaces, but when the finds were not related to any other finds, and the architecture was not clear enough, we refrained from discussing it in this Chapter 24 It appears as if the tabun (of whatever type) was originally on the roof, and this is the reason why while large fragments were found within the wall fall, none were found intact on the floor or in an occupational debris (for hearths on second storeys, see for example Kramer 1979:147–148, and additional references; see also Kramer 1982; David and Kramer 2001:274). 23

Débora Sandhaus helped us with the analysis, and while working on the pottery, added a number of complete vessels to our list (see also Chapter 2.4). 22 One of them was unearthed on top of the wall fall, in a locus that was still defined as topsoil (L.802). 21

217

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

4.20 Map of distribution of complete and nearly complete vessels

Grinding Stones: Two grinding stones were found in the eastern part of the building, which functioned as a courtyard. One grinding stone was found in the western wing, near the surface, and one was found in room 15 (the same room in which 6 cooking vessels were unearthed).

every house.25 Still, not a single loom weight was found at Kh. er-Rasm.26 This, of course, could be accidental, due to the fact that the site was not totally excavated. It is possible that in another (yet unexcavated) room a large number of loom weights will be found in the future. Still, in the majority of the building no loom weights nor spindle whorls were found, and we believe that the abundance of such finds at other sites, e.g., at the farmsteads of Mazor (Amit and Zilberbod 1998), Tirat Yehuda (Yeivin and Edelstein 1970:61), in the village of Horvat ‘Etri (Zissu and Ganor 2002:18),27 as well as the nearby center of Maresha (Orit

The above shows clear signs of differentiation between different rooms. Before discussing these patterns, however, we would like to discuss two types of finds that were not found at the site, and this, we believe, is a striking phenomenon. Weaving

We would like to thank Prof. Amos Kloner for his cooperation and help regarding unpublished data from Maresha. 26 Theoretically, two small stone weights could have been loom weights, but this is seriously doubted (see Peshin, Chapter 3.3 above). Not only is this very unlikely, but the vast majority of loom weights in this period were made of clay, and such items were completely missing at Kh. erRasm, casting further doubt about the possibility of the use of the stone weights for weaving. In any event, the absence of clay loom weights at Kh. er-Rasm stands out in relations to other sites. 27 The exact dating of the finds was not published, but the site was rural throughout its existence. 25

One type of finding, which is typical of most domestic setting of all periods, including the Hellenistic period, is loom weights (mainly of clay). At Maresha, for example, loom weights were found in many rooms in which domestic activity took place (Orit Shamir, personal communication). There was not a single residence at Maresha without loom weights, and they were unearthed in more than one room in

218

Avraham Faust and Adi Erlich

Shamir personal communication), makes this observation significant (below).

Both evidences of agricultural activity and of weaving appear, therefore, to have been common in rural sites of the Hellenistic period. The lack of small finds relating to agriculture and even rock-cut installations in and around Kh. er-Rasm, as well as the absence of evidence of weaving, should therefore raise some questions.

Agriculture We have hardly any evidence of agricultural activity at the site. This observation relates to both the absence of small finds indicating agricultural activities and the absence of installations. We found one large stone which appears to be a weight for an olive press, but it was found out of context (it was probably earlier in date). No other evidence was unearthed (there is a columbarium in a nearby cave, several dozens m. from the structure; [see Chapter 1.3]. This, if dated to the same period, might also be counted as indirect evidence of an agricultural installation). It is, of course, possible that more evidence is lying in Cave H, below the building, as is often the case in this period. But as already noted, the subterranean complex collapsed and fell out of use before the site’s final phase. The collapse of the cave explains the strange location of the single olive press weight that was unearthed – in the courtyard and out of what was probably its original context.

In light of the above mentioned lines of evidence in relation to the question posed earlier we will now discuss what type of site was Kh. er-Rasm in the second century BCE. Discussion What could have been the function of the building unearthed? Isolated buildings are known in many periods and regions, and were usually interpreted in one of the following ways: 1) A farm: According to Clark, a farm is “any tract or tracts of land or of water, varying greatly in extent, worked as a unit, used for cultivation of crops or for the rearing of livestock or fish, under individual or collective management”, and a farmstead is “the land and buildings of a small farm” (Clark 1993:114, our emphasis, A.F. and A.E.). Farmsteads are therefore isolated structures located in the midst of agricultural areas, and their inhabitants are usually the people who cultivate and use the land. The farmhouse can take many forms. Farmsteads, or small farms, should be differentiated from estates of various kinds (Faust 2003).

Evidence for agricultural activities is usually found in abundance in rural sites, including from the Hellenistic period: The nearby site of Aderet can serve as an example – here, an olive press with all its installations was unearthed, as were several small wine presses (Yogev 1982).28 The same is true regarding other sites, e.g., at Mazor, where an olive press and a rock-cut wine press were found (Amit and Zilberbod 1998), at Tirat Yehuda, where a large olive press was found (Yeivin and Edelstein 1970), at Horvat ‘Etri, where wine presses and an olive press in secondary use were found (Zissu and Ganor 2002:18),29 at Qalandia, where an olive press and six wine presses were uncovered (Magen 2004:35–38, 40, 43–45), and at Har Adar, where a wine press was unearthed (Dadon 1997:74).

2) An estate differs from a farmstead both in size and social composition. Roberts defines an estate as when “a powerful landowner has control over the people and resources of a large tract of land” (Roberts 1996:57). Estates, in contrast to farmsteads, extend over large areas and belong to the wealthy and whether the owner lives there or not, the residents may include servants, hired laborers, slaves, etc. (see also Roberts 1996:15ff.; 56–60). While the structures can assume various forms, at least some of them are expected to show a high level of building and perhaps even luxury (see also Faust 2003).

Discussion: Tirat Yehuda, Mazor, Aderet (outside Judea) as well as Horvat ‘Etri, Har Adar and Qalandia (within Judea) are all Hellenistic farmsteads. The first two, just like Kh. er-Rasm, were destroyed in the second century BCE, while Aderet ceased to exist at an earlier stage and the last three continued to exist at least into the first century BCE.

Farmsteads and estates represent different socioeconomic systems (Safrai 1998:8–9). One of the implications of an estate is that in many cases the surpluses or “profits” leave the countryside and go to the city (for a similar, albeit not identical, system in the ancient Near East, see MagnessGardiner 1994). The farmstead, on the other hand, is an integral part of the rural sector (Faust 2003).30

The olive press at Aderet is located on the surface, unlike many other Hellenistic olive presses in the Shephelah which are often subterranean. This may be due to the specific formation of rock in Aderet, Adulam formation of Zor‘ah form which is composed of kirton and flint, a formation which is less suitable for quarrying than the soft Maresha formation of the Zor‘ah form located west and south of it. Another theoretical explanation for the surface location of the olive press at Aderet is the relatively early date of the site within the Hellenistic period, perhaps prior to the large-scale quarries at the Shephelah, but this should be ruled out since it seems that at Kh. er-Rasm the hewing started relatively early in the Hellenistic period. While the above cautionary data should be kept in mind, it is also possible that in farmsteads, when space considerations were not a problem, olive presses were located on the surface, and not below it. Only future excavations of more farmsteads will allow us insight into this issue. In any case, there are several agricultural installations at Aderet that are absent from late second century BCE Kh. er-Rasm which is only slightly later and only a few km to the east. 29 It should be noted that the exact dating of the installations at Horvat ‘Etri is not certain. 28

3) A fort: A fort is usually a fortified structure that was used to house soldiers (and hence requires some form of political complexity in order to build and maintain it). Forts are usually viewed as fortified structures, and while this is indeed true in many cases, this was not always so. Forts that 30 Note that Tal, in his recent detailed survey (Tal 2006) did not differentiate between farmsteads and estates; he referred only to what he called “agricultural estates” – a category which included all isolated structures whose economy was based on agriculture.

219

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

were supposed to withstand an enemy siege were indeed fortified. Others, which only served as a garrison, and when no strong enemy was expected, could be left without real fortifications. Though not always heavily fortified, forts were probably always enclosed structures. Forts were also usually expected to be symmetrical, though it is not certain whether this was the case in the Hellenistic period (cf., Tal 2006:138–163; more below). Some forts were located within settlements, of course, but in many instances they were isolated structures, located in the countryside.

organization of space, however, might be similar to forts. Moreover, caravansaries are expected to be fortified, at least outside the settled area, and this might further complicate differentiating between them and forts. In some cases, buildings could serve both purposes at the same time. For example, the forts along the Nabatean roads in the Negev served in most cases to control the road as well as to host the caravans (Cohen 2000). While additional possibilities or combinations may exist (Faust 2005a:193–207; see also below), we can summarize the similarities and differences between the four types in the following manner:

4) An inn: An inn, or caravansary, is a structure intended to serve as a haven for travelers – a secure place to sleep and eat (see detailed discussion in Casson 1994; see also Thareani-Sussely 2007). Many such structures can be found within settlements, but many others are located along the roads which connected various settlements (e.g., Stern 1997; Casson 1994; Thareani-Sussely 2007). Inns are found along the main roads in the Greek world from at least the fifth century BCE, and were more common later on (Casson 1994:87). Many of the country inns were located at about a day’s travel from one another or from the nearest town, i.e., about 8–10 miles apart (Casson 1994:189–190, 201; note that this data refers to good Roman roads).31 At times, the services were well organized, while in others they were more sporadic. In most cases, however, the inns were quite humble to say the least, and their services very limited (e.g., Casson 1994:87, 88, 203–204). Aside from a resting place, the caravansary also provided security for the travelers. In most cases the caravansary is composed of a system of rooms around a central courtyard (Stern 1997; Casson 1994, see more below). Stern (1997:22) notes that it is similar to a fort, but without the fortifications. Stern (ibid) also notes that of the later caravansaries, those of the second millennium CE, about 30% are “small” ones, whose area was not more than 1,000 square meters. Inns and caravansaries hosted a wide range of activities, from hostels to markets, and could also serve as garrisons, brothels, etc. (Stern 1997; Casson 1994; Kleberg 1957).

Both estates and farmsteads are mainly agricultural sites, and in this they differ from caravansaries and forts. It appears that the archaeological means of differentiating between farms and estates are the size of the agricultural area, when known, and an examination of the number, size and quality of the farm house(s), their composition and, at times, their contents (Faust 2003). An estate would have a large agricultural area, while a farmstead would have a relatively small area that could be worked by a family (small or large, see also below); the products (or income, depending on the type of economy) from it would be sufficient for the needs of the family unit. An estate would include several buildings, or one very large building, which could house the owner and his family (whether throughout the year or only at certain times), the manager, and in many cases his family, and workers. The owner’s house would be relatively opulent, and a difference is expected to be seen between the residence of the owner (or manager, or both) and the dwellings of the workers. The farmhouse would suffice for a single family and only a few workers, if any. Forts are always enclosed buildings, and in most cases they are fortified. They are a state endeavor, in many cases uniformly built and usually symmetrical. A farmstead is less formidable. An estate might also be enclosed and symmetrical, but would differ according to the apparent wealth of its owner and, as we have seen, the amount of evidence of agricultural activities.

Distinguishing among the Various Options

A caravansary could also be fortified and symmetrical, but this is not always the case, and most inns were more humble.33 When a building is formidable, fortified and symmetrical, and has no or little evidence of agricultural activities it should be considered a fort,34 or if there is evidence that hints at a caravansary, as a combination of both. When the structure is less formidable, not necessarily less luxurious, but still shows no real signs of being an agricultural site, it could have been an inn, but this will require more consideration of other finds that can support or refute this hypothesis.

How can we distinguish among the various options, and decide whether an isolated structure is a farmstead, an estate, a fort or an inn? A fort is expected to be relatively fortified and symmetrical. A farmstead is expected to be less fortified and not necessarily well built. Estates can be well built, but one expects to find evidence of luxury in them, which is less likely to be found within forts. Inns or caravansaries are expected to be organized differently from an agricultural site.32 Evidence for agriculture at such sites is expected to be less common, and one should find rooms organized around a yard, in order to host both guests and animals. Such an

A consideration of all of the above should help us decipher the function of the site in question.

According to Dorsey (1991:45) the distance between inns in the Persian period was 10–15 miles, depending on the topography, and the distance in the Roman world much larger (25–35 miles, though much closer in inhabited regions). 32 Thareani-Sussely (2007) had recently devoted a lengthy discussion to ancient caravanseries (see more below). 31

We are not discussing city inns, as this is irrelevant for the present discussion. 34 Note that there can be agricultural installations in a fort as in the olive press at the fort in Moa (Cohen 1987:28), but when no such installations are found, a farmstead or an estate must be ruled out. 33

220

Avraham Faust and Adi Erlich

Khirbet er-Rasm Reexamined

garrisons by Bacchides, the Seleucid field commander, after defeating Judah, the Maccabee, in the battle of El‘asa (Fischer 1995:87–88; Roll 1996). Some of these forts were in towns, but others are identified with sites in the countryside. One of these forts is Timna in the valley of Pharaton, which is identified as the Elah Valley (Roll 1996:512). Timna is indeed identified to the east of this valley and not on its western edge like Kh. er-Rasm, but this list of forts gives us an insight into fortifications erected by the Seleucids in Judaea and its surrounding during their battles with the Hasmoneans. At any event, the contribution of the list to the present discussion is limited.

When we attempt to analyze the function of Kh. er-Rasm’s central building, the following points should be considered. Agriculture: Khirbet er-Rasm as a Farmstead or an Estate The lack of clear evidence of agricultural activity seems to render the first and second options (a farmstead or an estate) quite weak. After all, most of the building was excavated and, moreover, its immediate surroundings were carefully surveyed in a search for rock-cut installations. While it is possible that some installations lie in the subterranean complex, during the structure’s last phase (which is the phase discussed here) the caves were already out of use. The dearth of agricultural finds clearly indicates that the site was different from Aderet, Mazor, Tirat Yehuda, Qalandia, Har Adar and other contemporary farmsteads, some of which are located in the same region or in adjacent areas.

2) The second source is the archaeological evidence. Several Hellenistic forts have been found in the Land of Israel (Tal 2006:138–163). Acco-Ptolemais, for example, was defended by forts erected in Upper Galilee (Aviam 2004), and fortified sites have been found also further south. We will concentrate on such buildings in the region of Kh. er-Rasm, in Judaea and Idumaea. Two forts which were excavated, and both of which also appear on Bacchides’ list, are Horvat Eqed near Emmaus and Beth-Zur in the Hebron hills. The two demonstrate how contemporaneous forts in the region can differ in plan: Horvat Eqed is built in an irregular fashion on the summit of a hill, and is surrounded by a wall and round towers (Fig. 4.21; Fischer 1995; Gichon 1993), while Beth-Zur is an almost symmetrical square building, slightly larger than Kh. er-Rasm (Fig. 4.22; Sellers 1933: pl. II; see discussion on the possible different phases in Tal 2006:150–152). Recently, Zissu (2004) suggested that Horvat Tura, a fortress which was surveyed west of Jerusalem, should be attributed to Simon the Hasmonean. The plan of the fortress is again different from other forts, such as Horvat Eqed and Beth-Zur (Zissu 2004:9, fig. 6).

Loom Weights: Families at Khirbet er-Rasm The absence of loom weights seems to indicate that no women were present on a permanent base at the site, at least during its last stage of existence, or to be more accurate, that no typical female activities were conducted at the site before it was destroyed – after all weaving became the ultimate symbol of femininity in many traditional societies, including in the ancient Near East and in the Greek world (cf. Murdock 1937; Murdock and Provost 1973; Hoffner 1966; Bourdieu 1979; Barber 1991; Carr 2000; Malul 1996; Cassuto 2004; see below for some reservations). Furthermore, even if we should assume that loom weights and/or spindle whorls will be discovered in the site in the future, in a room that has not yet been excavated, their absence in most rooms is a clear indication that the site differed from other Hellenistic rural sites (above). This stands out also in comparison with the finds from Maresha, where loom weights were frequent. It appears that we cannot reconstruct any “regular” social organization at Kh. er-Rasm, e.g., a number of families sharing the site, whether as part of an extended family or not. This also seems to rule out a farmstead or an estate.

There are a few other military constructions in southern Idumaea. Tel Maresha is thought by Tal (2006:149–150) to enclose a fort, which others interpret as an administrative building with a shrine in its court (Avi Yonah and Kloner 1993). Tel Beersheba is also thought to occupy a fort (Tal 2006:153–154). As for non-urban forts, Tel Arad and Kh. ‘Uza in the Negev demonstrate again the variety and inconsistency in military buildings during the Hellenistic period, although both represent a reuse of forts from earlier periods. The first is a square building with some rooms and a court occupied by one large tower at its corner (Herzog 1997:248–250), while the second followed the Iron Age square outline of the building except for the western new line, and added rooms surrounding a large court inside (Beit-Arieh 1993; Tal 2006:154). The Hellenistic users of the fort at Kh. ‘Uza used the already existing buttress-like towers, and also added another one at the new western wall of the building, which emphasizes the defensive character of the building. Despite the difference between the two nearby buildings, they must have served similar purposes.

Khirbet er-Rasm as a Fort The initial plan of the site (Fig. 4.11) makes it tempting to suggest that it was used as a fort, as it is similar to a number of Iron Age II forts or estates, for example at Kh. Abu et-Twein and Deir Baghl (Mazar 1982; see also Faust 2005a:199–206, and the discussion in Chapter 4.1). Before continuing, it is worth examining the information we possess on forts in the region during the Hellenistic period: There are two main sources regarding forts in Qoele Syria during the second century BCE.

It is possible that Seleucid forts in Judaea vary so greatly in plan because of a process in which existing structures were refurbished into forts (Roll 1996:513). The variety of military buildings in the region during the Hellenistic

1) The first is the list in the first book of Maccabees (1 Macc. 9:50–52) of the forts restored and inhabited with

221

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

4.21 Plan of Horvat Eqed (after Gichon 1993: 416; courtesy of the Israel Exploration Society, and Prof. Moshe Fisher)

Era is also attested in Tal’s typology (2006:155–13). If Kh. er-Rasm served as a fort, it would fit Tal’s second group of military buildings: square buildings defined by him as “fortresses”.

forts described above. It owes its symmetric fort-like shape to what we believe to be an earlier stage of the building. It has neither towers nor fortifications, neither in the original stage nor in the last phase under discussion. Tal (2006:158) mentions the administrative role of forts in addition to their military function, and believes that such buildings served for collecting and storing taxes. The picture drawn from the rather poor and ascetic findings from Kh. er-Rasm, the almost complete absence of imports and the small number of storage vessels, certainly does not suit an administrative fort serving the Seleucid governors in the northern part of Idumaea.

It should be noted, however, that when the site was destroyed it was not used in its original plan any longer. At this time it was clearly used in a different manner; it was asymmetric, and probably not fortified, therefore ruling out a fort. While it was probably still surrounded on four sides by (some) walls, only the northern, western and southwestern wings were used as buildings, and the eastern and southeastern parts of the building, while (partially at least) built, served as part of the courtyard (with an installation and a tabun even in the room by the gateway).

Another parameter which should be examined is the proximity of a fort to a major road. This parameter is, of course, valid for an inn as well (see below). The forts restored by Bacchides all controlled major roads to and

Kh. er-Rasm is therefore different even from the varied

222

Avraham Faust and Adi Erlich

is actually a low summit located between two much higher ones, Tel ‘Azekah and Kh. Shiqlon. The former overlooks the Elah Valley from the north, and occupies a fortress dated by Dagan (2000:88–89, 232–233) to the Hellenistic period. The latter overlooks the area from the south and is thought to be a fort that overlooked the road during the Roman period (Roll and Dagan 1988; no Hellenistic remains were found; Dagan, personal communication). Kh. er-Rasm which is topographically inferior to both of them, is certainly connected to the crossroads on its foothill, but not as a fort. The fortified high summit of Tel ‘Azekah, if indeed dated to the Hellenistic period, suits this purpose much better, and clearly, there was no need for an additional fortress nearby. It is quite clear, in light of the above, that Kh. er-Rasm’s plan, lack of fortification and position are not suitable for a fort. Khirbet er-Rasm as an Inn Inns in the Greek world were usually square or oblong, with a courtyard in the center, used for animals or vehicles (when relevant), surrounded by a shallow continuous structure (Casson 1994:88, 202–203; see also ThareaniSussely 2007:127). Although at Kh. er-Rasm not all the surrounding structure was used for habitation, the plan is very similar to Casson’s description. The courtyard in Kh. er-Rasm is very spacious and with no signs of any human activity, hence, it was suitable for accommodating animals. It seems, therefore, that an inn is a reasonable explanation (see more below), not only because all the other suggestions were ruled out but since the pattern observed in the finds seems to support this suggestion.

4.22 Plan of fortress at Beth-Zur (after Funk 1993: 261; courtesy of the Israel Exploration Society)

from Jerusalem (Roll 1996:510). Kh. er-Rasm appears to fit this parameter, as it is located near the Elah Valley (Fig. 4.23), next to a junction of roads leading to Maresha on the south, Emmaus on the north, Jerusalem on the east and perhaps another road leading westward, to Ashdod and Jamnia, both Hellenistic towns (Fig. 4.24). But Kh. er-Rasm is not the best spot for viewing the crossroads. It

4.23 Aerial view of Kh. er-Rasm, looking east toward the Elah Valley

223

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

installation was used for storage, one can also speculate that it was used as a service area for the guest rooms, perhaps even as a latrine or privy of sorts, and that the cooking pots are actually chamber pots (cf., Baruch and Amar 2004:41–43; Herr 1997:120), and there is evidence for their importance in inns as well (e.g., Casson 1994:88, 209). Not much is known about rural latrines in the region, but Baruch and Amar recently summarized the knowledge of such features (both in rural and urban settings). They note that aromatics and oils were used to overcome the odor in the latrine (Baruch and Amar 2004:43). While this is speculative, perhaps the bottle unearthed in the area was used for just such a purpose. The only items associated with women that were found at Kh. er-Rasm, the cosmetics (kohl) sticks, were not found in the northern wing (where all the other metal objects were found); these might have therefore have belonged originally to female guests, rather than to women who lived in the house on a permanent basis. This is, therefore, still in line with the observation regarding the absence of evidence of “typical” feminine activities at the site. It is also possible that women comprised a large percentage of the guests at the site; inns also provided additional services to the guests. Most inns also provided prostitutes (Casson 1994:204, 207, 211, see also pp. 37, 207, 215; Kleberg 1957:89–91; Dorsey 1991:43), and as Casson (1994:211) noted, in many instances the maids provided those services, for an additional fee as this was the only entertainment available for guests at a country inn (Casson 1994:211), like Kh. er-Rasm. The likelihood of this suggestion is even strengthened if we remember that two cosmetic sticks were found at Kh. er-Rasm (one of which was found in area B in room 1, which functioned as part of the courtyard, near its entrance, and one on the surface near the hiding cave complex). The two cosmetic sticks seem quite a lot given the paucity of metal objects at the site which include an arrowhead, obviously belonging to the destruction phase and not to the final occupation, as well as (sometimes fragments of) a hinge, a nail, two rings, and other objects. Could it be that the sticks were used by such women?36

4.24 Map of ancient roads during the Hellenistic period

According to such an interpretation, the northern wing is where the innkeeper lived. It is here that we find the majority of artifacts, including concentrations of finds that were probably kept here and in specialized rooms. This also appears to be the area where food was prepared. Also, almost all the metal objects, including coins, were found there. This is probably because it is here where people dwelt on a permanent basis, and hence more valuables were lost. The western and southwestern wings were used for the guests. Here, on the basis of the rooms that were excavated, and taking into account the fact that no baulks were removed, it can be reconstructed that a typical room included a bowl (or bowls), a lamp and a bottle, and it is likely that this was the typical assemblage with which the “guest-rooms” were equipped (cf., Casson 1994:208).35

The location of Kh. er-Rasm, about 11 km from Maresha as the crow flies, also strengthens the proposed interpretation, and this will receive a more detailed discussion below. A possible problem with this interpretation, though not an overwhelming one, is that one might have expected a woman to reside even in an inn, at least in the innkeeper’s part of the structure. Many if not most innkeepers in antiquity were females, and this seems to have been a very old tradition (Casson 1994:87, 208, see also pp. 37, 206, 207; Dorsey 1991:44). Still, they were probably not regarded as “ordinary women”, and it is doubtful whether typical domestic female activities were expected of them. It

Interestingly, in the southern part of room (space) 9, two cooking pots, a bowl and a bottle were found in and near a rounded installation. Room 9 is bordered on the south by a kirton wall, which was added to the main building later, probably during the period being discussed here. While the room continued northward, the area under discussion is much deeper than the rest of the space, and is bounded to the north by a sharp sloping rock, whereas the northern part of this space is much higher. While we assume that the Theoretically, one could suggest that the northern wing was used to store vessels, while the southwestern wing was used for living. While this is possible, it doesn’t explain the lack of evidence of agricultural activity, the absence of loom weights, nor the scarcity of storage jars in the northern wing, which, according to such a suggestion was used for storage.

Although it is much more difficult to assess the significance of this observation, it appears that the number of bottles unearthed in Kh. er-Rasm is relatively high (Sandhaus, Chapter 2.2.1). Again, they may have been used by women, but not necessarily reflecting typical domestic activities, but rather belong with cosmetic sticks.

35

36

224

Avraham Faust and Adi Erlich

is likely that the cosmetic kohl sticks and perhaps the large number of bottles characterize those women better than the domestic loom (for the latter’s symbolic importance, see Cassuto 2004 and many references).

the region as a supplier of prostitutes seems to be in line with the above interpretation of Kh. er-Rasm as an inn.

There are additional lines of evidence which supports the interpretation of Kh. er-Rasm as an inn:

As noted above, caravansaries or inns, when located outside settlements, are likely to be situated near roads. What can we say regarding Kh. er-Rasm’s situation in this regard?

The Location of Khirbet er-Rasm

Petrographic Finds

The Elah Valley, above which the site is located (Fig. 4.25; see also Figs. 4.23, 4.24), was a major crossroads throughout history (e.g., Aharoni 1979:58, 59; Dorsey 1991:154–155, 189–191; Ziv 2005). It is commonly thought that during biblical times several roads met at the valley, just below the site, i.e., the longitudinal road from Maresha to BethShemesh (Aharoni 1979:58; Dorsey 1991:154), and the east-west road from Ashkelon to Jerusalem, or Bethlehem (Dorsey 1991:189–191; Aharoni 1979:59). Moreover, it is likely that several east-west routes converged near the Elah Valley (see also Ziv 2005).

Interestingly, while the pottery unearthed at the site is, generally speaking, very poor,37 petrographic analysis seem to indicate that it was brought from a relative distance, and from various sources (Cohen-Weinberger, Chapter 2.5). While the site is located in the Shephelah, the pottery came from all over southern Israel – some of it is local, but some came from the highlands, and some even from the coastal plain. We believe that the dispersed origins of the pottery, which stand in sharp contrast to the overall poor quality of the vessels, does not result from extensive trade, but rather from the fact that the site is located on a major junction, and that it had many guests. Four Hellenistic cooking pots were examined, and these came from three distinct sources, none of them in the immediate vicinity of Kh. er-Rasm: one vessel was brought from as far as the coastal plain, another from the hills farther east (although similar vessels made of terra rossa are frequent in the Shephelah), and the third either from the northern Negev/southern Shephelah or, more probably, from the inner coastal plain. It seems to us that the findings support the interpretation of the site as an inn rather than as a farmstead or the like.

Admittedly, it is very difficult to trace ancient, pre-Roman, roads; the route of such roads is hypothesized almost solely on the basis of topography and ancient settlements (Roll 1996:509). But the later Roman roads which crossed the region also indicate that the site is situated just above the meeting point of two of the most important roads of the Shephelah (Roll 1983:139; 1996:510). The road from Beth Guvrin (near Hellenistic Maresha) to Jerusalem crossed the valley, as did the road from Emmaus to Beth Guvrin (e.g., Dagan 2000:78–79). Another road apparently crossed the valley slightly northward, using the trail to Emmaus, and then turning west, toward the coastal plain, using the Nahal Ha’Elah.

Zenon Papyri

In addition, it was also possible to go due west from the site itself, or via the small wadi just north of it, toward the Lower Shephelah and the coastal plain (eventually reaching Ashdod, which was, by the way, destroyed in similar circumstances; Dothan 1971:22).

Another kind of historical evidence regarding the Hellenistic period that should be cautiously taken into consideration is the Zenon papyri (although we should remember that they predate the discussed phase in Kh. er-Rasm by more than a century). The travels of the Ptolemaic officials in Qoele Syria are well documented in these papyri (Tscherikower 1937; Durand 1997), as is their trade in slaves in Marisa (Maresha); unfortunately we know nothing about the facilities they used in that region for overnight stays between journeys. However, the documentation of many officials in the country applying for travel costs from the Ptolemaic administration points indirectly to the need for inns on roads, and especially on crossroads. We believe Kh. er-Rasm has all the geographical conditions to serve as a lodge.

It is therefore clear that the westernmost part of the Elah Valley was a major junction in many periods in antiquity, probably throughout history. Kh. er-Rasm is located on a salient hill (not the steepest, though, see above) just west of the valley, about half a km as the crow flies, from the valley plain. It is not clear, of course, where the ancient junction was, but even today’s junction is less than 1.5 km from the site. It is likely that the ancient one was where the wadi enters the valley, and was therefore less than half the distance from Kh. er-Rasm.

Zenon papyri also testify to the trade in female slaves as prostitutes who were dragged from place to place (Tscherikower 1937:9). While not proving anything regarding Kh. er-Rasm, of course, the frequent mention of

It appears, therefore, that Kh. er-Rasm was situated at a very convenient spot for an inn. It is likely that due to security reasons in the region at the time, an inn would be located on hills, and not in the valley itself. Kh. er-Rasm is situated on top of a hill, which could be well guarded. But of all the relevant hills west of the valley, Kh. er-Rasm is the least steep, and was, therefore, the easiest to climb (cf. Tel ‘Azekah and Kh. Shiqlon; see above in the discussion

No imported vessels were found among the rich corpus which is dated to the second century BCE, and the only imported sherd was among the limited amount of pottery dated to the Early Hellenistic period. 37

225

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

4.25 Aerial view of Kh. er-Rasm, looking east toward the Elah Valley

about forts). It is likely that it was chosen precisely because of this location.

of sleeping accommodations and food preparation and consumption areas. Thareani-Sussely also mentioned four “additional features” that “are of secondary importance”. These include animal holding pens, security posts or fortresses, trade centers and a water supply system.

Moreover, not only is the situation of Kh. er-Rasm suitable for an inn, so is its location in relation to other settlements. Kh. er-Rasm is located (as the crow flies) about 11 km from Maresha to the south, about 30 km from Ashdod to the west, almost 30 km from Jerusalem to the northeast, and 18 km from Emmaus (and 19 from Gezer) to the north. This is a very convenient location for an inn (cf. Casson 1994:189–190, 201; Dorsey 1991:45).

Clearly, all the most important features exist at Kh. erRasm, and even several of the secondary features can be identified at the site. The structure is close to major trade routes and is located near a junction; it is set apart from the local population, outside Tel ‘Azekah; we have seen above that the plan of the building and the finds within it indicate the existence of sleeping accommodations; and the functional analysis suggested that there was a separation of food preparation and consumption. As far as the “additional features” are concerned, the courtyard was useful as an animal pen, and water cistern(s) was (were) found nearby (and perhaps a few others remained in use beneath the building, in case some of the caves were still usable at the time), though this is not surprising. We have no evidence for a security post, though, nor for the use of the building as a trade center, although the existence of cooking pots from various sources might hint at the latter option (cf., Thareani-Sussely 2007:136; though the variety at Kh. er-Rasm is much more limited). The relatively large

While none of this proves the interpretation of the site as an inn, it clearly allows for it and reinforces the theory. Discussion Interestingly, Thareani-Sussely (2007) suggested a model for identifying caravanserais in the archaeological record. While this is not the place for a detailed discussion, we would like to note that she listed a number of features that are expected to be present at such structures (pp. 126–128). Four features “are of primary significance; and hence they are the most important for identifying ancient caravanserais”. Those features are: proximity to trade routes, separation from the local population, existence

226

Avraham Faust and Adi Erlich

percentage of lamps in the assemblage (7.5%) is also in line with Thareani-Sussely’s (2007: 136) interpretation.

evidence for the daily presence of women. The latter exists, to a very limited extent, only outside the northern wing. These finds seem to rule out an interpretation of the site as a farmstead or an estate or as a fort, and it seems to us, therefore, that the most likely interpretation of the site is that it served as an inn of some sort.

It appears that various lines of evidence converge, indicating that Kh. er-Rasm was most probably an inn during the late second century BCE. Alternative Interpretations

This conclusion, however, will remain preliminary for the time being since we do not have comparable material from this period and the discussion is therefore somewhat speculative. We hope, however, that the present analysis will enable future discussions, and that when additional buildings are excavated the present paper will serve as a starting point with which to examine the new data. The new information, in turn, will help to reanalyze our data, and either refute, confirm, or refine our conclusions. Only future discoveries in Idumaea and nearby regions will enable us to assess the present interpretation, and to advance research into the social and political composition of Idumaea.

Viewing the site as something between a farmstead and an estate, with the northern wing representing the lodge of the owner, is very tempting. According to such an interpretation most of the finds were located here because here lived the wealthy family (or person), while the southwestern wing was inhabited by workers only. The dearth of evidence for agriculture and for the typical activities of women (even in the northern wing), however, render this option unlikely.38 It is also possible that a fifth option should be considered. Perhaps the building served in a capacity which is somewhere in between options 2 and 3, i.e., as an agricultural estate of the “state”. Perhaps the inhabitants were “corvee labor” recruited to work for the state (cf. Faust 2005a:193–207). This might explain the absence of women in the building, as well as the differentiation observed between the northern wing, where the “officer” / overseer stayed, and the western and southern wings, in which the workers dwelled. Currently, however, we know nothing about the social composition of Idumaean society (or other contemporary polities and societies), so we cannot evaluate even the likelihood of the existence of such institutions. Furthermore, the lack of agricultural installations in the vicinity of the building seems to make this option less plausible, as the laborers were expected to engage in such activities. We therefore believe that some sort of an inn is a better theory.

4.2.4 Khirbet er-Rasm during the Late Second Century BCE in Relation to Its Surroundings The aim of the present chapter is to examine the site at Kh. er-Rasm during the second century BCE, its main period of occupation, in relation to its surroundings. By surroundings, we refer mainly to the human landscape, i.e., other settlements, forts, and roads, but also to the natural environment, as far as it can be reconstructed from the archaeological evidence. The discussion will begin by noting some of the methodological problems of such an enterprise, which is based to a large extent on the results of surveys. After that, we will summarize the data on the human features in the area in which the site was embedded. Notably, we have suggested above that Kh. er-Rasm was an inn at the time, and we should therefore examine its location in regard to both the major roads which crossed the region and the major cities of the period. Here we will summarize this data, and will add more information on neighboring settlements of various kinds, as it is known both from the survey and from the excavations. The third part will add the information we possess on the “natural” landscape (whether “impacted” by humans or not).

Summary: The Function of Khirbet er-Rasm The excavations at Kh. er-Rasm have revealed a large square isolated structure, dated to the Hellenistic period. The building was destroyed in the late second century BCE, probably by the Hasmoneans during their campaign in Idumaea. Based on the dispersal of the finds within the site, it appears that the main part of the building was the northern wing. Here most of the pottery was kept and food prepared. Furthermore, it is only in this wing (and in the nearby part of the courtyard) that coins were found, probably representing loss during the site’s period of existence. The absence of such finds in the western and southern wings seems to indicate that activity there was less permanent, and hence fewer coins were lost.39 In addition, there is practically no evidence for agricultural activity at the site, and hardly any

Methodological Problems with the Data As noted, most of the available data regarding the contemporaneous settlement around Kh. er-Rasm is derived from surveys. The region around Kh. er-Rasm was surveyed by Yehuda Dagan, as part of a long term and meticulous project of surveying the Shephelah. Some of the survey maps have already been published (e.g., Dagan 1992a; 2006), and much of the data appears in Dagan’s doctoral dissertation (Dagan 2000; note that the region of Kh. er-Rasm has not been published yet – we are grateful

The absence of loom weights could theoretically be explained by suggesting that there were no women at the structure, only a male overseer. This would make the suggestion more similar the fifth option discussed below. The lack of evidence for agriculture, however, makes this suggestion untenable. 39 We should note, however, that this is relevant only for the building’s last phase. Since we do not know how long this phase lasted, as we don’t know when the collapse of the cave system took place (and although 38

we tend to date it to some point in the second century), this could have happened at the beginning of the second century and theoretically even earlier.

227

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

to Yehuda Dagan for giving us information on some of the sites, below). All in all, some 255 Hellenistic sites were discovered in this survey (Dagan 2006:41*).

system of the Hellenistic period. We will only present the (limited) data, with a minimal attempt at reconstructing the overall settlement system.

The Shephelah survey is one of the most detailed surveys carried out in Israel. Generally speaking, this is precisely the type of data which is usually used for reconstructing settlement patterns. Still, for a number of reasons, surveys are a very problematic source of information for such purposes, and this is especially acute in regard to the Hellenistic period. We shall describe the problems from the more general to the specific:

These reservations are relevant only for the data on settlements as derived from surveys. The data on roads is speculated on the basis of both later roads and data on settlement in various periods, and is not likely to have changed much between the two sub-phases of the Hellenistic period. The data from excavations, both from other sites and from Kh. er-Rasm, is also much more secure and can be used more easily (we are referring to data on the environment, as data on settlement are very limited).

1. Many sites are inevitably missed during the survey, and are not identified at all. Other sites can be identified, but some of their periods of occupation, including the Hellenistic period, might be missed. This, however, is true of all surveys, and the errors are “random” (Faust and Safrai 2005, and references). We could still have used the data, with the appropriate reservations, but there is another, much greater problem. 2. The term “Hellenistic period” encompasses a long period of time. We know that there were major changes in the transition to Hasmonean rule in some parts of the country, when the former Hellenistic sites were destroyed or abandoned and new sites were established afterwards; this is the situation, for example, in Samaria’s western foothills, and it seems to be the case in the Shephelah too (see Chapter 5.1). There is a clear change, and in many regions even a break, between the rural settlement of the Early Hellenistic period, and that of the Late Hellenistic period (Hasmonean period). The archaeological data indicates quite clearly that in most areas that were conquered by the Hasmoneans the previous rural system was abandoned, and replaced with another one, which lasted until the Jewish revolts. This is the situation at Kh. er-Rasm, and as we shall see below (Chapter 5.1), probably in the entire region. Clearly, examining both pre-Hasmonean and Hasmonean systems together is meaningless and even misleading. The survey, for obvious reasons, does not present us with data to differentiate the two systems, and hence should be treated with extreme caution. 3. Even during the earlier phases of the Hellenistic period, not all the sites existed throughout the period, and not all were therefore contemporaneous. This can be exemplified by reference to the excavated site of Aderet (Yogev 1982) – the site is “Early Hellenistic” (and preHasmonean) which might be seen as similar to Kh. er-Rasm. Still, the site was probably abandoned during the third century BCE or early second century,40 and while the settlement at Kh. er-Rasm existed at the time, this is much earlier than the main period of occupation discussed here.

Roads and Settlements in Khirbet er-Rasm’s Vicinity Roads: The road system around Kh. er-Rasm was discussed in detail in Chapter 4.2.3, and here we would like to mention briefly, once again, the main roads that crossed the region. The Elah Valley, above which the site is located, was a major crossroad throughout history (e.g., Aharoni 1979:58, 59; Dorsey 1991:154–155, 189–191; Ziv 2005). During biblical times several roads converged in the valley, just below the site, i.e., the longitudinal road from Maresha to Beth-Shemesh (Aharoni 1979:58; Dorsey 1991:154), and the east-west road from Ashkelon to Jerusalem, or Bethlehem (Dorsey 1991:189–191; Aharoni 1979:59). Moreover, it is likely that several east-west routes converged near the Elah Valley (see also Ziv 2005). A similar picture can be seen when examining the Roman roads which crossed the region, and the valley served as the meeting place of two of the most important roads of the Shephelah (Roll 1983:139; Roll 1996:510). The road from Beth Guvrin (near Hellenistic Maresha) to Jerusalem crossed the valley, as did the road from Emmaus to Beth Guvrin (e.g., Dagan 2000:78–79). Another road apparently crossed the valley slightly northward, using the trail to Emmaus and then turning west, toward the coastal plain, using the Nahal Ha’Elah. In addition, it was also possible to go due west from the site itself, or via the small wadi just north of it, toward the lower Shephelah and the coastal plain. Khirbet er-Rasm in Relation to the Major Settlements: The major settlements at the time were located at quite a distance from Kh. er-Rasm. The nearest city, Maresha, was located some 11 km to the south, as the crow flies. The nearest center in the north was Emmaus, some 18 km away (and Gezer was 19 km away). To the west, the nearest center was Ashdod, almost 30 km away, while Jerusalem was the nearest center to the northeast, and was located some 30 km away. We have already seen that the distances, too, were suitable for an inn. Khirbet er-Rasm in Relation to Other Hellenistic Settlements in Its Vicinity: In the survey (Dagan 2000), about 25 sites which yielded some Hellenistic remains were discovered in the immediate vicinity of Kh. er-Rasm (Fig. 4.26; 5 km as the crow flies; the following are their numbers in Dagan’s list of Hellenistic sites: 13, 36, 39, 41,

These reservations should be borne in mind while conducting the discussion, and we will therefore not attempt at any comprehensive reconstruction of the settlement On the basis of personal examination of the material from Aderet by our team, in conjunction with Dr. Jon Seligman.

40

228

Avraham Faust and Adi Erlich

4.26 Sites with Hellenistic remains in the vicinity of Kh. er-Rasm (based on the data provided by Dagan's Shephelah survey)

44, 47, 50, 57, 60, 62, 63, 66, 70, 74, 77, 78, 79, 83, 84, 85, 86, 89, 91, 92, 93), but most of them were defined as finding spots only, and not as settlements, not even as farmsteads. Only six sites in this region were defined as settlement sites, and these are (from north to south): Horvat Khusham (Kh. Kheishum; #13), Horvat Immer (Kh. Abu Amira; #41), Kh. el-Muzeria (#44), Tel Yarmuth (#50), Tel ‘Azekah (#60) and Socho (#79).

activity, or perhaps it was a resting place on the main road that crossed the Elah Valley. Due to the abovementioned problems in establishing the contemporaneous existence of all the sites, we cannot even attempt to reconstruct the system in which Kh. er-Rasm was embedded. We shall, therefore, address only one site, whose relevance cannot be denied: Tel ‘Azekah. Luckily, Tel ‘Azekah is not only the nearest “large” site to Kh. erRasm (ca. 1 km north of the site), but it was also excavated, and despite the fact that the excavations took place over a century ago, we have a relatively large amount of data to discuss; on the basis of the existing evidence it is clear that the site existed during the second century BCE. Its nature, however, is not clear, and it is likely that it was quite limited in size. Dagan notes that the excavators attributed one phase of the famous fort to the Hellenistic period, and, moreover,

Of these, the closest site to Kh. er-Rasm is Tel ‘Azekah, and it appears to have been the central site on whose periphery Kh. er-Rasm was located (below). It is likely that the other sites identified by Dagan were farmsteads or villages, while the find spots indicate some agricultural activity or something of very marginal nature: the nearby “find spot” at Kh. Jannaba at-Tahta (site #74), in which only a few sherds were discovered, might indicate some agricultural

229

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

4.27 Modern-day olive trees in Kh. er-Rasm

he himself believes that both phases should be dated to this period (Dagan 2000:232–233; most scholars attribure at least one phase to the Iron Age). If this is indeed the case, and even if only one of the phases should be dated to the Hellenistic period, Tel ‘Azekah was probably a central site, and Kh. er-Rasm was probably a client, at least as far as market and exchange were concerned. This is probably also where the inhabitants of Kh. er-Rasm fled to in 112/111 BCE.

per year (Kloner and Sagiv 1993; Kloner 2001:120–122; 2003:70–71). Although no wood samples have been yet published from Maresha, it is clear that olive trees were very common in the vicinity of the site. Kh. er-Rasm also had an oil industry at a certain point (see below). This might be in accordance with Zenon papyri, which document olive oil as one of the products the Land of Israel exported to Egypt (Tscherikower 1937:8, 22–23; Berlin 1997:6–8). This was probably the role of the region already during the third century BCE, and was probably a major part of the region’s economy during both the third and the second centuries. Olive trees continued to be dominant at the site through modern times (Chapter 3.12), and today, too, the site is covered with olive trees (Fig. 4.27).

Natural Surroundings Our best information on Kh. er-Rasm’s environment is derived from the analysis of the wood remains (charcoal; Liphschitz, Chapter 3.7). Some 38 boxes of charcoals were sent for testing and about 94% of the samples were from olive trees. It is clear that the majority of trees that were available for building were olive trees. While not ruling out other crops in the area, e.g., vines which could not have used for building, this clearly means that Kh. er-Rasm’s immediate surroundings were filled with cultivated orchards and the local “natural” trees were not readily available for building purposes.

We should note that during the late second century BCE the inhabitants of Kh. er-Rasm were probably not engaged in olive oil production, but it appears that olive trees were still the most readily available trees for construction. 4.2.5 Economy and Subsistence at Khirbet er-Rasm during the Second Century BCE The aim of the present chapter is to examine the information on the economy of Kh. er-Rasm, and on the subsistence patterns of the site’s inhabitants.

It is quite clear, therefore, that the entire area was cultivated intensively. Olive trees were probably also the leading crop in Hellenistic Maresha, as its main agricultural industry consisted of oil presses. Over 22 oil factories have been found in Maresha, which supplied about 9000 liters of oil

As far as economy goes, we have already suggested that Kh. er-Rasm served as an inn during the last phase of its

230

Avraham Faust and Adi Erlich

occupation (the late second century BCE). At this time, the site appears to have served guests, and earned its income from this trade (Chapter 4.2.3).

in the Hellenistic Shephelah, see, Fig. 4.5). It is plausible that in the Early Hellenistic period the production of olive oil was the major economic activity at the site, and that most of the production took place within the caves. This is also in line with the importance of the region as an olive oil supplier to Egypt at the time (as mentioned above). This changed, of course, when the caves collapsed and the structure’s function changed. Still, we should note that although no evidence for this was unearthed in the excavations, it is not impossible that the production of olive oil continued to be a factor in the economy of the site even during the late second century BCE.

We do not have any evidence for agricultural or industrial activity at Kh. er-Rasm at the time, as no finds relating to such activities were unearthed in the excavations (more below). If the limited architectural evidence (from the cave) that exists should be viewed as indicating some agricultural production, this was probably only supplementary activity. It is more than likely, however, that most of the limited architectural finds are related to the earlier Hellenistic phase (Chapter 4.1; see more below). As we have seen, most caves around the main building (e.g., Cave D, with possible evidence for olive oil production) should be dated to the earlier phases. Only the columbarium in Cave F (above), which was the last stage of use in that cave in antiquity, was probably dated to the late second century BCE, and it could have served as a supplementary income for the owner.

The importance of olives in the region’s economy seems to be supported by the wood samples discovered in the destruction layer. We have seen that about 94% of the charcoal that was uncovered came from olive trees (Chapter 3.6). This data demonstrates that olive trees were the main type of tree, of those that could have been used for construction, in the surroundings of Kh. er-Rasm, probably throughout the Hellenistic period.

Regarding subsistence we are only in a slightly better situation.41 The faunal remains (Shoam, Raban-Gerstel and Bar-Oz, Chapter 3.5), though very limited, comprised 75% sheep and goats and 25% cattle (MNI; the NISP ratio was 80/20). Shoam, Raban-Gerstel and Bar-Oz note that the “high ratio of juvenile cattle suggests that they were slaughtered primarily for their meat while sheep and goats were probably used for secondary products, such as milk and/or wool. An animal economy based on large herds of sheep and goats exploited primarily for milk or wool, and less for meat, is the “traditional Middle Eastern subsistence pattern.”’

A columbarium that was unearthed to the south of the main building might indicate that raising pigeons had a role in the economy. As noted above, the dating of the cave is uncertain, but it is likely that its use as a columbarium corresponds to the late second century BCE occupation (Chapters 1.3, 4.1). In summary, the available evidence regarding the economy of Kh. er-Rasm is quite vague, and even the data about subsistence is rather limited. From what has been unearthed, it appears that the finds, though limited, indicate consumption typical of a Mediterranean agricultural economy. It is likely that some agricultural activity took place at the site also during the last phase prior to its destruction, although no clear evidence has been discovered yet. There is better evidence for agricultural activity in earlier phases of the main bulding, probably before the collapse of the caves system (above).

The seeds found at Kh. er-Rasm were limited in number and make any conclusion rather tentative, but, as Weiss (Chapter 3.6) concluded: “(G)enerally, this plant assemblage hints at a typical Mediterranean diet, encountered already at other sites in the area. It includes crop plants: cereals, pulses and fruit, as well as weeds of cultivated plants and wild plants. Cereals are represented by wheat and barley, four grains each, pulses by three lentils seeds, and fruits by eighty three olive stones and three grape pips. In addition, there are eleven seeds of wild plants and weeds”.

4.2.6 Summary The site changed it character after the collapse of the cave system, probably during the second century BCE. During this time (phase 2c) some of the earlier parts of the building, although still standing, served as part of the courtyard.

While the amount of evidence is very sparse, it all points to a typical Mediterranean subsistence. It is possible that the very limited faunal and botanical evidence unearthed at the site is another indication that agriculture was not the primary occupation at the site, but it is more likely that the reasons for the scarcity of the finds should be sought elsewhere.

It appears that the family, probably a wealthy one, that lived here earlier had left the site, and it then became an inn, near the major junction at the Elah Valley. The impending Hasmonean campaign of 112–111 BCE (John Hyrcanus) caused the inhabitants to flee, taking their valuables with them but leaving the rest behind. They planned to come back, but were unable to do so. The site was subsequently destroyed, and not resettled for a significant period of time.

Interestingly, one olive-press weight that was found out of context (Fig. 4.4) strengthens this view that at an earlier phase there was agricultural activity at the site, most likely concentrated on olive oil production (for similar presses 41 Since all the finds in the Hellenistic structure were analyzed together, the evidence can also refer to earlier phases of the Hellenistic period, and not only to the late second century BCE.

231

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

4.3 Khirbet er-Rasm after the Second Century BCE 4.3.1 Reuse and Later Occupation in the Early Roman Period Following the destruction of the Hellenistic settlement in the late second century, the site was deserted for some 100–150 years. During this time the processes of collapse continued, and the walls crumbled. This is evident by the few Early Roman sherds that were uncovered within the wall fall (four sherds were found within the fall, one of them even deep within the fall [CP9 at L.822]42). The lack of later, e.g., Byzantine and Medieval pottery, in the wall fall indicates that by that time the collapse had already ended. A few sherds could, of course, have reached any layer in various ways, but we believe that the appearance of Early Roman sherds only, though few, within the wall fall is no accident, and together with the almost total lack of later finds indicates that during this time the wall continued to collapse.43

4.28 Stone vessels from room 8

Interestingly, another type of find, which is probably connected with this chapter in the site’s history, is the hiding burrow, connecting Cave D with Cave G (Chapter 1.3. Note that it is possible that there were other such burrows, perhaps in Cave I). The features of the hideout burrow, such as connecting already existing caves, a curve, change of levels, and blocking from the inside are all characteristics of such systems (Kloner and Tepper 1987:40–71; Kloner and Zissu 2003:182–183). Our tunnel is relatively short, but it still contains the main typical features. The hideout systems found in the Shephelah are dated to the first–second centuries CE, and they are attributed either to the Great Revolt, or, in most cases, to the Bar Kokhba revolt (Kloner and Zissu 2003:186–190). Our burrow should be later than the two caves it connected, and there is no reason to predate it to the Early Roman period, the usual date for this phenomenon. Most of the hiding complexes are related to nearby residences (Kloner and Zissu 2003:181). This could be also the case in Kh. er-Rasm. While Caves D and G should be attributed to the main period of use of the main building during the Hellenistic period (above), the burrow, which is clearly later, should be attributed to the more provisional presence in the Early Roman period, which is expressed in the scanty evidence in room 8. As for the debate surrounding the hiding tunnels and their exact dating, Kh. er-Rasm cannot solve this problem.

It is during the Early Roman period, moreover, that a phase of reoccupation took place at Kh. er-Rasm (phase 4). In one of the rooms (room 8; see Fig. 1.42), we uncovered a surface on which finds from this period were unearthed. The finds include some pottery, including a complete cooking pot (CP9), as well as two stone vessels typical of the period (Fig. 4.28; see Chapter 3.3).44 The Early Roman pottery was buried within what appeared like a wall fall composed of nari (though we tend to view it partially as a stoning mound). This heap of stones buried the Early Roman material, and the latter was lying on top of what turned out, when excavated, to be an earlier (Hellenistic) wall fall, composed also of many kirton blocks (Fig. 1.20). Stratigraphically, the Roman phase is clearly situated on top of the Hellenistic one, and the difference between the later phase and the Hellenistic occupation is also attested by the relative absence of kirton stones in the former. Unfortunately, the excavations ended before we could enlarge the exposed area to the north. The nature of the Roman occupation (phase 4) is unclear (see more below), but it was fairly small and concentrated in the western wing (where it appears that the settlers leveled the ground in the room(s) which they used).

The above mentioned pottery, stone vessels, hiding complex and coins all seem to be a result of this phase of reoccupation. The reoccupation probably took place during the first century CE, as attested by the coins of Agrippa I that were found not in situ, or perhaps slightly later, toward the early second century CE (which is the latest possible date for those finds).

A limited amount of Early Roman pottery was unearthed also in other parts of the structure (mainly topsoil), along with two coins of Agrippa I.

Interestingly, if the connection between stone vessels and a Jewish presence is substantiated (e.g., Magen 2002:162– 164; Amit, Seligman and Zilberbod 2001:102), then the settlers at this time were Jews. This is no surprise, since the Shephelah became Jewish following the Hasmonean conquest (see also Chapter 5.1). At any event, activity at the site at this time was quite limited, and its nature is unclear.

The fact that no pottery later than Hellenistic was found below the wall fall, and that only a few Roman sherds were found within the wall fall, indicates that the stratigraphy was basically correct, and moreover, that the wall fall was so massive that it appears to have limited the number of sherds that penetrated lower layers as a result of various processes (e.g., burrowing of rodents, worms, the operations of roots, etc.). 43 Only one late sherd was found within the falls (not their lower part). This is a very small quantity, and the sherd had probably reached the level through an animal burrow or some similar process. 44 This cooking pot is of the same type as two more complete cooking pots found in area C (below). The whole assemblage from this room and this phase is dated from the Herodian period up to the Bar-Kochva revolt (see chapter 2.2). 42

The stepped structure at area C probably belongs also to this phase (below). It is likely that the failure of either the 232

Avraham Faust and Adi Erlich

4.29 Plan and section of tower from Kh. er-Ras (courtesy of Prof. Shimon Dar)

first Jewish Revolt or the Bar Kokhba revolt put an end to this phase in the history of Kh. er-Rasm.

and Judea (Hirschfeld 2000:685–696, 709–720; Zissu 2001:260–262). Those towers are always part of the rural buildings, either in the corners or elsewhere (Dar 1986: figs. 2–18; Hirschfeld 2000: fig. 40). Moreover, the towers are often divided into rooms.

4.3.2 The Stepped Stone Structure (Area C): Function and Date The stepped stone structure, discussed in Chapter 1.2, is difficult to date. As mentioned earlier, we have referred to the possibility that it was built in the Hellenistic period, but noted that it is more likely that it was erected later.

A different type of “tower” is a much smaller structure, either square or round, which served as a field tower (Fig.  4.29). Many field towers dated to the Hellenistic period were found in Samaria (Dar 1986:88–125; Tal 2006:120). The field towers all have rooms on top, with walls, entrances, floors and roofing (Dar 1986:105–107). Although common in Samaria, the Hellenistic and Early Roman field towers appear also in other parts of the country, including Judea (Tal 2006:120).

The unique stepped structure raises questions regarding both its function and its date. Although the date should be addressed first, as it determines the place of the following information within the broader discussion of the site’s history (i.e., the present section of the report), we must begin with a discussion of the function, as it seems to have some bearing on the dating of the structure.

Other square towers, which are isolated buildings, are related to fortifications or military buildings. At Yoqne‘am, a square tower dated to the Late Hellenistic or Early Roman period was unearthed (Tal 2006:17–18; Ben-Tor, Avissar and Portugali 1996:11–12). Additional small isolated forts of the Hellenistic period were found in the Land of Israel, all subdivided into rooms, and serving as small forts rather than one solid tower (Tal 2006:155, 158, fig. 21).

The Function of the Structure The Structure as a Tower The structure at area C is unusual and raises questions regarding its nature. The first option we would like to consider is the use of the structure as an observation tower. Towers which served as observation points are common in fortified farms and manor houses of the Hellenistic and Early Roman periods. Frequently a square building of a farm or estate is accompanied by a tower, as attested in Samaria (Dar 1986:8–16, figs. 2–18; Tal 2006:121–125)

The structure in area C deviates from all the examples above, and is not suited to be an observation post for defensive purposes. First, in the abovementioned buildings the tower is either part of the building it defends, or located within its architectural boundaries, whereas in area C it is

233

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

4.30 Nefesh structure from Suweida in Syria (after Brünnow and Domaszewski 2004 [1909]:97–100

separate from the main building and situated about 25 m from it. It is too far to serve the building, but too close to serve as a field tower. Second, the location of area C is not the preferred spot for an observation post. The main building occupies the summit, while the structure at area C is lower. The view seen from it is only to the south and west, whereas the main building on the summit also faces northward and eastward. Third, the structure is relatively large and solid, requiring a great effort to construct. It seems to have required too much work for a simple tower in a small non-urban settlement. A much smaller shomera-like building would have been enough. The structure is not divided into spaces or rooms like both defensive towers and field towers. Last, the upright facing walls and the pyramid above it neither resemble the above examples nor suit the purpose of observation. There are no signs of wall lines on top of the upper surface of the pyramid. Climbing the structure is possible from the north and perhaps from the east as well, but the structure does not seem to be intended for climbing. For these reasons, we believe the structure to be something other than a tower or any other kind of building designated for observation, defense or agriculture.

observation point but rather as a structure meant to be seen. The unique shape of the structure leads us to the conclusion that it served as a funerary marker, nefesh, which is typical of the Hellenistic and Roman periods in Syria and the Land of Israel (Gawlikowski 1970; Triebel 2004; Hachlili 2005:339–353). Nefesh is a common term in Hebrew and other Semitic languages, referring to tomb markers, steles, towers, mausoleums, memorials and other funerary contexts (Avigad 1954:66–73). It is a common term in epigraphic and historical texts of the Hellenistic and Roman East (Triebel 2004:117–233). The nefesh structures are diverse in terms of shape, size and decoration, but their significance remains the same (Gawlikowski 1970:29). The types of tomb monuments vary geographically, for example, built towers typical of Palmyra (Will 1949; Gawlikowski 1970) and stone-cut monuments and façades characteristic to Nabatea (McKenzie 1990). Of all forms, the pyramidal nefesh was highly favored in the Land of Israel during the Hellenistic and Early Roman periods (Avigad 1954:131– 132; Triebel 2004). The pyramids were either hewn or built, real or engraved, square or conical, stepped or sloping in straight lines, on top of a tomb or without a tomb, serving merely as a memorial.

The Structure as a Funerary Marker Tomb Markers (nefesh, nefashot pl.) in the Levant

Pyramidal Nefesh Structures in Syria and Nabatea

As already mentioned above, the structure overlooks the south and the west, but it is also prominent and can be viewed from these directions from a great distance. Therefore, we believe that the building served not as an

The pyramid-like nefesh at Suweida in the Hauran, Syria (Fig. 4.30) is composed of a solid square building on top of two steps, decorated with Doric pilaster and relief of armors

234

Avraham Faust and Adi Erlich

in the intercolumniation, surmounted by a Doric frieze and cornice, and capped by a massive stepped pyramid which is only partially preserved (Brünnow and Domaszewski 2004 [1909]:97–100). This remarkable building, dated to the first century BCE or CE (Avigad 1954:67; Gawlikowski 1970:23), and has neither entrances nor tomb in it (Gawlikowski 1970:23; Triebel 2004:137). The term nefesh is mentioned in a bilingual Greek-Aramaic inscription inscribed on the monument, which was dedicated by a husband named Adinath to his wife Hamrath.

Jerusalem of the Late Second Temple period (see summary in Triebel 2004:75–81). The three built and stone-cut nefashot in the Kidron Valley are the most famous. The monument of Bnei Hazir is accompanied by a detailed inscription starting with the statement “This is the tomb and nefesh…,” distinguishing the grave from the nefesh. Indeed, the façade is composed of two architectural elements – a Doric façade and a missing nearby nefesh, which was restored by Avigad into a pyramid (Avigad 1954:73–78) but thought by Barag to be similar to a Nabatean tower tomb (Barag 2002:39–44). The two other nefashot in the Kidron Valley are capped by a pyramid (Tomb of Zecharia; Avigad 1954:79–90) or tholus (Yad Avshalom tomb; Avigad 1954:91–130). Other Jewish pyramid-like tomb markers in Jerusalem are capped by a pyramid, such as Jason tomb (Rahmani 1967), and the so-called tomb of the kings or tomb of Adiabne (Vincent 1954:347–361, pl. XCVII). All these are well-built tomb markers or memorials, with mixed Hellenistic and Egyptian architectural elements and decorations. The pyramid surfaces of the Jerusalem nefashot are flat, slanting upwards in straight or curved lines, unlike our stepped structure.

In Nabatea, nefesh representations which are engraved on rocks are portrayed as pyramids on top of square pedestals, bearing inscriptions on the pedestal (Avigad 1954:68–69; Triebel 2004: pls. XXXII–XXXVI; Hachlili 2005:350–351). These nefesh forms resemble obelisks rather than true pyramids. The typical Nabatean so-called Pylon tombs are square cut tombs crowned with one or two rows of crowsteps, each resembling a small pyramid (McKenzie 1990: pls. 10–13, 157a, 158a–b). The crowsteps are typical of Syrian architecture, and they are primarily related to temples, as in the Bell temple in Palmyra and elsewhere. Their depiction on tombs is part of the common concept of the tomb as a shrine for the dead. Crowsteps on tombs appeared at Amrit as early as the sixth century BCE (Gawlikowski 1970:12–15).45 Another type of Nabatean tombs, the step tomb, is reminiscent of a pyramid, but is an inverted pyramid, flanked by two side steps (McKenzie 1990: pls. 157b, 158c). So are the so-called Proto Hegr and Hegr types (McKenzie 1990: pl. 157). None of these Nabatean types resemble our built and stepped pyramid, but the idea and concept of a pyramid ascending on top of a grave or depicted in small scale, is similar.

Pyramidal Nefesh Structures in the Judean and Hebron Hills A structure both much more similar and much closer to Kh. er-Rasm, is located in Kh. Midras, about 4.5 km south-southeast of our site (Rachmani 1964:223–228). The nefesh of Kh. Midras (Fig. 4.32) is built above a tomb with a courtyard and an arched dromos. The upper structure is a square building measuring 10 by 10 m, with upright walls built of three courses with the pyramid on top. Unlike our structure, the one at Kh. Midras is built out of fine dressed stones, and in one side the upright wall ends with a cornice. The Kh. Midras structure is slightly bigger, more regular and more neatly built, but apart from these differences, the resemblance between the two structures is remarkable. Rahmani dated the complex of tomb and nefesh to the third century CE, but he does not exclude an earlier dating to the first or second centuries CE (Rahmani 1994:31, note 20).

A still different nefesh type in Nabatea, much closer in geographical terms, greatly resembles the pyramid at area C. The necropolis at Mampis (Kurnub) in the Negev, dated to the first–second centuries CE, yielded among other monuments a few solid structures of nefesh built of ashlar stones. In most cases their stones were robbed in later periods, except for tomb no. 119 which collapsed into its burial shaft and therefore was saved from robbers (Fig. 4.31). This tomb attests to the original form of similar structures as stepped pyramids (Negev 1971; 1986:78–79; 1993:892; Triebel 2004:113–115). Negev marked the spots in the necropolis where funerary meals took place (Negev 1986:76, fig. 39). We will return to this point below.

There are also artistic representations and graffiti of pyramidal nefashot coming from the region of the Shephelah and the Hebron region. An ornamented Jewish lamp from this region depicts a nefesh (Sussmann 1982:56–57, No. 60). Ossuaries with depictions of nefashot are attributed to the Hebron and Shephelah regions (Rahmani 1994:31). Two ossuaries in the collections of the State of Israel are decorated with pyramidal nefashot, one said to come from Jerusalem (Rahmani 1994:133, No. 231, pl. 33) and the other perhaps originating in the Hebron hills (Rahmani 1994:182–183, No. 473, pl. 70). The latter depicts a pair of pyramidal nefesh structures standing between two stylized trees, in a way that resembles the appearance of Area C nowadays (Fig. 1.67). Of course, the olive trees flanking the structure are recent ones, but the evidence from the site through the ages indicates the presence of olive trees nearby in many periods, including in antiquity (e.g., Liphschitz, Chapter 3.7, and Sasson, Chapter 3.12).

Pyramidal Nefesh Monuments in Judea The seven pyramids built by Simon the Hasmonean in Modi‘in to commemorate his family, as attested in 1 Macc. 13:27–29, are the earliest evidence of the use of such nefashot by Jews. Thereafter nefesh monuments occurred in Interestingly, Mckenzie (1990: 115) notes that in Medain Saleh, wherever there are inscriptions on tombs, the rather simple pylon tombs are inhabited by women, while other more complex and ornamented types are owned by men. However, the monumental stepped nefesh in Sueida discussed above is dedicated to a woman and yet it is quite splendid, and is even decorated with typical masculine sculptures of military equipment in bas relief (Gawlikowski 1970: 22-23). 45

235

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

4.31 Necropolis of Mampis with pyramidal nefesh (after Negev 1993: 892; courtesy of the Israel Exploration Society)

4.32 Nefesh structure at Kh. Midras

236

Avraham Faust and Adi Erlich

The lamp and ossuaries discussed here are portable objects, of uncertain source and provenance. But there is also insitu evidence for pyramidal tomb representations in the Shephelah itself, including graffiti of nefashot engraved on walls of Early Roman burial caves in the Beth Guvrin region (Triebel 2004: pls. XXIII–XXV). A loculi tomb in Kh. El-‘Ein (Fig. 4.33), located two km north-east of Beth Guvrin, is decorated with two graffiti of nefesh structures similar to those on ossuaries (Zissu 1999:86–87; Triebel 2004: PL. XXIII, Nos. 3–5). Zissu suggests that yet another graffito in a tomb at Iyye Nahash, 2.5 km east of Beth Guvrin, also depicts a nefesh, despite its schematic Λ shape (Zissu 2000). The foremost graffito in connection to the Kh. er-Rasm structure comes from Horbat Egoz, some 10 km south-southeast of Beth Guvrin. On one of the wall in the entrance hall to the cave, a schematic wreath was engraved, over two stepped pyramidal structures (Zissu 1997). The left structure is composed of alternating low and high steps, terminating at the top with a tower-like structure. This could explain the idea of the structure at area C: the high steps standing for the upright wall, and the tower-like structure which is missing from our building and could have stood on top of the flat upper surface which still remains. However, our structure is simpler, without alternating components. All these representations are Early Roman in date (first to early second centuries CE), but there are also two possible graffiti of nefashot in Early Hellenistic tombs (Erlich 2009:64): One in Maresha (Peters and Thiersch 1905:20, fig. 4) and the other in Za‛akuka, 6 km east of Maresha (Kloner, Regev and Rappaport 1992:29–31, fig. 8). However, in the Shephelah, like the Levant as a whole, although the pyramidal nefesh had roots in ancient times, it became common only during the Early Roman period.

structure in area C are the nearby pyramid at Kh. Midras and the Nabatean pyramids at Mampis. A major component which differentiates those parallels from our structure is the grave itself. The pyramid at Mampis was erected on a burial shaft, while the pyramid at Kh. Midras is adjacent to a burial cave below. We have no proof of a tomb related to the area C structure. However, we did not remove the structure, so there could be a tomb underneath (The GPR examination of the structure was inconclusive in this regard; see http://www.mnemotrix.com/geo/shepela.doc). Another scenario is of a tomb not directly beneath the pyramid but just as in Kh. Midras, a nearby burial cave, perhaps to the north of the structure, where we were limited in our excavation by a bush covering a large area and preventing us from digging. Working in a JNF park, we could not cut down trees; the latter could only be trimmed by the JNF inspectors. In such a case, the nefesh is a solid memorial for a tomb located nearby but not underneath, just as in the stepped nefesh at Suweida in the Hauran which is a funerary monument with no signs of an interior grave (as discussed above), or the Zecharia tomb in Jerusalem which is a solid structure lacking a tomb (Avigad 1954:87–90). In any event, although we did not find a tomb, the stepped structure at area C seems to have been a pyramidal nefesh. The Date of the Structure We shall now turn to the problematical matter of dating the structure. The major problem is that the structure has stood almost exposed from its construction until today. It was not covered by later phases or altered by modifications, and the only changes were some collapses and the modern section cut in its west side. The loci excavated on top of it and around it are all regarded as topsoil, and therefore are useless in terms of dating. Hence, the considerations for dating the structure are mainly stylistic, historical and circumstantial in a sense.

As shown above, the examples from Jerusalem and the Judean Shephelah are quite numerous. Most of the cases are attributed to Jews inhabiting Judea, apart from two uncertain graffiti in tombs in Hellenistic Idumea, and perhaps also the Kh. Midras pyramid if it is indeed dated to the third century (as Rahamani suggested at the time), but this dating is questioned. Triebel (2004) emphasizes the special meaning and use of pyramidal nefesh structures among the Jews during the Early Roman period. There is a view which relates the origin of the nefesh to the Nabatean realm, from which the Jews adopted this custom (Hachlili 2005:350). In this regard, Hellenistic Idumaea could have been the link between the Nabatean and Judean custom, as Idumaea was located between the two and had close relations with the former. Idumaea itself could have been the origin of the nefesh phenomenon, similar to the loculi tombs which were probably adopted by the Jews from the Idumaeans (Kloner and Zissu 2007:78–79). This would explain the rather frequent appearance of pyramidal nefashot in the Shephelah, from the Early Hellenistic until the Early Roman periods or even later, representing the transition from Idumaea to Judea.

The structure is accessible from the main building and maintains similar directions, so both were certainly related to one another. We will examine it in relation to the two relevant stages of use on the site: the major phase of the nearby main building during the Hellenistic period and the later use of the building during the Early Roman period, sometime during the first century or the beginning of the second century CE (above). The Duration of Pyramidal nefesh Structures in the Land of Israel: Origins and Influence As scholars have shown, nefesh monuments were common in the Levant from the Late Iron Age to the Roman Period (Gawlikowski 1970; Triebel 2004). We are concerned, however, with pyramidal nefesh structures, and more specifically with stepped pyramidal structures. The earliest pyramidal monument in the Land of Israel, dated to the Iron Age II, is the tomb of Pharaoh’s daughter in Silwan in Jerusalem (Avigad 1954:18–31; Ussishkin 1993:43–62; Triebel 2004:63–67). This monumental

A Nefesh With or Without a Tomb To sum up the evidence, the closest examples to the stepped

237

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

4.33 Graffito of nefesh from Kh. El-‘Ein (after Zissu 2005: 32; courtesy of the Israel Antiquities Authority)

238

Avraham Faust and Adi Erlich

tomb, with its straight pyramidal and Egyptian cornice, is thought to be the source of inspiration of the Second Temple period nefashot in the Kidron Valley (Ussishkin 1993:337). Another view which we mentioned above, attributes the origin of the Jewish nefashot to the Nabateans.

form of our structure as a chronological indication. The relatively provincial style of our structure may be attributed to the rustic setting, rather than to an early date. We shall now turn to examine the evidence for a later date of the structure, i.e., the stage of re-occupation in the main building during the Early Roman period (see above). As was demonstrated above, the closest parallels to our structure come from the Judean Shephelah and are dated to the Early Roman period. The Kh. Midras nefesh was thought to be from the third century CE, but as it was only surveyed and not excavated, there is no reason to post-date it to the second century CE (Rachmani 1964:228; 1994:31, note 20). The stepped pyramids at Mampis are also of similar date, as is the monumental structure at Suweida. All the examples of stepped pyramidal nefesh structures are dated from the first century BCE to the early second century CE, and in regional as well as ethnic terms, are a Judean and Jewish phenomenon (though not limited to them, see Triebel 2004).

When we examine the earliest possible date for our structure, we should take into consideration the beginning of the renewal of the nefesh phenomenon in the Hellenistic period. Berlin (2002) considers the Second Temple period pyramidal nefashot to be a legacy of the Hasmoneans, who were the first to display monumental tombs among Jews, as attested in the book of Maccabees (above). In her opinion, the Hasmonean kings, who wanted to become integrated into Hellenistic politics, were influenced by large pyramidal mausoleums in Asia Minor (Berlin 2002). The stepped pyramid at Belevi near Ephesos bears a resemblance to our structure, except for its monumentality and the surrounding columns which are absent in our modest structure (Praschniker and Theuer 1979; Hoepfner 1993). Still, the phenomenon of pyramidal structures should not be attributed to a Hellenistic innovation imported by the Hasmoneans, but should rather be viewed as a local tradition which was part of a widespread Eastern Hellenistic custom. The evidence of the eighth–sixth centuries BCE in Jerusalem, as well as in Phoenicia and Syria (Triebel 2004:63–71) shows that the Hasmoneans did not create a new practice, but rather followed a local tradition while demonstrating their integration into the Hellenistic culture. Therefore, our structure does not necessarily post-date the Hasmonean conquest of the site, and theoretically, it could belong to the main stage of the building during the Hellenistic period. If we take into consideration the graffiti in Early Hellenistic tombs in Idumaea mentioned above, the nefesh at Kh. er-Rasm could have been a slightly later example of the same tradition.

If the structure was erected during the Early Roman period, it would explain its proximity to the main building. During this period the occupation at the site was limited, as shown by the excavations of the main building (above), and activity at the site took place near and around area C, the caves and room No. 8 in the west side of the structure. If the structure was built around the first century CE, its location suits the axis of activity in this period. The last consideration regarding an Early Roman date is based on the two complete Early Roman cooking pots of the same type, found on top of the structure within the accumulation of debris. They are of the same type as the complete cooking pot found in area A of the main building (see Chapters 2.2 and 4.3.1). For some reason, these two pots were placed on the structure, and left there (probably broken) until it went out of use. There are also relatively more Early Roman sherds in the area, than is usual in the site (below).46

Another support for a Hellenistic date could be the rather simple shape of the structure, which differentiates it from the more complex and elaborate Early Roman pyramidal structures. But this argument is unconvincing, for more than one reason: first, there are two dressed stones near the structure, and the upright walls could have been coated with plaster. Second, there are also unadorned nefesh structures in the Roman period, as in Mampis. Not all the stepped pyramids were monumental mausoleums with architectural decoration such as the Belevi and Suweida monuments. The fine ashlar stones of the nearby Kh. Midras nefesh are actually in secondary use, as attested by the incorrect placing of the chiseled side of some stones (Rachmani 1964:226–227, pl. 22.4). Finally, although the masonry in Hellenistic monuments in the Land of Israel is indeed less monumental and is more provincial than masonry from Herodian times onward, it is inaccurate to draw a linear graph progressing from simplicity and ineptness to complexity and delicacy. The decorations of Nabateans tombs, for example, show an opposite trend, changing from ornate to simple throughout the first century CE, and simple as well as ornate types are found in proximity (McKenzie 1990:120). Therefore, we cannot rely on the rather simple

One can argue that the structure had existed for a long period, i.e., it was built in the Hellenistic period, and after it fell out of its original use people used it in another way and left two cooking pots there. Still, we would like to suggest that not only could the pots date the structure, but that they are related to the activity which took place in it. The two complete cooking pots are important as they are the only complete vessels found in the vicinity of the structure, and the fact that both date to the same period adds to their significance (they are actually of the same type – CP9). Moreover, only one additional complete vessel from this period (Early Roman) was unearthed at Kh. erRasm and this might hint that the structure was the focus of activity at that time (it should be noted that in addition to the two complete cooking pots, five additional sherds of Early Roman cooking pots were found, all belonging to Notably, the number of Iron Age sherds is also relatively large (hence supporting the suggestion, above chapter 4.1.5, that the main building was “cleaned” by Early Hellenistic settlers). 46

239

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

the same type, hence strengthening the importance of this find).47 Cooking pots are very common in burial sites. In Jewish tombs in Jerusalem, lamps, bottles and juglets, as well as cooking pots, are the most common finds (Kloner and Zissu 2007:125–129). Lamps were used for light, bottles and juglets for perfume, and one can only wonder regarding the use of the cooking pots. Kloner and Zissu maintain that funerary meals did not take place inside the tombs, so the pots served either for food for the dead, or for heating water to wash the bodies (Kloner and Zissu 2007:127–129; Hachlili 2005:381–383). In the present context, funerary meals are a reasonable assumption, as the cooking pots were found on the memorial and not in the tomb itself. Funerary meals were common next to the tombs and funerary markers at the necropolis of Mampis (Negev 1986:76, fig. 39). Cooking pots outside of domestic use are common in ritual usage, for example in the Hellenistic phase of the Sanctuary of Pan at Banias, where numerous cooking pots are related either to offerings or, as is more likely, to dining at the sanctuary (Berlin 1999:30–31). The two cooking pots could have been in use by the relatives who commemorated their beloved deceased, and were subsequently left there. Such a scenario, along with all the abovementioned considerations, may lead us to dating the structure to the Early Roman period, probably to the first century CE.

Idumaea to Jewish Judea. During the early Roman period the site was reoccupied in a limited way. Occupation was identified only in one room (room 8) in the main building, using its partially preserved outline. The Agrippa coins found nearby (out of context) hint at the date of this phase, during the first century CE. To the south of the main building a stepped pyramid-like structure was erected, probably to commemorate a deceased person. In both places, room 8 in the building and the stepped structure, the main (only?) type of ceramics found and dated just to the Early Roman period is of the same cooking pot. In room 8 the pot was found together with typical Judean stone vessels, while on the pyramid, two complete cooking pots were found. Perhaps the nature of the activity at the site was related to the stepped-stone structure which functioned as a grave marker, and perhaps the presence of the pot in room 8 was also connected to this monument. Room 8 could have been in use for the relatives visiting the tomb marker from time to time. The cooking pots could have served for meals or water. This limited activity may have been connected to a nearby site, perhaps Kh. Shiqlon which is situated just east of Kh. er-Rasm, where the evidence indicates the existence of a settlement during the Roman period (Dagan personal communication). The Early Roman phase at Kh. er-Rasm probably came to an end in the great revolt or the Bar Kokhba revolt, when the short hideout burrow was cut.48 Most of the finds can be dated both to the first and to the early second centuries (for the pottery, see Sandhaus, Chapter 2.2; for the stone vessels, see Peshin, Chapter 3.3, and also Magen 2002:162–164; Geva 2006b:193), and hence both revolts could have been responsible for the cessation of the activity at the site.

Area C: A Summary The structure in area C has a few distinctive features which turns it into a nefesh although no tomb has been revealed as yet. It is square and solid, has upright walls capped with a stepped pyramid, and is prominent (see reconstruction in Fig. 4.34). It resembles pyramidal nefesh structures in the Shephelah region, most of which are related to Early Roman Judea. Although it is possible that the structure was built during the Hellenistic period, in conjunction with the Hellenistic phase of the main structure, and that in the Early Roman period people used the structure for undetermined purposes, we tend to date it to the phase of reoccupation in the Early Roman period (architectural phase 4 of the main building). We conclude that this structure is also part of the Jewish – Judean phenomenon of nefesh structures. This funerary marker was erected within a rural area with agricultural and limited residential activities. An earlier date in the Hellenistic period is less plausible according to the given evidence.

Interestingly, while we cannot resolve the question as to when this Early Roman occupation ended, the ceramic finds at area C could lend some support to an earlier date, in the first century CE. Among the finds in this area are a number of sherds of R-JT3. This juglet can be dated from the end of the second century BCE to the first century CE (see Sandhaus, Chapter 2.2). The sherds could, therefore, belong to the last Hellenistic phase at the site (phase 2c). However, since many of the finds in area C belong to the Early Roman period, and the concentration of R-JT3 in this area is greater than expected, it is more plausible that the sherds reached this area when the stepped structure was built in the Early Roman period. These sherds therefore lend some support to an earlier date within this period for the finds, but are not conclusive enough to settle the issue.49

Summary – Khirbet er-Rasm in the Early Roman Period The Early Roman period in Kh. er-Rasm (phase 4) represents the transition of this region from Hellenistic 47 We have seen that all the loci in area C were mixed and should be treated as topsoil. Out of 27 identified indicative sherds (including the two complete cooking pots) that were unearthed on and near the building, 7 were of CP9! Not only is this a very high percentage, but those are practically the only sherds that can be attributed with certainty to the Early Roman period (3 other sherds, of JT3, can be dated to this period, but also to the late second century BCE, and hence are not indicative for our purposes; after all, such sherds were also found within the building and it is therefore difficult to know when the pieces unearthed in areas C were broken in the Late Hellenistic or Early Roman period)

The only finding near the entrance of the burrow, which was looted before the dig, was a wall fragment of an ossuary. 49 One should also take into consideration the possibility that the reoccupation started early, and the juglet under discussion represenst this earlier phase within the Early Roman period, that continued until the BarKokhba Revolt, some time after the juglet went out of use. While we find it less likely, this is still a possibility. 48

240

Avraham Faust and Adi Erlich

4.34 Isometric reconstruction of the nefesh at Kh. er-Rasm

241

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

4.35 Plough marks on W.41

4.3.3 Agricultural Activity, Firing Zone, and a Park: The Site to the Present

area in and around Kh. er-Rasm (Ajjur was also settled at the time, but Byzantine remains were more limited [Dagan in press], and the site is farther away from Kh. er-Rasm, and is therefore a less likely candidate). It is likely that the ruins were used as a resting place at the time. At any event, the scant remains from this period, including the little pottery and even the two coins, should probably be attributed to such activity. The finding spots of some of these finds are next to openings in Cave H. Perhaps the activity in the site was connected to the deserted cave (again, most late sherds were found on the surface).

We do not have any evidence for permanent human occupation after the first–second centuries CE. The findings from the site after the second century CE are very few, and include one casserole (R-CS3) of the fourth– fifth centuries CE that was found in Square I49 (L.165), and two coins of the same period, one of them found in the opening of the cave at Square J50. Sherds of two types of bowls, dated roughly to the same period (R-BL5, dated from about 200 to the first half of the sixth century CE, and R-BL6, dated from the late second–early third century to the sixth century CE) were also found. Additional finds date to even later periods. A sherd of a storage jar (R-SJ9) is dated to the sixth–seventh centuries, and a bowl (R-BL7) is dated to the Early Moslem period. A basin (R-BS1) is dated to the Umayyad–Abbasid periods, and a bowl (R-BL8) is dated to the ninth–tenth centuries. A jug (R-JG9) dated to the 13th–15th centuries CE was also unearthed. A coin of a similar date was found in the topsoil of Square H52 in the courtyard. Most of the finds were found on the surface or in the survey, and only a few were found in the excavations, almost exclusively in the topsoil and in mixed loci.

After the Byzantine period (beginning with the Early Moslem period) the intensity of settlement in the region declined (e.g., Dagan 2006:45*–46*), and it is likely that agriculture was less extensive at the time. The coin and the few sherds indicate limited human activity in or near the site resulting either from agricultural activity or perhaps even from some travelers passing through the region. Kh. er-Rasm was still part of the lands of another settlement, although it is not clear how it was used. On the basis of the survey data (Dagan in press) it is likely that the area belonged to the inhabitants of Kh. Ajjur, which seems to have been a settlement from the Early Moslem period onward.

It is likely that after the Early Roman period, the site was not used for habitation and served mainly for agriculture. On the basis of the available evidence from surveys, it appears that the nearest rural site during the Late Roman / Byzantine periods was Kh. Shiqlon (Dagan in press). The site is located only some 400 m from Kh. er-Rasm, and it is most likely that its inhabitants cultivated the agricultural

During the Ottoman period, which witnessed some settlement growth (initiated probably already in the Mamluk period; see Grossman 1994; Dagan 2006:46*), and on which we have more information, the site was part of the agricultural area of Ajjur. The available evidence suggests that Ajjur existed throughout the Ottoman period (Grossman 1994:195, map 15; see also Sasson, Chapter

242

Avraham Faust and Adi Erlich

3.12 and Dagan in press). While this was probably the situation already much earlier, it is clearly the situation in the early twentieth century when the area was mapped. It appears that during the Ottoman period the region was less extensively cultivated, as can be seen in Guérin’s (1869:103) description – part of the area was used for growing olives, but in others he reported Pistacia lentiscus – much of this area was probably, therefore, used for grazing. According to the British Mandate period maps, part of the area was used for growing olives, while other parts were used for cereals and for grazing (Sasson, Chapter 3.12).

Although the date cannot be determined, evidence for postHellenistic agricultural activity at the site can also be seen in the plough marks on the upper flat side of several kirton “blocks” in the upper courses of the kirton walls (Fig. 4.35). The nearby village of Ajjur was abandoned during Israel’s War of Independence (1948), and it appears that Kh. erRasm was later part of a military firing zone (many used cartridges were found during the excavations) until the area became a JNF park in the early 1990’s.

243

Part V:

Conclusions

Avraham Faust and Adi Erlich

5.1 The Hasmonean Policy toward the Gentile Population: The View from Khirbet er-Rasm

is rare in Jewish sites, as is that of decanters (Sandhaus, ibid). The oil lamps, on the other hand, are similar to those found in Jewish sites (Ambar-Armon, Chapter 2.3), but it is possible that the Idumaeans were similar in this respect to the Jews. At any event, the information is very limited, and with the lack of good comparative data these are only preliminary observations, and any further discussion would be too speculative.2 Only when data from more sites will be available, will we be able to compare and contrast various behavioral patterns in a way that might prove worthwhile. Currently, this is not the case. Under these circumstances, it is clear that we do not intend to identify the ethnic identity of Kh. er-Rasm’s inhabitants.

In previous parts of this report we preferred to stick to a more archaeological / anthropological analysis, and did not dwell on the historical importance of Kh. er-Rasm. This was due to the type of data available and the detailed archaeological information in our possession, which is ideal for archaeological discussion, as well as our inclination toward this kind of analysis. Moreover, the nature of such a small site, where no “history” is expected to have taken place leads to more archaeological (and less “historical”) lines of interpretation.

What we would like to establish, however, is that the site was within the boundaries of Idumaea, and that whatever the settlers’ identity was, they were probably not Jews, and were regarded by them as Gentiles. We would like to claim that the site shared the fate of non-Jewish (including Idumaean) sites.

Still, there is one historical event for which the detailed information from Kh. er-Rasm proved to be of utmost importance. We are referring, of course, to the circumstances surrounding the destruction of the site in the late second century BCE and the debate over the Hasmonean policy toward the Gentile population in the conquered areas.

The exact boundaries of Idumaea during the Hellenistic period are unclear, and the northern and western borders are often drawn differently on maps of the Persian and Hellenistic period (e.g., Aharoni et al. 2002:155, map 210; Kasher 1988:5, map 1 versus p. 45, map 7; and also Rainey and Notley 2006:280 versus p. 329). The Elah Valley is sometimes regarded as the boundary between Idumaea and Judea (e.g., Dagan 2000:229, 233), and Tel ‘Azekah is often regarded as one of the northern “sites” of Idumaea (Kokkinos 1998:60, but note that the map puts Tel ‘Azekah within Judea), which leaves Kh. er-Rasm, located south-southwest of Tel ‘Azekah, within Idumaea, though probably near the border. The region is even more clearly in Idumaea according to other interpretations (e.g., the maps in Rainey and Notley 2006:323, 329). It is quite obvious, therefore, that despite the limited information we possess, Kh. er-Rasm was, according to all these who discussed the boundaries of Idumaea, within its limits (Fig. 5.1).

5.1.1 Khirbet er-Rasm as an “Idumaean”, Non-Jewish Site The study of ethnic identity on the basis of the archaeological evidence is notoriously difficult (e.g., Shennan 1989; Jones 1997; Emberling 1997; Faust 2006b and many references). Ethnicity is subjective and in a continual process of change, negotiation and redefinition. It is, therefore, very difficult to identify ethnicity archaeologically – difficult, but not impossible. Ethnicity can be deciphered archaeologically either through the identification of ethnic markers (symbols) or through the identification of ethnically specific behavior (McGuire 1982; see also Faust 2006b). The former is arbitrary, and is almost impossible to identify without additional, mainly textual, knowledge, or extremely detailed knowledge of all facets of material culture and its symbolic and social importance. The second avenue of identification of ethnic groups on the basis of the material record – that of recognizing ethnically specific behavior – is more susceptible to archaeological analysis. Still, it requires some knowledge of other types of social behaviors, e.g. those relating to class and gender. Clearly, only very detailed information on the societies under discussion can enable us to tackle the question of ethnicity in the archaeological record.

The finds at Kh. er-Rasm, and here we refer mainly to the large assemblage of complete pottery, are very similar to those of Maresha. While this is by no mean evidence for the ethnic identity of the inhabitants, as the connection between material objects and identity is very complex (e.g., McGuire 1982; Jones 1997; Faust 2006b, and additional literature), it does indicate that the site was integrated within this region, and economically at least, was part of it.

Unfortunately, although the relatively detailed data from Kh. er-Rasm may look promising in this respect, the lack of any comparable information from other sites renders such an enterprise almost an impossibility at this stage.1 We should note, however, that in addition to the circumstantial evidence to be discussed below, there are some hints that indicate that the settlers were not Jews. The finding of table amphorae (Sandhaus, Chapter 2.2.1) seems to hint at the presence of non-Jews, as this finding

Moreover, the history of the site – which is the subject of this section – also indicates that it was not Jewish. The issue will be discussed at great length below. The following discussion is not dependent only on the results of excavations at Kh. er-Rasm, which is, as we shall see, just one site out of many, and theoretically could We should also add that according to most sources, the ethnicity of the settlers in northern Idumaea was not necessarily Idumaean, nor was it permanent (cf. Kokkinos 1998:90–92; Shatzman 2006:222), hence trying to prove it to be Idumaean might be a wrong path altogether. 2

We are referring to the second century BCE inhabitants, of course. For the identity of the settlers in the reoccupation phase (Early Roman period; phase 4 in the main building), see Chapter 4.3 and Chapter 5.2. 1

247

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

5.1 Map of Idumea during the Hellenistic period

have developed even without these excavations. But it is our excavations that were the trigger for the following analysis, and hence the importance of Kh. er-Rasm in this connection.

I. During the latter’s rule (134–104 BCE) Judea expanded significantly. Initially, it appears that John Hyrcanus had to acknowledge the limitation of his power in relation to the Selucids. Later, when the geo-political circumstances changed, he began to expand his kingdom. The expansion started in Transjordan, and later extended to Samaria and Idumaea. The conquest of Idumaea was accompanied by the conversion of the local population (below). All in all Judea expanded to most of the inner regions of the Land of Israel, i.e., the hilly and mountainous regions between the Negev and northern Samaria (Rappaport 1984:211–219; Kasher 1988:44-78; 1990:116-131). Hyrcanus also opened the way for the conquest of northern areas as well, and the expansion of Judea continued even after his death. Judas Aristobulus, Hyrcanus’s son, although he ruled for only a year, conquered the land of the Itureaans (Rappaport 1984:219–222; Kasher 1988:79-85; 1990:132-137). The expansion of Judea also continued during the time of Alexander Jannaeus, but this period is beyond the scope of the present discussion.

Historical Background The Hasmonean state was established following a long process that was initiated by the Hasmonean Revolt, led by Judah Maccabee, in 164 BCE. Jonathan, Judah’s brother, headed the revolt from the defeat at the battle of Elasa and until its final success. Simon, the last surviving son of Mattathias, replaced Jonathan after the latter was captured, and was the head of Judea for less than a decade (142–134 BCE). He fortified Judea and its borders, including the region of Lydda (Hadid), “cleansed” it of foreign armies and “opened” a gate to the Mediterranean (Jaffa). Simon laid the foundation for the Hasmonean state (Rappaport 1984:206– 211; Kasher 1988:42-43; 1990:105-115), but was murdered in 134 BCE, and was replaced by his son, John Hyrcanus

248

Avraham Faust and Adi Erlich

The Current Debate over Hasmonean Policy in the Conquered Areas

2002:9–10, note 11). This approach is supported by the archaeological evidence from Maresha, where the site was destroyed in 112/111 BCE (Kloner 1991; 2003:5-6, and see more below).

The Hasmonean treatment of the population of the conquered regions has received a great deal of scholarly attention (e.g., Schwartz 1991; 1993a; Cohen 1999:110– 119; Safrai 2000; Shatzman 2003; 2005; Ronen 2003, and more). On the basis of the written sources (e.g., Maccabees 1; Maccabees 2; the Book of Jubilees, as well as Flavius Josephus and Greek writers), many scholars believed that the relations between Jews and non-Jews were hostile, and that whenever they could, the Hasmoneans inflicted a heavy blow on the Gentile settlement.

Such interpretation was very common in the past, and as we have seen is still quite common today. However, this view of the Hasmonean policy was challenged a long time ago, and today there is another prevailing interpretation. Already in the 1960’s, Rappaport (1967 [1993]) suggested that the population in the region should be divided into a number of groups: nomad tribes, mainly in the south, the inhabitants of the various poleis, and the local, “oriental” population, i.e., the population that did not become part of the poleis. Rappaport, and following him others (e.g., Ronen 2003; Cohen 1999:116–117; see also Schwartz 1993a), stressed the contrast between the urban, Hellenized population, and the rural “oriental” population, as well as the similarity between the latter and the Jewish population.3

According to Shatzman, who recently published a thorough article summarizing the relationship between Jews and Gentiles during the first generations of the Hasmonean rule, the latter initially practiced a severe policy and tended to kill Gentiles (e.g., Shatzman 2003:149, 156). Jonathan and Simon, on the other hand, switched to a “softer” policy, and refrained for various reasons (perhaps resulting from Judea’s international relations) from killing or removing the local population, and in many cases were satisfied with only removing symbols of the Gentile cult (ibid.). John Hyrcanus changed this policy, and initiated a policy of forced conversion (ibid., 149–150). This aggressive policy might have been a result, according to Shatzman, of an attempt to implement a policy similar to that of Judas (perhaps as a result of criticism that was leveled at his predecessors, who abandoned it), as well as from the State of Judea’s international relations, which made it possible (ibid., 157). In light of the archaeological evidence, and especially the finds from Maresha, Shatzman believes that John Hyrcanus’ expansionist policy was carried out mainly toward the end of his reign, from 113 BCE onward (see already Kloner 1992; Barag 1992-1993). Shatzman points out a change in policy at the time, when instead of expelling the Gentile population, the Jews converted them, even by force.

Ronen (2003) developed Rappaport’s approach. Following Maresha’s excavator (e.g., Kloner 2003:15–16) he showed that in many important details there is similarity between Jewish and Idumaean customs. For example, an early miqweh-like installation was discovered in Maresha (Kloner 1991:81; Ronen 2003:127), as was an ostracon with an early Idumaean marriage contract, which closely resembles the later Jewish contracts (Ronen, ibid.; Eshel and Kloner 1996). Another connection between Jews and Idumaeans can be seen in burial customs. The burial in loculi caves, which was typical in the Maresha region during the third and second centuries BCE (Kloner 1991; 2003:21–30), continued to be practiced in Judea in the later Hellenistic and Early Roman periods, and some see Idumaea as the link between early Hellenistic Alexandria and Second Temple Jerusalem (Oren and Rappaport 1984:149–151; Kloner and Zissu 2007:78–79). Circumcision is also attributed to the Idumaeans (Kasher 1988:57–59), and some scholars point out that since the Idumaeans were circumcised (Cohen 1999:115–116), forced circumcision (as described by Ptolemy) was not really necessary.4

As far as the Idumaeans are concerned, the historical sources appear to be contradictory. While some (e.g., Strabo) might suggest reconciliation between Jews and Gentiles and a willingness to convert, others (Flavius Josephus and an unknown source named Ptolemy) speak of confrontation and forced conversion (Cohen 1999:110–116). Shatzman (2003) recently published a detailed analysis of the three sources, and concluded that Josephus and Ptolemy are more reliable than Strabo. He suggests that the testimony of the latter is not connected with the Hasmonean expansion and is anachronistic and irrelevant in the present context, in contrast to the other two sources, which speak of various forms of forced conversion and seem to be much more grounded in the historical and archaeological reality of the time (see more below). Kokkinos (1998:90–92), like Shatzman, tends to accept the testimony of Josephus and Ptolemy, and not that of Strabo, and claims that many of the inhabitants of the northern and western parts of Idumaea were Hellenized Phoenicians who were indeed forced to convert to Judaism. Goodman (1994:75–76), too, gives more credit to Josephus’ description (see also Bar Kochva

In light of the similarity in various practices between the two groups, Ronen suggested that the rural Idumaean population was very similar to the Jewish one. The mere fact that the conversion succeeded so easily, and that after There are also some scholars who believe that the Idumaeans were not a real “people” (or “ethnos”), but rather a Jewish (or Judean) population which separated itself from the Jewish population during the Persian period, and developed a different identity (Kochman 1981). Kochman also claimed that the story about the conversion of the Idumaeans is fictitious, and that it was simply the reunification of the latter with the Jews. This suggestion was justly rejected by Ronen (2003:126), who showed that despite the similarities between Jews and Idumaeans (below), the finds at Maresha as well as textual evidence indicate that the Idumaeans were a distinct ethnic group. We should note that Kochman’s discussion of the Idumaeans’ ethnicity is very problematic and is full of internal contradictions. Whatever the origin of the Idumaeans is, if and when they saw themselves as a distinct group, they were a distinct ethnic group. 4 Note that according to Shatzman (2005:222) and Kokkinos (1998:90– 92) many of Idumaea’s inhabitants were not circumcised Idumaeans but rather Phoenicians and other Hellenized peoples who did not practice circumcision. 3

249

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

a relatively short period the entire population of Idumaea became Jewish, is also used to prove the similarity between the two groups (e.g., Rappaport 1984:215, 217–218). There was, according to this view, no hostility between the groups, and Ronen (2003:129) explains the destruction of Maresha “by the gap between the rural sector and the Hellenized urban population”. He believes that the Hasmoneans fought only with the latter, and the people of Maresha preferred to go into exile, probably to Egypt. Cohen also claims that the villages of Idumaea were in alliance with the Hasmoneans against the exploitation by the Hellenized cities (Cohen 1999:116–117). Kasher agrees with this view, and stresses the long affinity between the oriental, rural societies of the Jews and the Idumaeans, who struggled against the urban, Hellenized Phoenician society. He believes that the textual description of hostilities are exaggerated, and in the areas that were annexed by the Hasmoneans there was a process of conversion willingly, in continuation of the long tradition of cooperation between Jews and Idumaeans (Kasher 1988; 2002:169–170).

policy in Idumaea, attempted to harmonize the historical sources (ibid., 116), and claimed that the evidence for destruction refers to the urban sector, while the evidence for willing conversion of the Idumaeans should be understood as reflecting the reality in the rural sector. The latter, Cohen suggested, joined the Jews willingly, and even on their own initiative (ibid., 118). Safrai (2000:71–72) offered a somewhat different interpretation. He agreed that some of the descriptions of the process are exaggerated, but claimed that one should not speak of a Hasmonean “soft” policy, and in his words (p. 72 – our translation): “In summary, it is difficult to speak of ‘conversion’, let alone forced conversion, but neither of a policy of mercy and cooperation. The Hellenistic centers and pagan temples in those regions were destroyed and the religious tolerance of the Hasmoneans was very limited”. Focusing the Problem What we have seen above are two distinct approaches, and although many nuances can be identified, there are basically two different reconstructions of the Hasmonean conquest. Did the Hasmoneans harm the non-urban Gentile population, or did the rural Gentiles, who were culturally similar to the Jews, gladly accept the conquest, and perhaps even initiate it (and were, therefore, not harmed by it)? It should be noted that although Samaria and Idumaea are two distinct regions, many scholars discussed both, and even if the discussion concentrated on one region, evidence were brought from the other (e.g., Schwartz 1991; 1993a; Cohen 1999, and many others).

Rappaport (1984:214–215; our translation) summarized the conquest of Idumaea in the following words: The conquest of Idumaea was accompanied by the conversion of its inhabitants, and this is the first time that such a practice is reported by a Hasmonean ruler. It should be noted that the source stresses the conquest of Adorah and Maresha separately from the subjugation the Idumaeans in general, and this is in line with the fact that those two Hellenistic cities and the Idumaeans were not the same ethnically and culturally… it appears that while Adorah and Maresha opposed the Jewish occupation of Idumaea, the latter [the occupation] was not accompanied by too much hostility on the part of the Idumaean-Semitic population, and perhaps was even accepted with some support. The conquest of Idumaea by John was not a conquest from any point of view, but rather its liberation from the rule of the two Hellenistic cities: Adorah and Maresha.

As strange as it may seem, although the situation in the non-Jewish rural sector lies at the heart of the debate (as there is no debate that the few cities were destroyed), no scholar has treated the finds from this sector in detail. Safrai (2000) is a notable exception, but even he did not discuss the excavated sites.5 The entire discussion was based on assumptions regarding the possible Hasmonean policy and the logic behind it, on the basis of the few and seemingly contradictory textual sources, and on an attempt to refer to the findings in the urban sector (which, in the case of Idumaea, for example, consists of the evidence from Maresha only).

Schwartz (1991) also stressed the distance between rhetoric and practice, and claimed that relations between Jews and Gentiles were not as hostile as the impression given by some of the sources. In an expanded discussion of JewishSamaritan relations, Schwartz (1993a) claimed that the Hasmonean expansion to this region was accompanied by destruction of the centers of the (Samaritan) reformist elite, and speculated that not only that this destruction did not harm the majority of the Samaritan population, but that it was to their satisfaction (Schwartz 1993a:17–18). Schwartz pointed out that the archaeological evidence for destruction is concentrated at Mount Gerizim and at Shechem, and not in the rural sector (ibid., 18), and suggested that the Hasmoneans applied different policies toward the rural and urban sectors, and did not harm the former (ibid., 20–21, note 9).

One has to admit that this problem is symptomatic of the archaeology of ancient Israel. The rural sector receives little serious attention, rural sites are excavated almost only in salvage excavations, and in many cases are only partially published (see London 1989; Faust 1995; 2000). This is despite their great importance for our ability to reconstruct the settlement history of the Land of Israel (e.g., Faust and Safrai 2005). We believe that the finds from Kh. er-Rasm, along with additional small sites that were excavated over the years, Other scholars, e.g., Schwartz, also appear to have referred to the results of the surveys (not excavated rural sites), but less systematically. 5

Cohen (1999:110–119), in his discussion of the Hasmonean

250

Avraham Faust and Adi Erlich

5.2 Map of the Land of Israel with sites mentioned in discussion

Additional non-urban sites that were excavated in Idumaea include, in addition to Tel Beersheba and Lachish that will be discussed below, the following:

hold the key to solving the debate over Hasmonean policy in Idumaea and Samaria. 5.1.2 Khirbet er-Rasm as Representative of the Situation in the Non-Urban Sector

Tel ‘Arad: a Hellenistic fort was discovered at Tel ‘Arad. According to Herzog the finds from the Hellenistic period include pottery, ostraca and coins which are typical of the third–second centuries BCE (Herzog 1997:250) and it appears that the fort did not survive the Hasmonean conquest, and was abandoned until the Roman period (Herzog 1997:251).

We believe that the finds at Kh. er-Rasm are representative of the situation in Idumaea. We have seen that the site, which was probably a country inn at the time, was destroyed in the late second century BCE, and this seems to reflect on the Hasmonean policy in the region (for a map of sites discussed in this Chapter, see Fig. 5.2).

251

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

Kh. ‛Uza: the Iron Age fort at the site was abandoned in the early sixth century BCE, and after a gap of 400 years the site was resettled in the late third century. The settlement came to an end toward the close of the second century, and the excavator attributed the Hellenistic fortress to the time of the Seleucids (Beit Arieh 1993a:1496-1497; Fischer and Tal 2007; see also Tal 2006:154).

of the second century BCE or toward its end, and some of the excavators have explicitly attributed the abandonment/ destruction to Hasmonean activity. This dating is supported both by the ceramic evidence and the coins unearthed. In Tirat Yehuda, for example, the excavators (Yeivin and Edelstein 1970:67) wrote, when referring to the Hellenistic phase, that the farmstead was (re)established “at the end of the fourth or the beginning of the third century BCE, existed for about 150 years and was destroyed between 180–140 BCE, when the Maccabean war started. It is possible that the destruction of the farmstead was, as is evident by the thick burnt layer on top of the ruins, connected with the beginning of this war, that perhaps included raids on small and unfavorable places which were not fortified enough and not central, though important for the hostile government”. It should be stressed that the coins hint at a date closer to the end of this period (Rahmani 1970:69).

Another settlement which should be mentioned, although it is possible that it was an urban one, is Tel ‛Ira: the excavator believes that during the Hellenistic period the site was of urban nature, and states that the finds include pottery which is typical of the fourth–second centuries BCE, and coins from the time of Ptolemy II, Antiochus III, Antiochus IV and John Hyrcanus I (Beit Arieh 1993b:645). The findings in all those sites, most of which were not urban (though not typically rural), indicate that settlement had ceased to exist by the end of the second century BCE. The partial publication does not allow us to determine the nature of the destruction or abandonment, but the similarity to Kh. er-Rasm is striking, and should be addressed.

At Mazor the coins indicate that the farmstead was destroyed slightly later, in the time of John Hyrcanus I. After describing the finds, the excavators (Amit and Zilberbod 1998:63) wrote: “A thick layer of ash was recorded in many of the farmhouse rooms, and especially in the storerooms uncovered in Areas S2–3. Based on the recovered material, the construction of the farm was dated to the early second century BCE and its destruction to the late second century BCE, probably during the reign of John Hyrakanos”.

Moreover, examining the situation in other regions reinforces the identified trend, and clearly shows that many rural sites – practically all the excavated rural sites outside of Judea proper – were harmed during the Hasmonean conquest (see Fig. 5.2). Farmsteads in Western Samaria

As we have just noted, there are many other sites in which the finds indicate that they were destroyed or abandoned in the second century BCE, sometimes after centuries of occupation. Those also include the village at Rosh Ha‛ayin (Avner Levy and Torge 1999), Shoham bypass (Dahari and ‘Ad 2000:79–80 note that during the Hellenistic period a fort was built on top of the earlier site; ibid., 80), probably the farmstead at Ofarim (Riklin 1993), the settlement at Qula (Avissar and Shabu 1998) and two farmsteads at Kh. Burnat (Torge, Personal communication).

Relative to other rural sites, the farmsteads in western Samaria have received a great deal of scholarly attention. The phenomenon was identified during the Sharon survey of the 1970s, and it received a detailed treatment by Israel Finkelstein (1978; 1981). Later, many other farmsteads were surveyed (see also Kochavi and Beit Arieh 1994; Gophna and Beit Arieh 1997), and many were excavated in salvage excavations that accompanied the development in the area of Rosh Ha‛ayin, El‛ad, Shoham, and the Trans-Israel Highway (Highway 6). Among the excavated farmsteads with Hellenistic remains one can count the farms at Mazor (Amit and Zilberbod 1998), Tirat Yehuda (Yeivin and Edlestein 1970), and Ofarim (Riklin 1993), as well as the two farmsteads at Kh. Burnat west (Torge, personal communication) and the rural sites at Rosh Ha‛ayin (AvnerLevy and Torge 1999), at Qula (Avisar and Shabo 1998), and at the Shoham bypass (Dahari and ‘Ad 2000).

Finkelstein (1981:347), who was the first to identify, survey and describe the phenomenon, wrote that “it is not difficult to point to the cause for the abandonment of the sites: the Hasmonean revolts and the instability which ensued throughout the country in general, and in the discussed region in particular, from the mid-second century BCE onward”. This conclusion was based on the survey, but is supported today by the finds from the excavations in many sites in the region.

Many studies have been devoted to the farmsteads phenomenon, concluding that it began during the eighth century BCE. While there is a debate regarding the exact circumstances of its beginning (Finkelstein 1978; 1981; Eitam 1980; Dar 1982; Faust 1995; 2006a), we are concerned with its end. What is important is that the phenomenon of farmsteads (and most of the sites) continued until the Hellenistic period.

It appears that the farmsteads were indeed damaged during the years in which the area was encroached on by the Hasmoneans, until they came to an end during the rule of John Hyrcanus I. Even if the process lasted dozens of years, and even if there is some mistake in the exact dating of some of the sites, it is clear that the Hasmonean expansion brought an end to this old and well-established rural settlement system (see also Berlin 1997b:28; Safrai 1994:190).

The archaeological evidence indicates very clearly that most of the farmsteads were abandoned during the middle

252

Avraham Faust and Adi Erlich

It must be stressed that during the second century BCE new settlements were established in this region, but they differed both in form and distribution (e.g., ‘Ad 2000:52). They, like the village that was excavated recently below Tel Bareqet by David Amit, Hagit Torge and Peter Gendelman (2008), existed until the first and second centuries CE, when they were destroyed in the Bar Kokhba revolt (e.g., Kloner and Zissu 2003; ‘Ad 2000:57; see also the changes that took place at the Shoham bypass, Dahari and ‘Ad 2000, and the erection of the new structure at Ofarim, Riklin 1993). On the basis of the available evidence, it is clear that the settlers in the new settlements were Jewish, and the new system replaced the older one which was destroyed by the Hasmoneans during their expansion.6

above, and in which the data is more comprehensive. The evidence from Kh. es-Shuhara strengthens the overall picture of destruction and abandonment in the non-urban sector in tandem with the expansion of the Hasmoneans. An additional site which supports the presented Scenario is Nahf. Three small areas were excavated in this site, and the excavator concluded (Smithline 2008:99) that “(T) he ceramic assemblage, in conjunction with the stamped Rhodian amphora handle, indicates that the Hellenistic settlement ceased in the second half of the second century BCE. This would appear to coincide with the Hasmonean conquest of the Galilee”. The excavator believes that the inhabitants of the settlement at the time were pagans. The cessation of the Hellenistic settlement was followed by a settlement gap which lasted until the second or third centuries CE (ibid.).

The evidence from western Samaria, which relates to many sites which are clearly of a rural nature, seems to correspond with that from Idumaea (although the nature of the sites there is less clear, and most were not rural, but only small sites in the countryside).

Temples in the Countryside We have examined above the finds in rural (and other nonurban) sites in pagan areas during the Hellenistic period, and we have seen many sites in which destruction and abandonment were attributed to the Hasmoneans. Below we will examine the finds in a number of structures, which although not rural in nature were located within the countryside, and seem to have served it. We refer to the non-urban temples from the second century BCE that were destroyed by the Hasmoneans (Kokkinos 1998:92). Their destruction by the Hasmoneans is indeed natural and expected, but we believe that their destruction was harmful to the entire rural settlement system that surrounded them, and the destruction, therefore, reflected on the reality in the countryside as a whole.

Additional Excavations Recently, an additional salvage excavation, which could shed some light on the situation in the rural sector at the time, was published. Kh. es-Shuhara is located in the Upper Galilee, between Bar‛am and Yir‛on, near the Lebanese border. This was a farmstead or a small village from the Hellenistic period, in which, according to the excavators and on the basis of the finds, a non-Jewish population dwelled (Aviam and Amitai 2002:130–132). The excavators identified two distinct destruction layers from the Hellenistic period. The first was attributed to the struggle for power within the Seleucid kingdom, but the final destruction of the site was attributed to “the struggle between the Hasmonean state and the Seleucids, and the Hasmonean conquests in the Upper Galilee” (ibid., 130). After discussing the finds at the site and ceramic changes, they add that “this change is a consequence of the Hasmonean conquest, a severe blow to the Gentile settlements and the transfer of Jewish population from Judaea to the Galilee, and perhaps also conversion processes of small groups” (ibid., 132).

Two Hellenistic temples were found in the southern part of the country, within the area of Idumaea. The more southern one is the temple at Tel Beersheba, which was built near the fortress that existed at the site (Herzog 1993:142; Derfler 1984). The conquest of the temple and the removal of all its pagan symbols, which included various offerings (Giveon 1973), is attributed to John Hyrcanus I (Derfler 1984:188–193).7 The other temple from the second century BCE was found at Lachish, and the destruction was dated to about 150 BCE (Aharoni 1975:3–5; Tal 2006:68–69). Aharoni’s dating, however, is not based upon the evidence, and given the site’s proximity to Maresha it is more likely that the temple, built in the early second century BCE, was destroyed toward the end of this century. This is not the place to discuss the nature of the cult at the site, which Aharoni viewed as Jewish and others as pagan (Aharoni 1975:7–11), but it appears that the Hasmoneans destroyed this rural temple. The connection between Lachish and Maresha is also expressed in two identical pottery figurines, made from the same mould, that were found in both sites, and which belong to a unique type, with no additional parallels (Aharoni 1975: pl. 2:18; Erlich 2002:10–11, fig.

The finds are partial, and are open to different interpretations. Danny Syon (2002), for example, in light of his analysis of the coins unearthed at the site, believes that after the destruction of the Early Hellenistic site, the site was resettled by Jews in the Hasmonean period (the second phase). He suggests that the destruction of the site may have been related to other struggles which were not connected with Judea, although he does not rule out the possibility that the Hasmoneans were, after all, responsible for the destruction (ibid., 126*). As we have noted, the finds at Kh. es-Shuhara are partial, and it can be claimed that the interpretation is based to a large extent on the historical sources. Still, it appears that the site fits in with other excavated rural sites discussed

It appears that the second stage, which was attributed by Derfler to Jewish activity, was actually also pagan, and ceased as a result of the Hasmonean conquest (Tal 2006:70–71). 7

A new settlement system can be identified in other regions as well (see for example the site at Yad Binyamin; Weksler-Bdolah 2000). 6

253

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

fate of settlement during this period. One has to recourse, therefore, to the results of excavations.10

12; Erlich and Kloner 2008:43–45, pl. 24, no. 130). We believe that Maresha was a pagan urban site in Idumaea while Lachish was a pagan rural cultic site in this region (the finds in the latter site include only the temple and adjacent building), and both shared the same fate.

Intermediate Summary From the above it is quite clear that the settlement in the non-urban sector of the regions conquered by the Hasmonean suffered a major blow precisely at the time of this conquest. One can debate the exact reasons and agents of destruction in one site or another, as well as the exact dating, but possible modifications will not change the overall picture. The countryside was devastated in the second half of the second century BCE, in parallel with the Hasmonean expansion. It appears that the connection between the two, especially in light of the historical evidence, is clear and straightforward. If someone wishes to claim that it was not the Hasmoneans who were responsible for the destruction, that person bears the burden of proof.

Another temple which existed throughout the Persian and Hellenistic periods, and was damaged by the Hasmoneans, is the shrine that was excavated at Mizpe Yammim, in the Upper Galilee and in which evidence for the Phoenician cult was unearthed (Frenkel 1997). This temple, just like those in the south, was located in the countryside and served a local cult, which is represented in the situla dedicated to Astarte (Frankel and Ventura 1998). The destruction of the temples in the countryside could not have been viewed favourably by the local population, and the lack of any evidence for re-use in these cultic places can be seen as an evidence for the dramatic change this population underwent; whether this change involved total acceptance of the Jewish culture, or physical destruction of the settlements and the death or emigration of their inhabitants. The total destruction of the temples, and the cessation of their use, are evidence of the trauma experienced by the countryside and the rural sector. This strengthens Shatzman’s (2005:236; out translation) view regarding forced conversion in Idumaea: “… according to the available evidence it should have been expected that the population of Idumaea, both the Hellenized in the urban sector, as well as the rural one, would not have freely abandoned the traditional religion and cult and accepted the religion of its antagonists”.

There are two additional issues which strengthen the hypothesis which views the Hasmoneans as the agents of destruction. First of all, the fact that the previous, preHellenistic settlement system is replaced by a new one (see, e.g., in western Samaria, above), indicates that the change that accompanied the conquest was drastic. In many cases the “new” system was damaged, and even destroyed during the Jewish revolts, and this strengthens the identification of the “new” system as Jewish. Moreover, while the pattern of destruction in the nonJewish rural sector outside of Judea is clear, the situation within Judea is completely different, and this also supports the view that the cause for the change was the Hasmonean policy toward the non-Jewish population. In the following, we will, therefore, discuss the rural settlement within Judea.

A Note on the Identification of the Phenomenon in Surveys

5.1.3 Rural Settlement in Judea in the Second Century BCE

So far we have discussed only the information from excavations in non-urban sites, and not the results of the many surveys conducted throughout the country over the years.8 This is due to several factors. First of all, we should note that surveys are far less reliable for the purpose of examining settlement continuity than are excavations, including salvage excavations. When the results of many excavations are available, they should be preferred over the results of surveys (see also Faust and Safrai 2005). Second, in many areas, including those discussed here, the Hellenistic period was reported in the surveys as a single period, without any subdivision into the Early Hellenistic and Late Hellenistic (or Hasmonean) periods9 (this distinction is even more difficult, theoretically, in the regions of Idumaea and Samaria, since the Hasmonean period began only after it started in the heartland of Judea). The surveys of Lod and Rosh Ha‘ayin (Kochavi and BeitArieh 1994; Gophna and Beit Arieh 1997), as well as the Shephelah survey (Dagan 2000) treated the entire period as one unit, and the results cannot be used to determine the

Examining the available evidence regarding rural settlement in Judea during the second century BCE reveals a clear pattern of continuity. A number of rural sites from the Hellenistic period were excavated over the years in Judea, including the following: Horvat ‘Etri: the site was excavated by Boaz Zissu and Amir Ganor, and the settlement existed continuously from the late Persian period through the Hellenistic period, until its destruction in Roman times. The excavators believe that during the Hellenistic period the site was Jewish, and that some of the miqwa’ot (ritual baths) were hewn in this time (Zissu and Ganor 2002:19–20). Kh. um el-‘Umdan: The excavators, Alexander On and Shlomit Weksler-Bdolah (2005) view the site as a Jewish village that was “established in the pre-Hasmonean 10 It should be stressed that this is not a criticism of the results of the surveys; identifying sub-periods in surveys is sometimes almost impossible. The limitations are built-in in the surveys, and we should be aware of them.

For a preliminary attempt to use the surveys for this purpose, see Safrai 2000. 9 See Chapter 4.2.4. 8

254

Avraham Faust and Adi Erlich

5.1.4 Discussion and Conclusions

Hellenistic period, by the end of the third century or beginning of the second century BCE, and existed continuously during the Hellenistic, Hasmonean and Early Roman periods” (On and Weksler-Bdolah 2005:107). Significant remains from the Early Hellenistic period were unearthed, indicating that the settlement grew and developed in the Hasmonean period (ibid.:108–111).

We have seen above that the available evidence show that the non-Jewish rural sector was severely damaged during the Hasmonean conquest. Admittedly, the finds do not cover all the relevant regions, but with the lack of any evidence for continuity in the rural sector, we believe that the suggestion that this sector was not damaged, is very unlikely and is totally lacking archaeological support. The available archaeological evidence, even if somewhat partial (and all in all we have discussed almost 20 excavated non-urban sites), clearly shows that the non-Jewish rural settlements were damaged in the second half of the second century, while sites in Judea show clear continuity at the time. Even if we cannot be certain as to the exact date of destruction of each site, nor regarding the precise causes, the overall pattern is quite straightforward, and indicates that the Hasmonean expansion caused serious destruction in non-Jewish settlements, urban and rural alike. It is not impossible that in some micro-regions the pattern was different, but the burden of proof lies with those who believe this was the case. Before such proof is presented, we must assume the damage was quite widespread.14

Har Adar: excavations at Har Adar exposed a number of structures by Michael Dadon. The structure in area B was built in the early second century BCE and served until the first half of the first century BCE (Dadon 1997:73), i.e., the structure existed throughout the turmoil of the late second century BCE.11 Qalandia: The site was established in the late third century BCE. It was originally used only seasonally. The site’s main period of occupation was during the second century BCE, and it was probably destroyed in the first Jewish revolt (Magen 2004:84). Kh. Kabar: Salvage excavations conducted at a structure about 200 m north of Kh. Kabar revealed remains from the Persian period to the end of the Hasmonean period, i.e., the first century BCE (Baruch 2006:49*–56*).

The fact that the later settlement system, that of the Hasmonean period, was different, strengthens the view that the agents of destruction were indeed the Hasmoneans.

Ramat Beth Shemesh: The site was identified during salvage excavations conducted by Emanuel Eisenberg at Ramat Beth Shemesh. The structure was probably established sometime during the first half of the second century BCE and continued to exist long after the middle or end of the century (not published; Eisenberg personal communication).12

We should also stress the different fate of the rural settlements in Judea, where we have identified impressive continuity (and growth) from the Hellenistic to the Hasmonean periods, and actually until the Jewish revolts. The data from Judea can serve as a control group, which shows the importance of the data from other regions.

The sites described above show very clearly that the settlement pattern in Judea differs greatly from that in the non-Jewish regions. Sites in Judea established in the Early Hellenistic period existed also after the second century BCE, and were usually destroyed only during the Jewish revolts.

The evidence does not indicate full destruction. The gaps in our data require caution. Moreover, the suggestion of religious incorporation between Jews and Idumaean has much appeal. Rappaport’s suggestion (1984:217; our translation) that not only there were Jews in the conquered regions, but even the non-Jewish population – those “Aramaic speaking Semites, a language spoken by Jews, that the Hasmonean ‘occupation’ ‘liberated’ from exploitation by the Hellenistic cities and made them part of a successful nation, culture and religion…” – was open to receive and accept religious and cultural influences, seems very convincing, and might hold some truth (see also Cohen 1999:118). Still, as we have seen above, such a representation is too schematic at best. We should not differentiate between the Hellenistic cities, supposedly

The fact that the pattern in Judea is so different from other sites discussed here,13 is proof that the Jewish and Gentile rural settlements went through a different process during the second half of the second century BCE. This supports the suggestion that the factor which determined the fate of the rural settlements (like the urban sites), was their religious, ethnic and political affinity. That is, Jewish sites were not affected at the time, while the Gentile settlements suffered during the Hasmonean expansion. We can only attribute the systematic destruction of the non-Jewish settlements to the Hasmoneans.

We should stress that it cannot be argued that rural sites have, inevitably, a short life-span, and a crisis can, therefore, be “identified” at practically every point in time. First of all, we should note that this claim runs contrary to the claim that there was continuity in the rural sector at the time. Moreover, the western Samaria farmsteads prove that small sites can exist for lengthy periods of time, and rural sites are not necessarily shortlived. The fact that those farmsteads, most of which existed continuously for more than 500 years, were destroyed in the second half of the second century BCE, only stresses the severe damage inflicted on the rural sector during the Hasmonean expansion. Finally, we have already shown that sites in Judea continued to exist at the time, hence disproving such a hypothesis. 14

Another structure in the near vicinity was destroyed in the first half of the second century BCE, probably a result of the battles that accompanied the revolt. 12 We are grateful to Emanuel Eisenberg for supplying us with the information, and to him and Alon de Groot for discussing these finds with us. 13 Not all the sites in Judea existed continuously, and it is likely that in the future more sites which differ from this pattern will be discovered. Still, it is not likely that the overall pattern will change significantly. 11

255

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

damaged by the Hasmonean, and the rural sector, allegedly accepting them willingly. It is clear that at least part of the rural sector was damaged during the Hasmonean conquest.

century BCE changed the settlement map of the region, and this trauma can be seen in the excavations of many sites, urban and rural alike. From this perspective this was a formative period in shaping the settlement in the region.

It should also be noted that even the non-Jewish cities maintained long-lasting and strong Semitic traditions, especially in the south, in Idumaea, and at times the cities maintained an independent cultural identity more than other sectors, due to their strength and long-lasting traditions (Kasher 1990:14-26). It seems that the dichotomy suggested by Rappaport and Ronen between Hellenized towns and the Semitic countryside needs reevaluation. In Idumaea, for example, only six km as the crow flies east of Maresha, at a site called Za‘akuka, and which was probably part of the city’s hinterland, a burial tomb from the Maresha type was discovered (Kloner, Regev and Rappaport 1992). We must note that we are not addressing only a general typological similarity to caves in Maresha, but the inscriptions in both sites are in Greek. The distance of the site from Maresha overturns a suggestion that the buried were an urban family from Maresha. This type of burial was new in the region during the Early Hellenistic period, and had its roots in Alexandria or Phoenicia. Not only is this cave similar to the Maresha caves, but the graffiti, the finds and the onomasticon of names found, all indicate Hellenized population, and all this is the countryside, and at some distance from the major city.

We must also note that only continued research, with more stress on the rural sector, will allow us to examine how widespread the damage was and to evaluate the relationship between the Hasmoneans and the non-Hellenized local population. In the meantime all we can say is that there was damage.

5.2 A Brief History of the Site from the First Settlement to the Creation of the British Park Kh. er-Rasm is a small site, located on a salient hill (some 330 m above sea level), south of Tel ‘Azekah, and west of the Elah Valley. The evidence for the earliest human activity at Kh. er-Rasm is dated to the Chalcolithic period, but the meager remains from this period are insufficient for learning anything about the nature of the human occupation at the time. The site was then abandoned for several millennia, until the resumption of human activity toward the end of the Iron Age I. The evidence from this period, as well as that of the early Iron Age II, is very limited, and is composed of a few sherds and one wall at best (W.21; this occupation corresponds with phase 0 in the main building).

Along the same lines, we should also point toward a paradoxical argument of those who believe that the rural sector in Idumaea was not destroyed due to the similarity of lifestyle between the inhabitants of the latter and the Jews in Judea. Ironically, the proof for this assumed “similarity” is based on the miqweh (ritual bath) and the marriage contract found at Maresha (e.g., Ronen 2003:127), as well as the burial practice discussed above. While there can be no doubt that the above does show some “connections” between the populations, we must stress that the finds were unearthed in the central city – in Maresha, and not in the rural sector. Ronen also agrees that Maresha was destroyed by the Hasmoneans, despite the ritual bath and other elements of ritual behavior (i.e., the marriage contract) found there. It is clear, therefore, that while the relations between Jews and Idumaeans deserve further discussion (in both urban and rural settings), there is nothing in the archaeological record to support the claims that there was more similarity between rural “Idumaeans” and Jews, in contrast to the situation in urban “Idumaea”.

The number of sherds dated to the later part of the Iron Age II is larger. This, and especially the architectural features of the main building (which was originally built as a square and symmetrical building using only nari stones) and its (dated) architectural parallels, led us to conclude that it is most likely that the building was built during this period, probably in the eighth century BCE. The original plan of the building was very impressive (phase 1 and probably 1a in the main building). It was a symmetrical building (some 30 x 30 m), with only one entrance (in the eastern part of the southern wing), through what appears like a small gateway. Moreover, a wall, built just outside the gateway, served as an access ramp from the east and west, but at the same time prevented direct access from the south. All these features attest to the importance and even the relative grandeur of the main building (Fig. 4.11). On the basis of an analysis of similar structures, we assume that Kh. er-Rasm served, at this time, as a center of an estate. The data, however, is too partial to allow us to be certain of this interpretation, let alone to enable further analysis.

It is quite clear that the archaeological evidence supports the view that the relations between Judea and its neighbors were hostile, and that the Hasmonean expansion was accompanied with destruction,15 not only in the urban sector, but also in the rural one – to the best of our knowledge the Hasmonean revolt and the subsequent processes led to an almost complete collapse of the Gentile rural settlement in many parts of the Land of Israel. The Hasmonean conquests of the second half of the second century and early first

The site was used during the Persian period, though it is not clear if the building itself was occupied or not, and whether it was used continuously or was resettled during the late Persian period. Idumaea. The issues raised by Schwartz (1991) are, therefore, worthy of a separate discussion, and the archaeological evidence cannot negate such a possibility. Still, we must stress that even if Schwartz is correct, the lack of hostility did not, in practice, have much impact on the Hasmonean policy.

This could have resulted from political and practical considerations, and not necessarily from hostility between the Jews and the population of 15

256

Avraham Faust and Adi Erlich

The structure was probably rebuilt during the Early Hellenistic period, either during the second half of the fourth century BCE or during the first decades of the third century BCE (phases 2a and 2b in the main building).16 At this time, the structure was re-established – its rooms were cleaned, and walls and roof rebuilt. This was accompanied by several changes to the main building, the most obvious of which included the use of a large amount of kirton blocks in the construction of the superstructure of many walls, and at times even in the construction of new kirton walls. It is likely that a complex of caves below the building was hewn at this time, supplying building stones for the rebuilding of the structure, and providing spaces that could have been used for working and storage. It is likely that the caves were used, partially at least, for the production of olive oil.

After this phase, the site was abandoned permanently, and was used merely as part of the agricultural area of nearby settlements, and perhaps as a resting place. During the late medieval and modern periods, that nearby settlement was Ajjur. Following the abandonment of Ajjur by its inhabitants during Israel’s War of Independence in 1948, the use of the area gradually changed; it was not used for agriculture anymore, and was subsequently turned into a military firing zone. From the 1950’s, trees were planted in the nearby region by the JNF, and in the early 1990’s an area of some 40,000 dunams, which encompasses the site, was turned into a large park (the JNF’s British Park). We hope that the excavations will turn Kh. er-Rasm into an important feature within the park, where visitors will be able to learn about its past.

The structure of the Early Hellenistic period was formidable and symmetrical, and we concluded that it is likely that it housed an Idumaean village strongman, like those described in the Zenon papyri.

Concluding this summary, we would like to refer to the usual treatment of small sites in the countryside in the archaeology of the Land of Israel. Admittedly, unlike the past “tel archaeology”, rural sites receive some attention today, mainly through the surveys. But treatment of small sites is rudimentary, and the sites are usually simply analyzed by size and possible demographics. We will allude to this briefly below (Chapter 5.3), and here we would only like to claim that such treatment does not really bring the “people without history” to the fore. It is simply another way to avoid dealing seriously with the countryside. Rural settlements vary greatly in form and function, and only when they are dealt with seriously can the data be genuinely illuminating. In the present monograph we have seen that one countryside site changed its function through time, and we hope that this study will exemplify the importance of paying appropriate attention to such sites.

At some point during the Hellenistic period, most probably during the second century BCE, the cave system collapsed, and the rooms above it ceased to function. From this time onward (phase 2c), the building ceased to function as before (Fig. 1.18). Its symmetrical features were lost, and some of the rooms were not used anymore; even some of the rooms with standing walls ceased to function as such – the room near the gateway, for example, was used as part of the courtyard, with tabun and kirton installation, but with very little else (this stands in contrast to the finds in most rooms). On the basis of the many finds that are dated to this phase and that were unearthed below the wall fall and the collapse of the second century BCE (below), we reconstructed the building during this phase as an inn, which served the travelers along the main roads that met at the valley just below Kh. er-Rasm.

5.3 Endnote: The Importance of the Excavations at Khirbet er-Rasm Most excavations are conducted in large and central sites, and hardly any attention is paid to rural sites. The latter are usually excavated, whenever necessary, in salvage excavations, but those are usually published only partially, if at all – some sites are published in short articles, while some only in brief notes, and still others are not published at all. Only rarely do the publications exceed the scope of articles, and this is usually done in salvage excavations of large and central sites. The lack of importance attributed by scholars to this type of excavations can be seen not only by the partial nature of publication, but also by the absence of any detailed discussion of the results of salvage excavations in general (see Faust and Safrai 2005; forthcoming).

During the late second century, probably as result of the impending Hasmonean campaign of 112/111 BCE, the inhabitants of the site left it and sealed its rooms (though leaving much of the rooms’ content), as they probably fled to nearby Tel ‘Azekah (phase 3 in the main building). The site was then destroyed, and its roof and walls collapsed (the historical importance of this event, and the significance of the excavations at Kh. er-Rasm in this respect, were discussed in some details, in Chapter 5.1). A brief period of reoccupation took place during the Early Roman period, probably in the first century CE (though we should note that it is possible that the entire phase should be dated to the early second century BCE). During this period (phase 4 in the main building) a small part of the main building was reused, and it is likely that this is when the pyramid-like stone structure in Area C was built. It is also likely that a simple hiding complex that was unearthed should be dated to this phase.

The Kh. er-Rasm excavations differ, as we aimed at excavating precisely such an “unimportant” rural site. Though we had some surprises, regarding both the date of the main phase of occupation and the nature of the site during its last phase, we believe that the results of the excavations can be used to show the importance of excavating small sites. Many issues, including historical ones (e.g., the Hasmonean policy toward the Gentile population), were addressed in the present report in a more

It is also possible that it was rebuilt at a slightly earlier period, and that this is the nature of the Persian period activity. 16

257

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

detailed manner than in the past (see more below), and only future excavations of more small sites and their publications will allow us to advance the research into those issues.

We should also note that the detailed information available from Kh. er-Rasm allowed us to refer, even if relatively briefly, to processes of site formation, including the processes that accompanied the abandonment, destruction and collapse of the site, as well as to other methodological issues which rarely receive any attention in archaeological reports of this period. The detailed information available also allowed us to study the function of the site, mainly during its last phase of occupation in the late second century BCE.

We should also note, in this regard, that the results of the excavations at Kh. er-Rasm indicate that the common interpretation of small sites, identified in surveys simply as farmsteads, is far too simplistic. Small sites are usually treated as though similar to each other, with a similar density coefficient, etc. The excavations at Kh. er-Rasm, however, indicate that not only is the countryside composed of many different types of “small sites”, but even the function of the same site changes through time. While not saying that survey results should be disregarded, we believe the results of the excavations call for more caution in the analysis of small sites, and stress the need for more excavations before the findings in surveys can be properly studied.

We believe that while there will no doubt be debates over some of our conclusions, we have shown the importance of excavating and publishing small sites, and hope the present monograph will advance the study of various aspects of the Hellenistic period, and to a lesser degree, of other periods as well.

It seems to us that the present report is the “first” final report of its kind of a site from the Hellenistic period. This is the only site dated to this period whose publication is relatively complete, and which includes detailed reports of all the finds (e.g., 14C, flints, shells, petrography, woods, botanical finds, animal bones, etc.), hence allowing future discussion of various aspects of the period’s social, ethnic, economic and environmental aspects during the site’s various phases. While our discussion suffered from the lack of comparative data (and many of the contributors referred to it, e.g., BarYosef, Chapter 3.8, and Boaretto, Chapter 3.9), and as a consequence some conclusions were preliminary, we hope that our report will give future studies some comparative data that will allow them to stand on firmer ground.

One of our conclusions is that excavating small sites has several advantages. Not only are those sites underrepresented in the archaeological discussion, and hence the contribution of each such site to archaeology is greater (as paradoxical as it may sound) than excavating “one more” large site, but due to their smaller size it is possible for the researcher to encompass all the data, and to try to integrate it in a manner that is not possible when excavating larger sites. Given the current state of knowledge about the various types of sites, we believe that excavating small sites has much more potential to contribute new data (especially when comparing costs) and to advance archaeology than the excavation of large urban sites.

258

Abbreviation

s

AASOR

Annual of the American Schools of Oriental Research AJA American Journal of Archaeology BAR Biblical Archeaology Review BASOR Bulletin of the American Schools of Oriental Research BCH Suppl. Bulletin de correspondance hellénique, supplément ESI Excavations and Surveys in Israel IAA Reports Israel Antiquities Authority Reports

IEJ INJ JBL JRA Suppl. NEAEHL PEQ ZDPV

259

Israel Exploration Journal Israel Numismatic Journal Journal of Biblical Literature Journal of Roman Archaeology Supplementary Series The New Encyclopedia of Archaeological Excavations in the Holy Land (E. Stern ed.). Palestine Exploration Quarterly Zeitschrift Deutschen Palästina Vereins

Bibliography

‘Ad, U., 2000, Landscape and Pattern in the lowland of Samaria in the Byzantine Period, Unpublished M.A. thesis, The Hebrew University of Jerusalem (Hebrew). Adan-Bayewitz, D., 1988–1989, “Kefar Hananya – 1987”, ESI 7–8, p. 108. Adan-Bayewitz, D., 1995, “A Lamp Mould from Sepphoris and the Location of Workshops for Lamps and Common Pottery Manufacture in Northern Palestine”, The Roman and Byzantine Near East, JRA Suppl. 14, Ann Arbor, pp. 177–182. Aharoni, Y., 1975, Investigation at Lachish V, The Sanctuary and the Residency, Tel Aviv. Aharoni, Y., 1979, The Land of the Bible: Historical Geography, Philadelphia. Aharoni, Y., Avi-Yonah, M., Rainey, A.F., and Safrai, Z., 2002, The Carta Bible Atlas (4th edition), Jerusalem. Ambar-Armon, E., 2007, Oil Lamps in the Land of Israel during the Hellenistic Period in Light of the Finds from the Maresha Excavations: Conservatism and Tradition alongside Creativity and Innovation, unpublished Ph.D. dissertation, Bar-Ilan University (Hebrew). Amiran, R., 1956, “The Millstones and the Potter’s Wheel”, Eretz-Israel 8, pp. 46–49 (Hebrew). Amit, D., 1993, “Discoveries in the Study of the Bar Kochba Revolt”, in Z.H. Erlich and Y. Eshel (eds.), Judea and Samaria Research Studies Proceedings of the 2nd Annual Meeting – 1992, Kedumim-Ariel, pp. 215–227 (Hebrew). Amit, D., Seligman, J., and Zilberbod, I., 2000, “Stone Vessel Workshop of the Second Temple Period East of Jerusalem”, in Y. Hirschfeld (ed.), Ancient Jerusalem Revealed, Jerusalem, pp. 353–358. Amit, D., Seligman, J., and Zilberbod, I., 2001, “A Quarry and Workshop for Production of Stone Vessels on the Eastern Slopes of Mount Scopus”, Qadmoniot 122, pp. 102–110 (Hebrew). Amit, D., Torgu, H., and Gendelman, P., 2008, “Horvat Burnat, a Jewish Village in the Lod Shephelah during the Hellenistic and Roman Periods”, Qadmoniot 136, pp. 96-107 (Hebrew). Amit, D., and Zilberbod, I., 1998, “Mazor – 1994–1995”, ESI 18, pp. 61–63 (Hebrew). Anderson-Stojanović, V.R., 1987, “The Chronology and Function of Ceramic Unguentaria”, AJA 91, pp. 105–

122. Arnon, Y., 2003, Development and Continuity in the Early Islamic Pottery Types from the 7th Century to the 12th century CE: The Caesarea Data as a Study Case, unpublished Ph.D. dissertation, Haifa University, Haifa. Avi Yonah, M., and Kloner, A., 1993, “Maresha”, in E. Stern (ed.), NEAEHL 3, Jerusalem, pp. 948–957. Aviam, M., 1999, “Yodfat - Uncovering a Jewish City in the Galilee from the Second Temple Period and the Time of the Great Revolt”, Qadmoniot 118, pp. 92–101 (Hebrew). Aviam, M., 2004, “Hellenistic Fortification in the ‘Hinterland’ of ‘Akko-Ptolemais”, Jews, Pagans and Christians in the Galilee, Rochester. Aviam, M., and Amitai, A., 2002, “Excavations at e Shuhara”, in Z. Gal (ed.), Eretz Tzafon, Jerusalem, pp. 119–133 (Hebrew with English Summary). Avigad, N., 1954, Ancient Monuments in the Kidron Valley, Jerusalem (Hebrew). Avissar, M., 1996a, “The Hellenistic and Early Roman Pottery”, in A. Ben-Tor, M. Avissar and Y. Portugali (eds.), Yoqne‛am I: The Later Periods (Qedem Reports 3), Jerusalem, pp. 48–59. Avissar, M., 1996b, “The Oil Lamps”, in A. Ben-Tor, M. Avissar and Y. Portugali (eds.), Yoqne‛am I: The Later Periods (Qedem Reports 3), Jerusalem, pp. 188-197. Avissar, M., and Shabo, E., 1998, “Qula”, ESI 20, pp. 51*–53*. Avissar, M., and Stern, E.J., 2005, Pottery of the Crusader, Ayyubid, and Mamluk Periods in Israel (IAA Reports 26), Jerusalem. Avitsur, S., 1975, Daily life in Eretz Israel in the XIX Century, Jerusalem (Hebrew). Avner-Levy, R., and Torge, H., 1999, “Rosh Ha-‘Ayin”, ESI 19, pp. 40*, 58–59. Bahat, D., 1971, “Area K”, in M. Dothan (ed.), Ashdod II-III, Atiqot 9–11, pp. 168–180. Bar, O., Zilberman, E., Gvirtzman, Z., and Feinstein, S., 2006, Reconstruction of the Uplift Stages of the Mountain Ridge in Central Israel, Geological Survey of Israel, Report GSI/28/2006, Jerusalem (Hebrew). Barag, D., 1992-1993, “New Evidence on the Foriegn Policy of Hyrcanus I”, INJ 12, pp. 1–12. Barag, D, 2002, “New Developments in the Research of the

260

Avraham Faust and Adi Erlich

Tombs of the Sons of Hezir and Zecharias”, Qadmoniot 123, pp. 38–47 (Hebrew). Barber, E.J.W., 1991, Prehistoric Textiles: The Development of Cloth in the Neolithic and Bronze Ages with Special Reference to the Aegean. Princeton. Barkay, R., 1992–1993, “The Marisa Hoard of Seleucid Tetradrachms Minted in Ascalon”, INJ 12, pp. 21–33. Bar-Adon, P., 1980, The Cave of the Treasure, the Finds from the Caves in Nahal Mishmar, Jerusalem. Bar-Kochva, B., 2002, “The Conquest of Samaria by John Hyrcanus: The Pretext for the Siege, Jewish Settlement in the `Akraba District, and the Destruction of the City of Samaria”, Cathedra 106, pp. 7–34 (Hebrew). Bar-Nathan, R., 1981, “The Finds at Lower Herodium: Pottery and Stone Vessels of the Herodian Period (1st century B.C.–1st century A.D.)”, in E. Netzer (ed.), Greater Herodium, Qedem 13, Jerusalem, pp. 54–70. Bar-Nathan, R., 2002, Hasmonean and Herodian Palaces at Jericho (E. Nezer director), Final Reports of the 1973–1987 Excavations III: The Pottery, Jerusalem. Bar-Nathan, R., 2006, Masada VII. The Ygael Yadin Excavations 1963–1965, Final Reports. The Pottery of Masada, Jerusalem. Bar-Nathan, R., and Kamil-Gitler, R., 2002, Typology of the Hrodian 3 Pottery, in R., Bar-Nathan (ed.), Hasmonean and Herodian Palaces at Jericho (E. Nezer director), Final Reports of the 1973–1987 Excavations III: The Pottery, Jerusalem, pp. 145-192. Bar-Nathan, R., et al., in press, Bisan: Umayyad Theatre Pottery Workshop (IAA Reports), Jerusalem. Barron, J. B., 1923, Governmental of Palestine – Census of Palestine 1922, Jerusalem. Baruch, E., and Amar, Z., 2004, “The Latrine (Latrina) in the Land of Israel in the Roman-Byzantine Period”, Jerusalem and Eretz Israel 2, pp. 27–50 (Hebrew). Baruch, Y., 2006, “Buildings of the Persian Hellenistic and Early Roman Periods at Khirbat Kabar, in the Northern Hebron Hills”, Atiqot 52, pp. 49*–71*. Beck, H.C., 1928, Classification and Nomenclature of Beads and Pendants, Archaelogia 77, pp. 1–71. Beijerinck, W., 1947, Zadenatlas der Nederlandsche flora, ten behoeve van de botanie, palaeontologie, bodemcultuur en warenkennis. Wageningen. Beit-Arieh, I., 1992, “Horvat Radum”, Eretz Israel 23, pp. 106–112 (Hebrew). Beit-Arieh, I., 1993a, “‘Uza, Horvat”, in E. Stern (ed.), NEAEHL 4, Jerusalem, pp. 1495–1497. Beit-Arieh, I., 1993b, “‘Ira, Horvat”, in E. Stern (ed.), NEAEHL 2, Jerusalem, pp. 642-646. Beit-Arieh, I., 1999, Tel ‘Ira a Stronghold in the Biblical Negev, Jerusalem. Beit Arieh, I., 2007, Horvat ‘Uza and Horvat Radum: Two Fortresses in the Biblical Negev, Tel-Aviv. Beit-Arieh, I., and Cresson, B.C., 1991, “Horvat ‘Uza, a Fortified Outpost on the Eastern Negev Border”, Biblical Archaeologist 54, pp. 126–135. Ben-Arieh, Y., 1987, “The Villages in Sancak Gaza (including Jaffa and Ramla) in the Eighteen-Seventies”, Shalem 5, pp. 139-187 (Hebrew).

Ben-Dor, I., 1945, “Palestinian Alabaster Vases”, Quarterly of the Department of Antiquities of Palestine 11, pp. 93–112. Bennett, W.J., and Blakley, J.A., 1989, Tel el-Hesi. The Persian Period (Stratum V). Winona Lake. Ben-Shlomo, D., 2005, Pottery Production Centers in Iron Age Philistia: An Archaeological and Archaeometric Study, Unpublished Ph.D. dissertation, Hebrew University of Jerusalem, Jerusalem. Ben-Tor, A., Avissar, M., and Portugali, Y., 1996, Yoqne‛am I, The Late Periods, Qedem Reports 3, Jerusalem. Bentor, Y.K., 1945, Petrographic Investigations of the Upper Albian-Lower Cenomanian Near Jerusalem and Their Contribution to the Problem of Dolomitization and Quartzification, Unpublished. Ph.D. dissertation, Hebrew University of Jerusalem, Jerusalem. Berber, E.J.W., 1991, Prehistoric Textile, Princeton. Berlin, A.M., 1997a, “The Plain Wares”, in S.C. Herbert (ed.), Tel Anafa II, i. The Hellenistic and Roman Pottery (JRA 10), Ann Arbor, pp. 1–211. Berlin, A.M., 1997b, “Between Large Forces: Palestine in the Hellenistic Periods”, Biblical Archaeologists 60, pp. 2–51. Berlin, A.M., 1999, “The Archaeology of Ritual: The Sanctuary of Pan at Banias/Caesarea Philippi”, BASOR 315, pp. 27–45. Berlin, A.M., 2002, “Power and its Afterlife, Tombs in Hellenistic Palestine”, BAR 65/2, pp. 138–147. Bertsch, K., 1941, Früchte und samen. Stuttgart. Biger, G., and Liphschitz, N., 1999, “The Use of Wood in the Palestinian Arab House”, Cathedara 92, pp. 199–202 (Hebrew). Binford, L.R., 1981, “Behavioral Archaeology and the “Pompeii Premise”, Journal of Anthropological Research 37, pp. 195–208. Blakely, J., 1988, “Ceramics and Commerce Amphorae from Caesarea Maritima”, BASOR 271, pp. 31–50. Bliss, F.G., and Macalister, R.A.S., 1902, Excavations in Palestine during the Years 1898–1900, London. Bourdieu, P., 1979, The Kabyle House or the World Reversed, Algeria 1960, Cambridge, pp. 133–153. Breccia, E., 1912, Catalogue general des antiquites egyptiennes la necropolis di Sciatbi (Musee d’Alexandrie), Alexandrie. Bronk-Ramsey, C., 1995, “Radiocarbon Calibration and Analysis of Stratigraphy: The OxCal Program”, Radiocarbon 37, pp. 425–430. Bronk-Ramsey, C., 2001, “Development of the Radiocarbon Prograam OxCal”, Radiocarbon 43, pp. 355–363. Bruins, H.J., 1986, Desert Environment and Agriculture in the Central Negev and Kadesh-Barnea During Historical Times. Nijkerk. Brünnow, R.E., and Domaszewski, A., 2004 (after 1909), Die Provincia Arabia, Strassburg. Buchbinder, B., 1969, Geological Map of Hashephela Region, Israel. Sheets: Kefar Uriyya 14-13, Sha‛ar Hagay 15-13 and Beit Shemesh 14-12, Nes Harim 1512. Scale: 1:20,000 and Explanatory Notes, Jerusalem: Geological Survey of Israel.

261

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

Cahill, J.M., 1992, “Chalk Vessel Assemblages of the Persian/Hellenistic and Early Roman Periods”, in A. de Groot and D.T. Ariel (eds.), Excavations at the City of David 1978–1985 directed by Yigal Shiloh vol. III (Qedem 33), pp. 190–257. Cameron, C.M., 1990, “Pit Structure Abandonment in the Four Corners Region of the American Southwest: Late Basketmaker III and Pueblo I Periods”, Journal of Field Archaeology 17, pp. 27–37. Cameron, C.M., 1993, “Abandonment and Archaeological Interpretation”, in C.M. Cameron, and S.A. Tomka (eds.), Abandonment of Settlements and Regions: Ethnoarchaeological and Archaeological Perspectives, Cambridge, pp. 3–7. Cameron, C.M., and Tomka, S.A. (eds.), 1993, Abandonment of Settlements and Regions: Ethnoarchaeological and Archaeological Perspectives, Cambridge. Carr, K., 2000, “Women’s Work: Spinning and Weaving in the Greek Home”, Archeologie des textiles des origins au Ve siecle: Actes du colloque de Lattes, octobre 1999, pp. 163–166. Casey, J., 1986, Understanding Ancient Coins: An Introduction for Archaeologists and Historians, London. Casson, L., 1994 (1974), Travel in the Ancient World, Baltimore and London. Cassuto, D., 2004, The Social Context of Weaving in the Land of Israel: Investigating the Context of Iron Age II Loom Weights, Unpublished MA thesis, Bar-Ilan University, Ramat Gan. Ciolek-Torrello, R., 1989, “Households, Floor Assemblages and the ‘Pompeii Premise’ at Grasshopper Pueblo”, in S. MacEachern, D.J.W. Archer and R.D. Garvin (eds.), Households and Communities: Proceedings of the 21st Annual Chacmool Conference, Chacmool, pp. 201–208. Clamer, C., 1989, “Calcite-Alabaster Vessels”, in Z. Herzog, G. Rapp, and O. Negbi (eds.), Excavations at Tel Michal, Tel Aviv, pp. 345–349. Clark, A.N., 1993, The Penguin Dictionary of Geography, Harmondsworth. Cohen, R., 1987, “Excavations at Moa, 1981–1985”, Qadmoniot 77–78, pp. 26–31 (Hebrew). Cohen, R., 1993, “Negev”, in E. Stern (ed.), NEAEHL 3, Jerusalem, pp. 1119–1145. Cohen, R., 2000, “The Ancient Roads from Petra to Gaza in View of the New Discoveries”, in E. Orion, and Y. Eini (eds.), The Spice Roads, Sede Boker, pp. 28–77 (Hebrew). Cohen, S.J.D., 1999, The Beginning of Jewishness, Boundaries, Varieties, Uncertainties, Berkley, Los Angeles and London. Cohen-Weinberger, A., and Goren, Y., 2004, “LevantineEgyptian Interactions during the 12th to the 1fifth Dynasties based on the Petrography of the Canaanite Pottery from Tel el-Dab’a”, Ägypten und Levante 9, pp. 69–100. Conyers, L.B., and Goodman, D., 1997, GroundPenetrating Radar: An Introduction for Archaeologists, London. Coogan, M.D., 1984, “Numeira 1981”, BASOR 255, pp.

75–81. Crowfoot, J.W., Crowfoot, G.M., and Kenyon, K.M., 1957, Samaria-Sebaste III: The Objects from Samaria, London. Dadon, M., 1997, “Har Adar”, Atiqot 32, pp. 63–79 (Hebrew). Dagan, Y., 1982, “Artificial Caves in Judean Lowlands”, Nikrot Zurim 6, pp. 45–47 (Hebrew). Dagan, Y., 1992a, Archaeological Survey of Israel: Map of Lakhish (98), Jerusalem. Dagan, Y., 1992b, The Judean Shephelah in the Period of the Monarchy in Light of the Archaeological Excavations and Surveys, Unpublished MA thesis, Tel Aviv University, Tel Aviv (Hebrew). Dagan, Y., 2000, The Settlements in the Judean Shephela in the Second and First Millenia B.C., A Test Case of Settlement Processes in a Geographic Region, Unpublished Ph.D. dissertation, Tel Aviv University, Tel Aviv (Hebrew). Dagan, Y., 2006, Archaeological Survey of Israel: Map of Amazia (109), Jerusalem. Dagan, Y., in press, Archaeological Survey of Israel: Map of Beth-Shemesh, Jerusalem. Dahari, U., and ‘Ad, U., 2000, “Shoam bypass road”, ESI 20, pp. 56*–59*. Dan, J., 1990, “The effect of dust deposition on the soils of the land of Israel”, Quaternary International, 5: 107–113. Dan, J., Fine, P., and Lavee, H., 2007, The Soils of the Land of Israel. Eretz,, Tel Aviv. Dan, J., Yaalon, D.H., Koyumdjisky, H., and Raz, Z., 1972, “The Soil Association Map of Israel”, Israel Journal of Earth Sciences 21, pp. 29–49. Dan, J., Yaalon, D.H., Koyumdjisky, H., and Raz, Z., 1976, The Soils of Israel, Agricultural Research Organization, Bet-Dagan, Pamphlet 159. Dan, J., Yaalon, D.H., Koyumdjisky, H., and Raz, Z., 1977, The soils of Israel, Bulletin 168 of the Volcanic Institute of Agricultural Research. With 1:500,000 soil map of Israel, Bet Dagan (Hebrew). Danin, A., 2004, Distribution Atlas of Plants in the Flora Palaestina Area, Jerusalem. Dar, S., 1982, “Ancient agricultural farms near Wadi Beit‘Arif”, Nofim 16, pp. 47–60 (Hebrew). Dar, S., 1986, Landscape and Pattern, An Archaeological Survey of Samaria 800 B.C.E.–636 C.E., Oxford. David, N., and Kramer, C., 2001, Ethnoarchaeology in Action, Cambridge. Davidson, G.R., 1952, Corinth XII, The minor objects, New Jersey. De Vaux, R., 1953, “Les grottes de Murabba‛at et leurs documents: rapport preliminaire”, Revue Biblique 60, pp. 245–267. Deetz, J., 1967, Invitation to Archaeology, New York. Deonna, W., 1938, le mobilier délien, exploration archeologioue de Délos 18, Paris. Department of Surveys, 1985, Atlas of Israel, State of Israel Ministry of Construction, Tel-Aviv. Derfler, S., 1984, The Hellenistic Temple at Beersheva, Israel, Ph.D. dissertation, Michigan.

262

Avraham Faust and Adi Erlich

Dorsey, D.A., 1991, The Roads and Highways of Ancient Israel, Baltimore and London. Dothan, M., 1971, Ashdod II–III, The Second and Third Seasons of Excavations 1963, Atiqot 9–10. Dothan, M., and Ben-Shlomo, D., 2005, Ashdod VI. The Excavations of Areas H and K (1968–1969) (IAA Reports 24), Jerusalem. Dothan, M., and Freedman, D.N., 1967, Ashdod I: The First Season of Excavations 1962, Atiqot 7. Dothan, T., 1982, The Philistines and Their Material Culture, Jerusalem. Dunand, M., 1950, Fouilles de Byblos 1933–1938, Paris. Dunand, M., and Duru, R., 1962, Oumm el-Amed: une ville de l’époque hellénistique aux échelles de Tyr, Paris. Durand, X., 1997, Des grecs en Palestine au IIIe siècle avant Jésus-Christ, Le dossier syrien des archives de Zénon de Caunos (261–252), Paris. Eitam, D., 1980, The Production of Oil and Wine in Mt. Ephraim in the Iron Age, Unpublished M.A. thesis, Tel Aviv University, Tel Aviv (Hebrew). Elgavish, J., 1974, Archaeological Excavations at Shiqmona. Report No. 2: The Level of the Hellenistic Period Stratum H. Haifa (Hebrew). Elgavish, J., 1976,” Pottery from the Hellenistic Stratum at Shiqmona”, IEJ 26, pp. 65–76. Emberling, G., 1997, “Ethnicity in Complex Societies, Archaeological Perspectives”, Journal of Archaeological Research 5 (4), pp. 295–344. Enviroplan, 2004, Biosphere Reserve in the Judean Hills – Methodology Structure and Plan, Final Report, Tel Aviv (Hebrew). Erlich, A., 2002, “Terracotta Figurines of Hellenistic Maresha: Between Art and Craft”, Assaph, Studies in Art History, 7, pp. 1–16. Erlich, A., 2009, The Art of the Hellenistic Palestine (British Archaeological Reports International Series 2010), Oxford. Erlich, A., and Kloner, A., 2008, Maresha Excavations Final Report II, Hellenistic Terracotta Figurines from the 1989–1996 Seasons (IAA Reports 35), Jerusalem. Eshel, E., and Kloner, A., 1996, “An Aramaic Ostracon of an Edomite Marriage from Maresha, Dated 176 B.C.E.”, IEJ 46, pp. 1-22. Fantalkin, A., and Tal, O., 2004, “The Persian and Hellenistic Pottery of Level I”, in D. Ussishkin (ed.), The Renewed Archaeological Excavations at Lachish (1973–1994) IV, Tel Aviv, pp. 2174–2194. Faust, A., 1995, The Rural Settlement in the Land of Israel during the Period of the Monarchy, Unpublished M.A. thesis, Bar-Ilan University, Ramat-Gan (Hebrew). Faust, A., 2000, “Rural Community in Ancient Israel”, BASOR 317, pp. 17–39. Faust, A., 2003, “The Farmstead in the Highland of Iron Age II Israel”, in A.M. Maeir, S. Dar, and Z. Safrai (eds.), The Rural Landscape of Ancient Israel, (British Archaeological Reports International Series 1121), Oxford, pp. 91–104. Faust, A., 2005a, Israelite Society in the Period of the Monarchy: An Archaeological Perspective, Jerusalem:

Yad Ben Zvi (Hebrew). Faust, A. 2005b, “The Canaanite Village: Social Structure of Middle Bronze Age Rural Communities”, Levant 37, pp. 105–125. Faust, A., 2005c, “The Israelite Village: Cultural Conservatism and Technological Innovation”, Tel Aviv 32, pp. 204–219. Faust, A., 2006a, “Farmsteads in Western Samaria’s Foothills: A Reexamination”, in A.M Maeir and P. De Miroschedji (eds.), “I Will Speak the Riddles of Ancient Times” (Abiah chidot minei-kedem – Ps. 78:2b): Archaeological and Historical Studies in Honor of Amihai Mazar on the Occasion of his Sixtieth Birthday, Winona Lake, pp. 477–504. Faust, A., 2006b, Israel’s Ethnogenesis: Settlement, Interaction, Expansion and Resistance, London. Faust, A., 2008, “Settlement and Demography in Seventh Century Judah and the Extent and Intensity of Sennacherib’s Campaign”, PEQ 140, pp. 168–194. Faust, A., and Erlich, A., 2008, “Hasmonean Policy toward Jentile Population in Light of the Excavations at Khirbet er-Rasm and Additional Rural Sites”, Jerusalem and Eretz-Israel 6, pp. 5-32 (Hebrew). Faust, A., and Safrai, Z., 2005, “Salvage Excavations as a Source for Reconstructing Settlement History in Ancient Israel”, PEQ 137, pp. 139–158. Faust, A., and Safrai, Z., Forthcoming, The Settlement History of Ancient Israel: A Quantitative Analysis (Hebrew). Finkelstein, I., 1978, Rural Settlement in the Foothills and the Yarkon Basin in the Israelite-Hellenistic Periods, Unpublished M.A. thesis, Tel-Aviv University, Tel-Aviv (Hebrew). Finkelstein, I., 1981, “Israelite and Hellenistic farms in the Foothills and the Yarkon Basin”, Eretz-Israel 15, pp. 331–48 (Hebrew). Finkielsztejn, G., 1998, “More Evidence on John Hyracanus I’s Conquest: Lead Weights and Rhodian Amphora Stamps”, Bulletin of the Anglo-Israel Archaeological Society 16, pp. 33–63. Fischer, M., 1995, “Bacchides in Emmaus”, in A. Kasher and A. Oppenheimer (eds.), Dor Le-Dor, From the End of Biblical Times up to the Redaction of the Talmud, Studies in Honor of Joshua Efron, Jerusalem, pp. 87–91 (Hebrew). Fischer, M. and Tal, O., 1999, “The Hellenistic Period”, in I. Roll and O. Tal (eds.), Apollonia-Arsuf Final Report of the Excavations: Vol. I, The Persian and Hellenistic Periods (with appendices on the Chalcolithic and Iron Age II Remains), Jerusalem, pp. 223–261. Fischer, M., and Tal, O., 2007, “The Hellenistic and Roman Periods,” in I. Beit-Arieh (ed.), Hurvat `Uza and Hurvat Radum, Two Fortresses in the Biblical Negev (Tel Aviv University, Institute of Archaeology, Monograph Series 25), Tel Aviv, pp. 335–339. Fragman, O., Plitmann, U., Heller, D., and Shmida, A., 1999, Checklist and Ecological Data-Base of the Flora of Israel and its Surroundings, Jerusalem. Frankel, R., Ventura, R., 1998, “The Mispe Yamim

263

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

Bronzes”, BASOR 311, pp. 39–55. Frederick, C.D., and Krahtopoulou, A., 2000, “Deconstructing agricultural terraces”, in P. Halstead and C. Frederick (eds.), Landscape and Land Use in Postglacial Greece, Sheffield, pp. 79–94. Frenkel, R., 1997, “The Sanctuary from the Persian Period at Mount Mizpe Yamim”, Qadmoniot 113, pp. 46–53 (Hebrew). Funk, R.B., 1993, “Beth Zur”, in E. Stern (ed.), NEAEHL, 1, Jerusalem: pp. 259-261. Gaffney, C., and Gater J., 2003, Revealing the Buried Past: Geophysics for Archaeologists, Stroud, Gloucestershire. Gal, Z., 1991., “A Stone-Vessel Manufacturing Site in the Lower Galilee”, ‘Atiqot XX, pp. 25*–26*. (Hebrew). Galilli, E., Sharvit, Y., and Bahat-Silberstien, N., 1996, “Yavne Yam Underwater Survey”, Hadashot Arkheologiyot 106, pp. 117–8 (Hebrew). Gawlikowski, M., 1970, Monuments funéraires de Palmyre, Warszawa. Geva, H., 2003, “Hellenistic Pottery from Areas W and X-2”, in H. Geva (ed.), Jewish Quarter Excavations in the Old City of Jerusalem conducted by Nahman Avigad, 1969–1982. Volume II: The Finds from Areas A, W and X-2. Final Report, Jerusalem, pp. 113-175. Geva, H., 2006a, “Stone Artifacts”, in H. Geva (ed.), Jewish Quarter Excavations in the Old City of Jerusalem Conducted by Nahman Avigad, 1969–1982 Vol. III, Jerusalem, pp. 218–38. Geva, H., 2006b, “A Proposal for Jerusalemite Stone Vessel Typology of the Second Temple Periods”, in E. Baruch, Z. Greenhut and A. Faust (eds.), New Studies on Jerusalem, Volume 11, Ramat-Gan, pp. 193–200 (Hebrew, English abstract pp. 41*–42*). Geva, H., and Rosenthal-Heginbottom, R., 2003, “Local Pottery from Area A”, in H.Geva (ed.), Jewish Quarter Excavations in the Old City of Jerusalem conducted by Nahman Avigad, 1969–1982. Volume II: The Finds from Areas A, W and X-2. Final Report, Jerusalem, pp. 176-191. Gibson, S., 1983, “The Stone Vessel Industry at Hizma”, IEJ 33, pp. 176–188. Gibson, S., 2001, “Agricultural Terraces and Settlement Expansion in the Highlands of Early Iron Age Palestine: Is there any Correlation Between the Two?” in A. Mazar (ed.), Studies in the Archaeology of the Iron Age in Israel and Jordan (Series Journal for the Study of the Old Testament, Supplement Series: 331), Sheffield, pp. 113–46. Gichon., M., 1993, “‛Eqed, Horvat”, in E. Stern (ed.), NEAEHL Vol. 2 Jerusalem, pp. 416–417. Gitin, S., 1990, Gezer III: A Ceramic Typology of the Late Iron II, Persian and Hellenistic Periods at Tel Gezer (Annual of the Nelson Glueck School of Biblical Archaeology 3), Jerusalem. Giveon, R., 1973, “Egyptian Objects in Bronze and Faience”, in Y. Aharoni (ed.), Beer Sheba I, Tel Aviv, pp. 54–55. Glass, J., Goren, Y., Bunimovitz, S., and Finkelstein, I., 1993, “Petrographic Analysis of Middle Bronze Age III,

Late Bronze Age and Iron Age I Pottery Assemblasges”, in I. Finkelstein (ed.), Shiloh, The Archaeology of a Biblical Site, Tel Aviv, pp. 271–286. Golan, A., 1991, Wartime Spatial Changes, Former Arab Territories Within the State of Israel, 1948–1950, BenGurion Research Center, Beer Sheva (Hebrew). Gonen, A., and Tahal Group, 2004, Open Spaces Survey, Subject maps Survey, National Planning and Building Council, Ministry of the Interior, Planning Authority, Jerusalem (Hebrew). Goodman, M., 1994, Mission and Conversion, Proselytizing in the Religious History of the Roman Empire, Oxford. Gophna, R., and Beit-Arieh, I., 1997, Map of Lod (Archaeological Survey of Israel), Jerusalem. Goren, Y., 1995, “Shrines and Ceramics in Chalcolithic Israel: The View through the Petrographic Microscope”, Archaeometry 37 (2), pp. 287–305. Goren, Y., 1996a, “Principles of the Petrographic Study of Archaeological Finds and the Application of this Method in the Archaeology of Israel”, Qadmoniot 112, pp. 107–114 (Hebrew). Goren, Y., 1996b, “The Southern Levant in the Early Bronze Age IV: The Petrographic Perspective”, BASOR 303, pp. 33–72. Goren, Y., Finkelstein, I., and Na’aman, N., 2004, Inscribed in Clay, Tel-Aviv. Goren, Y., Kamaiski, E., and Kletter, R., 1996, “The Technology and Provenience of the Figurines from the City of David: Petrographic Analysis”, in D.T. Ariel, and A. De Groot (eds.), City of David Excavations. Final Report Vol. IV (Qedem 35), Jerusalem, pp. 87–89. Grant, E. and Wright, E., 1939, Ain Shems Excavations Part V, Haverford. Grosman, D., 1994, Expansion and Desertion: The Arab Village and its Offshoots in Ottoman Palestine, Jerusalem. Guérin, V., 1869, Description géographique, historique et archéologique de la Palestine, Judée, II, Paris. Gutfeld, O., and Nenner-Soriano, R., 2006, “Metal Artifacts”, in H. Geva (ed.), Jewish Quarter Excavations in the Old City of Jerusalem, Conducted by Nahman Avigad, 1969–1982, Vol. III, The Finds from Area E, Final Report, Jerusalem, pp. 272-282. Guz-Zilberstein, B., 1995, “The Typology of the Hellenistic Coarse Ware and Selected Loci of the Hellenistic and Roman Periods”, in E. Stern (ed.), Excavations at Dor, Final Report Volume I B, Areas A and C: The Finds (Qedem Reports 2), Jerusalem, pp. 289–433. Hachlili, R., 2005, Jewish Funerary Customs, Practices and Rites in the Second Temple Period, Leiden and Boston. Hart, S., 1988, “Excavations at Ghareh, 1986: Preliminary Report”, Levant XX, pp. 89–99. Hartmann, M., 1883, “Die Ortschafatenliste des Liwa Jerusalem in dem turkeschen Staatskalander fur Syrien auf das Jahr 1288 des Flucht. 1871“, ZDPV 4, pp. 102–149. Hayes, J.W., 1972, Late Roman Pottery, London. Hayes, J.W., 1985, “Sigillate Orientali”. In Enciclopedia dell’Arte Antica, Classica e Orientale, Rome, pp. 1–48.

264

Avraham Faust and Adi Erlich

Hayes, J.W., 1991, Paphos III: The Hellenistic and Roman Pottery, Nicosia. Hegmon, M., Nelson, M.C., and Ruth, S.M., 1998, “Abandonment and Reorganization in the Mimbers Region of the American Southwest”, American Anthropologists 100, pp. 148–162. Herr, L.G., 1997, “The Iron Age Period: Emerging Nations”, Biblical Archaeologists 60/3, pp. 114–183. Herzog, Z., 1993, “Beersheba, Tel Beersheba”, in E. Stern (ed.), NEAEHL Vol. 1, Jerusalem, pp. 167–173. Herzog, Z., 1997, “The Arad Fortresses”, in R. Amiran, et al. (eds.), Arad, Tel-Aviv, pp. 113–292 (Hebrew). Hillier, B., and Hanson, J., 1984, The Social Logic of Space, Cambridge. Hirschfeld, Y., 1995. The Palestinian Dwelling in the Roman-Byzantine Period, Jerusalem. Hirschfeld, Y., 2000, Ramat Hanadiv Excavations, Final Report of the 1984–1998 Seasons, Jerusalem. Hoepfner, W., 1993, Zum Mausoleum von Belevi, Archäologischer Anzeiger 1993, pp. 111–123. Hoffner, H.A., 1966, „Symbols of Masculinity and Femininity: Their Use in Ancient Near Eastern Sympathetic Magic Rituals“, JBL 85, pp. 326–334. Hoglund, K.G., 1992, Achaemenid Imperial Administration in Syria-Palestine and the Mission of Ezra and Nehemiah, Atlanta. Horwitz, L.K., Tchernov, E., and Dar, S., 1990, “Subsistence and Environment of Mount Carmel in the Roman/ Byzantine Period to the Middle Ages as Evidenced by Animal Remains from the Site of Sumaqa”, IEJ 40, pp. 287–304. Houghton, A., and Lorber, C., 2002, Seleucid Coins, A Comprehensive Catalogue, Vols. I–II, New York. Houghton, A., and Spaer, A., 1998, Sylloge Nummorum Graecorum. Israel I. The Arnold Spaer Collection of Seleucid Coins, London. Howland, R.H., 1958, The Greek Lamps and their Survivals: The Athenian Agora: Results of Excavations Conducted by the American School of Classical Studies and Athens, Vol. IV, Princeton. Inomata, T., and Webb, R.W. (eds.), 2003, The Archaeology of Settlement Abandonment in Middle America, Salt Lake City. Issar, A., 1968, “Geology of the Central Coastal Plain of Israel”, Israel Journal of Earth Sciences 17, pp. 16–29. Johnson, B., and Stager, L., 1995, “Ashquelon: Wine Emporium of the Holy Land”, in S. Gitin (ed.), Recent Excavations in Israel, A View to the West: Reports on Kabri, Nami, Miqne-Ekron, Dor and Ashkelon, Dubuque, Iowa, pp. 95–109. Jones, S., 1997, The Archaeology of Ethnicity: Constructing Identities in the Past and Present, London. Kahane, P., 1953, “Pottery Types from the Jewish Ossuary Tombs around Jerusalem”, IEJ 3, pp. 48–54. Karali, L., 1999, Shells in Aegean Prehistory (British Archaeological Reports International Series 761), Oxford. Kasher, A., 1988, Jews, Idumeans and Ancient Arabs. Relations of the Jews in Eretz-Israel with the nations of

the Frontier and the Desert during the Hellenistic and Roman Era (332 BCE–70 CE), Tübingen. Kasher, A., 1990, Jews and the Hellenistic Cities in EretzIsrael. The Relations of the Jews in Eretz-Israel with the Hellenistic Cities during the Second Temple Period (332 BCE–70 CE). Tübingen. Kasher, A., 2002, “On the Character and Origin of the Herodian Dynasty”, Cathedra 103:165-184 (Hebrew). Keay, S.J., 1984, Late Roman Amphorae in the Western Mediterranean, Oxford. Kent, S. (ed.), 1990, Domestic Architecture and the Use of Space, Cambridge. Kerkhof, V.I., 1969, “Catalogue of the Shechem Collection in the Rijkmuseum Van Oudheden in Leiden”, in A. Klasens (ed.), Oudheid Kundige Mede Delingen, Leiden, pp. 28–109. Killebrew, A., 1996, „Pottery Kilns from Deir el-Balah and Miqne-Ekron“, in J.D. Seger (ed.), Retrieving the Past (Essays on Archaeological Research and Methodology in Honor of Gus W. Van Beek), Mississippi, pp. 135–161. Killebrew, A., 1998a, “Ceramic Typology and Technology of Late Bronze II and Iron I”, in S. Gitin, A. Mazar, and E. Stern (eds.), Mediterranean Peoples in Transition, Thirteenth to Early Tenth Centuries BCE, Jerusalem, pp. 379–405. Killebrew, A.E., 1998b, Ceramic Craft and Technology during the Late Bronze and Early Iron Ages: The Relationship Between Pottery Technology, Style, and Cultural Diversity, Unpublished Ph.D. dissertation, Hebrew University of Jerusalem. Kingsley, S.A., and Raveh, K., 1996, The Ancient Harbour and Anchorage at Dor, Israel: Results of the Underwater Surveys 1976-1991 (British Archaeological Reports International Series 626), Oxford. Kislev, M.E., Melamed, Y., Simchoni, O., and Marmorstein, M., 1997, “Computerized Key of Grass Grains of the Mediterranean Basin”, Lagascalia 191–2, pp. 289–294. Kislev, M.E., Melamed, Y., Simchoni, O., and Marmorstein, M., 1999, “Computerized Keys for Archaeological Grains: First Steps”, in S. Pike and S. Gitin (eds.), The Practical Impact of Science on Near Eastern and Aegean Archaeology, London, pp. 29–31. Kislev, M.E., Simchoni, O., Melamed, Y., and Marmorstein, M., 1995, “Computerized Key for Grass Grains of Israel and Its Adjacent Regions”, in H. Kroll and R. Pasternak (eds.), Res archaeobotanicae, Kiel, pp. 69–79. Kleberg, T., 1957, Hotels, restaurants et cabarets des l’antiquite romaine: Etudes historiques et philologiques, Uppsala. Klein, R.G., and Cruz-Uribe, K., 1984, The Analysis of Animal Bones from Archaeological Sites, Chicago. Kloner, A., 1991, “Maresha”, Qadmoniot 95–96, pp. 70–85 (Hebrew). Kloner, A., 1993, “Mareshah (Marisa)”, NEAEHL Vol. 3, Jerusalem, pp. 951–957. Kloner, A., 1994, “Unique Hellenistic Juglet from Maresha (Marissa/Tel Sandahannah)”, in S. Drougou (ed.), Third Scholarly Conference on Hellenistic Pottery. Thessalonika. September 191, Athens, pp. 269–271.

265

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

Kloner, A., 1996, Maresha: A Guide, Jerusalem (Hebrew). Kloner, A., 2001, “The Economy of Hellenistic Maresha”, in Z.H. Archibald et al. (eds.), Hellenistic Economies, London and New York, pp. 103–131. Kloner, A., 2001–2002, “Water Cisterns in Idumea, Judaea and Nabatea in the Hellenistic and the Early Roman Period”, Aram 13–14, pp. 461–485. Kloner, A., 2003, Maresha Excavations Final Report 1, Subterranean Complexes 21, 44, 70 (IAA Reports 17), Jerusalem. Kloner, A., and Arbel, Y., 1998, “Maresha – Area 61 (Subterranean Complex)”, ESI 17, pp. 157–162. Kloner, A., Erlich, A., Vitto, F., and Shmuel, D., 1998, “Maresha – Area 61”, ESI 17, pp. 163–165. Kloner, A., Erlich, A., and Whetstone, S., 2000, “Maresha, the Lower City”, ESI 20, pp. 118*–119*. Kloner, A., and Eshel, E., 1999, “A Seventh-Century BCE List of Names from Maresha”, Eretz Israel 26, pp. 147–150 (Hebrew). Kloner, A., Finkielsztejn, G., and Arbel, Y., 1998, “Maresha – Area 100”, ESI 17, pp. 154–157. Kloner, A., and Hess, O., 1985, „A Columbarium in Complex 21 at Maresha“, `Atiqot (ES), 17, pp. 122–133. Kloner, A., Regev, D., and Rappaport, U., 1992, “A Hellenistic Burial Cave in the Judean Shephelah”, ‘Atiqot XXI, pp. 27*–50* (Hebrew, English abstract pp. 175–176). Kloner, A., and Sagiv, N., 1993, “The Olive Presses of Hellenistic Maresha, Israel”, in M.C. Amouretti and J.P. Brun (eds.), Oil and Wine Production in the Mediterranean Area (BCH Suppl. 26), Athens-Paris, pp. 119–136. Kloner, A., and Sagiv, N., 2003, “Subterranena Complexes 44 and 45”, in A. Kloner (ed.), Maresha Excavations Final Report 1, Subterranean Complexes 21, 44, 70 (IAA Reports 17), Jerusalem, pp. 51-72. Kloner, A., and Tepper, Y., 1987, The Hiding Complexes in the Judean Shephelah, Tel Aviv (Hebrew). Kloner, A., and Zissu, B., 2003, “Hiding Complexes in Judaea: An Archaeological and Geographical Update on the Area of the Bar Kochba Revolt”, in P. Schäfer (ed.), The Bar Kochba War Reconsidered, Tübingen. Kloner, A., and Zissu, B., 2007, The Necropolis of Jerusalem in the Second Temple Period, Leuven. Klute, A (ed.), 1986, Methods of Soil Analysis (2nd ed), monograph No. 9, Part 1. Madison, WI: American Society of Agronomy. Kochavi, M., 1972, “The Land of Judah”, in M. Kochavi (ed.), Judaea, Samaria and the Golan: Archaeological Survey 1967-1968, Jerusalem, pp. 19-89 (Hebrew). Kochavi, M., and Beit-Arieh, I., 1994, Map of Rosh Ha‘Ayin (Archaeological Survey of Israel), Jerusalem. Kochman, M., 1981, The Status and Extent of the Yehud Pahwe in the Persian Period, Unpublished Ph.D. dissertation, The Hebrew University of Jerusalem, Jerusalem (Hebrew). Kokkinos, N., 1998, The Herodian Dynasty, Origins, Role in Society and Eclipse, Sheffield. Kramer, C., 1979, “An Archaeological View of

Contemporary Kurdish Village: Domestic Architecture, Household Size, and Wealth”, in C. Kramer (ed.), Ethnoarchaeology: Implications of Ethnography for Archaeology, New York, pp. 139–163. Kramer, C., 1982, Village Ethnoarchaeology: Rural Iran in Archaeological Perspective, New York. Krause, H., 2005, The Distribution of Bronze Cosmetic Utensils in Masada, Unpublished M.A. thesis, Hebrew University of Jerusalem, Jerusalem (Hebrew). Kutiel, P., 2000, “Annuals’ Distribution and Properties in a Judean Mountains Terrace Site”, in Y. Eshel (ed.), Judea and Samaria Research Studies, 9, Ariel, pp. 37–381 (Hebrew with English abstract). Lapp, N.L., 1981, The Third Campaign at Tel el-Fûl: The Excavations of 1964 (AASOR 45, 1978), Cambridge, Mass. Lapp, P.W., 1961, Palestinian Ceramic Chronology 200 BCE–70 AD, New Haven. Lapp, P.W., and Lapp, N., 1968, “Iron II-Hellenistic Pottery Groups”, in O. Sellers, R.W. Funk, J.L. McKenzie, P. Lapp, and N. Lapp, The 1957 Excavation at Beth-Zur (AASOR 38), pp. 54–79. Levine, T., 1999, Pottery Assemblages of the Third and Second Centuries BCE from Upper Area 61 at Maresha. Unpublished M.A. thesis, Bar-Ilan University, RamatGan (Hebrew). Levine, T., 2003, “Chapter 6: Pottery and Small Finds from Subterranean Complexes 21 and 70, Chapter 7: Pottery and Small Finds from Subterranean Complex 44”, in A. Kloner (ed.), Maresha Excavations Final Report I: Subterranean Complexes 22, 44, 70 (IAA Reports 17), Jerusalem, pp. 73–136. Levy, J., 1998, The Bow and the Blanket: Religion, Identity and Resistance in Raramuri Material Culture, Journal of Anthropological Research 54, pp. 299–324. London, G., 1989, 1989, “A Comparison of Two Life Styles of the Late Second Millennium BC”, BASOR 273, pp. 37-55. Lyman, R.L., 1994, Vertebrate Taphonomy. Cambridge. Macalister, R.A.S., 1912, The Excavations of Gezer 1902– 1905 and 1907–1909 vol. 3, London. Magen, Y., 1988, The Stone Vessel Industry in Jerusalem in the Second Temple Period, Jerusalem. (Hebrew). Magen, Y., 1993, “Mount Gerizim and Samaritans”, in F. Manns, and E. Alliata (eds.), Early Christianity in Context: Monuments and Documents, Jerusalem, pp. 91–147. Magen, Y., 2000, “Jerusalem as a Center of the Stone Vessel Industry during the Second Temple Period”, in H. Geva (ed.), Ancient Jerusalem Revealed, Jerusalem, pp. 244–256. Magen, Y., 2002, The Stone Vessel Industry in the Second Temple Period: Excavations at Hizma and the Jerusalem Temple Mount (Judea and Samaria Publications 1), Jerusalem. Magen, Y., 2004, “A Second Temple-Period Viticulture and Wine-Manufacturing Agricultural Settlement”, in Y. Magen, D.T. Ariel, G. Bijovsky, Y. Tzionit, and O. Sirkis (eds.), The Land of Benjamin (Judea and Samaria

266

Avraham Faust and Adi Erlich

Meshorer, Y., 2001, A Treasury of Jewish Coins from the Persian Period to Bar Kokhba, Jerusalem. Meteorological Notes, 1967, Climatological Standard Norms of Rainfall 1931–1960. The Israel Meteorological Service, Ser. A. No. 21, Bet Dagan. Mienis, H.K., 1992, “Molluscs”, in A. De Groot and D.T. Ariel (eds.), Excavations at the City of David 1978–1985, Vol. III (Qedem 33), Jerusalem, pp. 122–130. Mills, E., 1931, Census of Palestine 1931, Jerusalem. Munsell, 1998, Munsell Soil Color Charts, Revised Washable Edition, New Windsor. Murdock, G.P., 1937, “Comparative Data on the Division of Labor by Sex”, Social Forces 15, pp. 551–53. Murdock, G.P., and Provost, C., 1973, “Factors in the division of Labor by Sex: A Cross-cultural Analysis”, Ethnology 12, pp. 202–225. Negev, A., 1971, “The Nabatean Necropolis of Mampis (Kurnub)”, IEJ 21, pp. 110–129. Negev, A., 1986, Nabatean Archaeology Today, New York and London. Negev, A., 1993, “Kurnub”, in E. Stern (ed.), NEAEHL vol. 4, Jerusalem, pp. 882–893. Nelson, M.C., and Schachner, G., 2002, „Understanding Abandonment in the North American Southwest“, Journal of Archaeological Research 10, pp. 167–206. Netzer, E., 1992, “Domestic Architecture in the Iron Age”, in A. Kempinski and R. Reich (eds.), The Architecture of Ancient Israel from the Prehistoric to the Persian Period, Jerusalem, pp. 193–201. Oked, S., 2001, “‘Gaza jar’, A Chronicle, and Economic Overview”, in A. Sasson, Z. Safrai and N. Sagiv (eds.), Ashkelon, A City on the Seashore, Tel Aviv, pp. 227–250 (Hebrew). On, A., and Weksler-Bdolah, S., 2005, “Khirbet Um elUmdan – A Jewish Village with a Synagogue from the Second Temple Period at Modiin”, Qadmoniot 130, pp. 107–116 (Hebrew). Oppenheim, B.M., 1939, Tell Halaf, Paris. Oren, E.D., and Rappaport, U., 1984, “The Necropolis of Maresha-Beth Govrin”, IEJ 34, pp. 114–153. Orton, C., Tyres, P., and Vince, A., 1993, Pottery in Archaeology, Cambridge. Peshin, R., 2001, Altars and Stone Vessels from Maresha, Unpublished Ph.D. dissertation, Bar-Ilan University, Ramat-Gan (Hebrew). Peters, J.P., and Thiersch, H., 1905, Painted Tombs in the Necropolis of Marissa, London. Petrie, W.M.F., 1937a, Anthedon, London. Petrie, W.M.F., 1937b, Funeral Furniture of Egypt and Stone and Metal Vases, London. Porat, N., 1987, “Local Industry of Egyptian Pottery in Southern Palestine during the Early Bronze I Period”, Bulletin of the Egyptological Seminar 8, pp. 109–129. Praschniker, C., and Theuer, M., 1979, Das Mausoleum von Belevi, Forschugen in Ephesos VI, Wien. Price, M.J., 1991, The Coinage in the Name of Alexander the Great and Philip Arrhidaeus, BMC, Zurich and London.

Publications 3), Jerusalem, pp. 29–144. Magness, J., 1993, Jerusalem Ceramic Chronology, Sheffield. Magness, J., 1995, “Pottery from Area V/4 at Caesarea”, AASOR: 133–145. Magness, J., 2003, “ Late Roman and Byzantine Pottery”, in H. Geva (ed.), Jewish Quarter Excavations in the Old City of Jerusalem, vol. II, Jerusalem, pp. 423–432. Magness, J., 2005, “The Roman Legionary Pottery”, in B. Arubas and H. Goldfus (eds.), Excavations on the Site Jerusalem International Convention Center (Binyanei Ha’Huma): A Settlement of the Late First to Second Temple Period, the Tenth Legion kilnworks and a Byzantine Monastery Complex. The Pottery and other Finds (Journal of Roman Archaeology Supplementary Series 60), pp. 69–194. Magness, J., 2006, “ Late Roman and Byzantine Pottery”, in H. Geva (ed.), Jewish Quarter Excavations in the Old City of Jerusalem, Vol. III, Jerusalem, pp. 184–191. Magness-Gardiner, B., 1994, “Urban-Rural Relations in Bronze Age Syria: Evidence from Alalah Level VII Palace Archives”, in G.M. Schwartz and S.E. Falconer (eds.), Archaeological Views from the Countryside: Village Communities in Early Complex Societies, Washington, pp. 37–47. Majcherek, G., 1995, “Gazan Amphorae: Typology Reconsidered”, in H. Meyza and J. Mlynarczyk (eds.), Hellenistic and Roman Pottery in the Eastern Mediterranean: Advances in Scientific Studies, Acts of the Second Workshop at Nieborow, Warsaw, pp. 163-178. Malul, M., 1996, “David’s Curse of Joab (2 Sam. 3:29) and the Social Significance of mhzyq bplk”, Aula Orientalis 10, pp. 49–67. Mayerson, P., 1992, “The Gaza ‛Wine’ Jar (Gazition) and the ‛Lost’ Ashkelon Jar (Askalônion)”, IEJ 42, pp. 76–80. Mazar, A., “The Excavations at Khirbet Abu et-Twein and the System of Iron Age Fortresses in Judah”, Eretz Israel 15, pp. 229-249 (Hebrew]) Mazar, A., 1982, “Iron Age Fortresses in the Judaean Hills”, PEQ 114, pp. 87–109. Mazar, A., and Panitz-Cohen, N., 2001, Timnah (Tel Batash) II. The Finds from the First Millennium BCE, (Qedem 42), The Hebrew University of Jerusalem. Mazar, B., 1971, “The Excavations in the Old City of Jerusalem near the Temple Mount – Second Preliminary Report, 1969–1970”, Eretz-Israel 10, pp. 1–34 (Hebrew). Mazar, B., and Dunayevsky, I., 1967, “Ein Gedi: Fourth and Fifth Seasons of Excavations Preliminary Report”, IEJ 17, pp. 133–143. McGuire, R.H., 1982, “The Study of Ethnicity in Historical Archaeology”, Journal of Anthropological Archaeology 1, pp. 159–178. McKenzie, J., 1990, The Architecture of Petra, Oxford. McNicoll, A., Smith, R.H., and Hennessy, J.B., 1982, Pella in Jordan I, Canberra. Meshorer, Y., 1985, City-Coins of Eretz-Israel and the Decapolis in the Roman Period, The Israel Museum, Jerusalem.

267

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

Pritchard, J.B., 1964, Winery, Defenses and Sounding at Gibeon, Philadelphia. Pritcahrd, J.B., 1988, Sarepta IV, Beyrouth. Rachmani, L.I., 1964, “A Partial Survey in the Adulam Area”, Yediot Bahakirat Eretz-Israel Weatiqoteha 28, pp. 209–231 (Hebrew). Rahmani, L.Y., 1961, “Jewish Rock-cut Tombs in Jerusalem”, Atiqot 3, pp. 93–127. Rahmani, L.Y., 1967, “Jason’s Tomb”, IEJ 17, pp. 61–100. Rahmani, L.Y., 1970, “The Coins from Tirat Yehuda”, Atiqot (HS) 6, pp. 68–69 (Hebrew). Rahmani, L.Y., 1994, A Catalogue of Jewish Ossuaries in the Collections of the State of Israel, Jerusalem. Rainey, A.F., and Notley, R.S., 2006, The Sacred Bridge. Carta’s Atlas of the Biblical World. Jerusalem. Rapoaport, U., 1967, “The Hellenistic Cities and the Judaization of the Land of Israel under the Hasmonean Rule”, in S. Perlman and B. Shimron (eds.), Doron, Tel Aviv, pp. 219-230 (Hebrew). Rapoaport, U.,1984, “The Hasmonean State”, in M. Stern (ed.), The Hellenistic Period and the Hasmonean State (332-37 BCE) in The History of Eretz Israel, Jerusalem, pp. 191-272 (Hebrew). Rappaport, U., 1993, “The Hellenistic Cities and the Judaization of the Land of Israel under the Hasmonean Rule”, in U. Rappaport and Y. Ronen (eds.), The Hasmonean State: Its History in the Hellenistic Period, Jerusalem-Tel Aviv, pp. 231-242 (Hebrew). Ravikovitch, S., 1969, Manual and Map of Soils of Israel, Map.Scale: 1:250,000, Jerusalem (Hebrew). Reese, D.S., 1997, “The Animal Bones and Shells”, in A. Leonard, Jr. (ed.), Ancient Naukratis, Excavations at a Greek Emporium in Egypt, Part I (AASOR 54), pp. 347–351. Regev, D., 2003, “Appendix 2: Typology of the Persian and Hellenistic Pottery Forms at Maresha-Subterranean Complexes 70, 21, 58”, in A. Kloner (ed.), Maresha Excavations Final Report I: Subterranean Complexes 22, 44, 70 (IAA Reports 17), Jerusalem, pp. 163–183. Reich, R., 2003, “Stone Vessels, Weights and Architectural Fragments”, in H. Geva (ed.), Jewish Quarter Excavations in the Old City of Jerusalem, Conducted by Nahman Avigad, 1969–1982, Vol. II, Jerusalem, pp. 263–291. Reimer, P.J., Baillie, M.G.L., Bard, E., Bayliss, A., Beck, J.W., Bertrand, C., Blackwell, P.G., Buck, C.E., Burr, G., Cutler, K.B., Damon, P.E., Edwards, R.L., Fairbanks, R.G., Friedrich, M., Guilderson, T.P., Hughen, K.A., Kromer, B., McCormac, F.G., Manning, S., Bronk Ramsey, C., Reimer, R.W., Remmele, S., Southon, J.R., Stuiver, M., Talamo, S., Taylor, F.W., van der Plicht, J., Weyhenmeyer, C.E., 2004, “IntCal04:Calibration Issue”, Radiocarbon 46, pp. 1029–1058. Reisner, G.A., Fisher C.S., and Lyon, D.G., 1924, Harvard Excavations at Samaria 1908–1910, Vol. I, Cambridge M.A. Reynolds, John M., 1997, An Introduction to Applied and Environmental Geophysics, Chichester, UK. Rice, P.M, 1987, Pottery Analysis: A Sourcebook, Chicago.

Riklin, S., 1993, “‘Ofarim”, ESI 13, pp. 53–54. Riklin, S., 1997, “Beit Arye”, Atiqot XXXII, pp. 7–10 (Hebrew). Riley, J.A., 1975, “The Pottery from the First Session of Excavation in the Caesarea Hippodrome”, BASOR 218, pp. 25–63. Roberts, B.K., 1996, Landscapes of Settlement: Prehistory to Present, London. Robinson, D.M., 1938, Excavations at Olynthus: The Hellenistic House vol. VIII, Baltimore. Roll, I., 1983, “The Roman Road System in Judaea”, in L.I. Levine (ed.), The Jerusalem Cathedra, 3, Jerusalem, pp. 136–161. Roll, I., 1996, “Bacchides’ Fortifications and the Traffic Arteries to Jerusalem in the Hellenistic Period”, EretzIsrael 25, pp. 509–514 (Hebrew). Roll, I., and Dagan, Y., 1988, “Roman Roads around Beth Govrin”, in E. Stern and D. Urman (eds.), Man and Environment in the Southern Shefelah: Studies in Regional Geography and History, Giva‛atayim, pp. 175–179 (Hebrew). Ronen, I., 2003, “The Finds from Maresha Excavations and the Conversion of the Idumeans”, in M. Mor, A. Oppenheimer, J. Pastor and D.R. Schwartz (eds.), Jews and Gentiles in the Holy Land in the Days of the Second Temple, the Mishnah and the Talmud, A Collection of Articles, Jerusalem, pp. 123–131 (Hebrew). Rosenthal, R., and Sivan, R., 1978, Ancient Lamps in the Schloessinger Collection (Qedem 8), Jerusalem. Rosenthal-Heginbottom, R., 1995, “Imported Hellenistic and Roman Pottery”, in E. Stern (ed.), Excavations at Dor, Final Report, Vol. IB: Areas A and C: The Finds, Jerusalem, pp. 183–288. Rosenthal-Heginbottom, R., 2003, “Hellenistic and Early Roman Fine Ware and Lamps from Area A”, in H. Geva (ed.), Jewish Quarter Excavations in the Old City of Jerusalem, Conducted by Nahman Avigad 1969–1982, II, Jerusalem, pp. 192–223, 232–255. Rostovtzeff, M., 1959, The Social and Economic History of the Hellenistic World, Oxford. Rotroff, S., 1987, “Two Centuries of Hellenistic Pottery”, in H.A. Thompson and D.B. Thompson (eds.), Hellenistic Pottery and Terracottas, Princeton, pp. 1–8. Rotroff, S.I., 1997, Hellenistic Pottery, Athenian and Imported Wheelmade Table Ware and Related Material: The Athenian Agora: Results of Excavations Conducted by the American School of Classical Studies at Athens, Vol. XXIX, Princeton. Safrai, Z., 1994, “The Shephela of Samaria”, in Z.H. Erlich and Y. Eshel (eds.), Judea and Samaria Research Studies, Proceedings of the 4th Annual Meeting – 1994, Kedumim-Ariel, pp. 189–208. Safrai, Z., 1998, “Ancient Fields Structures – The Villages in Eretz-Israel During the Roman Period”, Cathedra 89, pp. 7–40 (Hebrew). Safrai, Z., 2000, “The Conversion of Newly Conquered Areas in Hashmonean Judea”, in J. Schwartz, et al. (eds.), Jerusalem and Eretz Israel, Arie Kindler Volume, Ramat-Gan, pp. 70–88 (Hebrew).

268

Avraham Faust and Adi Erlich

Safrai, Z., and Sasson, A., 2001, Quarrying and Quarries in the Land of Israel, Elkanah (Hebrew). Sagiv, N., and Zissu, B., 1998, “Khorvat Tabaq at Tel Goded – A Jewish Site which was Destroyed after the Bar-Kokhba Revolt”, Judea and Samaria Research Studies Proceedings of the 7th Annual Meeting – 1997, Kedumim-Ariel, pp. 115–139 (Hebrew). Salles, J.F., 1985, “A propos du niveau 4 de Tel Keisan”, Levant 17, pp. 203–204. Sandhaus, D., forthcoming, “Part II: 1: The Pottery”, in G. Mazor and D. Sandhaus (eds.), Nysa –Scythoplolis: The Hellenistic City – Tel Iztaba (IAA Reports), Jerusalem. Sasson, A., 2002, “The Cave of Zedekiah – Types of Quarrying Methods in the Land of Israel”, in Y. Eshel (ed.), Judea and Samaria Research Studies, 11, pp. 345–358 (Hebrew). Sasson, A., 2003, “Maghar: A Village of Caves from the Ottoman Period in the Coastal Plain”, Bulletin of the Anglo-Israel Archaeological Society 21, pp. 11–38. Schick, C., 1896, “Zur Einwohnerzahl des Bezircks Jerusalem”, ZDPV XIX, pp. 120–127. Schiffer, M.B., 1985, „Is There a „Pompeii Premise“ in Archaeology?“ Journal of Anthropological Research 41, pp. 18–41. Schiffer, M.B., 1987, Formation Processes of the Archaeological Record, Albuquerque. Schmidt, E.F., 1957, Persepolis II, Contents of the Treasury and Other Discoveries, Chicago. Schwartz, S., 1991, “Israel and the Nations Roundabout: 1 Maccabees and the Hasmonean Expansion”, Journal of Jewish Studies 42, pp. 16–38. Schwartz, S., 1993a, “John Hyrcanus I’s Destruction of the Gerizim Temple and Judaean-Samaritan Relations”, Jewish History 7, pp. 9–25. Schwartz, S., 1993b, “A Note on the Social Type and Political Ideology of the Hasmonean Family”, JBL 112: 305–309. Segal, O., Kletter, R., and Ziffer, O., 2006, “A PersianPeriod Building from Tel Ya‘oz (Tell Ghaza)”, Atiqot 52, pp. 1*–24*. Sellers, O.R., 1933, The Citadel of Beth-Zur, Philadelphia. Shai, I., 2006, The Material Culture of Philistia during the Iron Age IIA: The Ceramic Assemblage of Tel es-Safi as a Test Case, Unpublished Ph.D. dissertation, Bar-Ilan University, Ramat Gan (Hebrew). Shatzman, I., 2003, “The Relations Between Jews and Gentiles in the First Generations of the Hasmoneans according Contemporary Sources”, in M. Mor, A. Oppenheimer, J. Pastor and D.R. Schwartz (eds.), Jews and Gentiles in the Holy Land in the Days of the Second Temple, the Mishnah and the Talmud, A Collection of Articles, Jerusalem, pp. 142–158 (Hebrew). Shatzman, I., 2005, “On the Conversion of the Idumaens”, in M. Mor, J. Pastor, I. Ronen and Y. Ashkenazi (eds.), For Uriel: Studies in the History of Israel in Antiquity Presented to Professor Uriel Rappaport, Jerusalem, pp. 213–241 (Hebrew). Shennan, J.S., 1989, “Introduction: Archaeological Approaches to Cultural Identity”, in J.S. Shennan (ed.),

Archaeological Approaches to Cultural Identity, London, pp. 1–32. Shott, M., 1996, “Mortal Pots: On Use Life and Vessel Size in the Formation of Ceramic Assemblages”, American Antiquity 61, pp. 463-482. Singer, A., 2007, The Soils of Israel, Berlin. Singer-Avitz, L., 1989a, „Stone and Clay Objects“, in Z. Herzog, G. Rapp, and O. Negbi (eds.) Excavations at Tel Michal, Israel, Tel Aviv, pp. 350–60. Singer-Avitz, L., 1989b, “Local Pottery of the Persian Period (Strata XI–VI)”, in Z. Herzog, G. Rapp, and O. Negbi (eds.), Excavations at Tel Michal, Israel, Tel Aviv, pp. 115–144. Sinopoli, C.M., 1991, Approaches to Archaeological Ceramics, New York. Sivan, D., 1996, Paleogeography of the Galilee Coastal Plain During the Quaternary (Geological Survey of Israel Report GSI/18/96), Jerusalem (Hebrew with English summary). Sivan, D., Potasman M., Almogi-Labin, A., Bar-Yosef Mayer, D.E., Spanier, E., and Boaretto, E., 2006, “The Glycymeris Query along the Coasts and Shallow Shelf of Israel, Southeast Mediterranean”, Palaeogeography, Palaeoclimatology, Palaeoecology 233, pp. 134–148. Smithline, H., 2008, “Results of Three Small Excavations in Nahf, Upper Galillee”, Atiqot 59, pp. 87–101. Sneh, A., Bartov, Y., and Rosensaft, M., 1998, Geological Map of Israel. Scale: 1:200,00. Sheet 2. Jerusalem. SNG. 1977, Sylloge Nummorum Graecorum. The Royal Collection of Coins and Medals, Danish National Museum, Egypt: The Ptolemies, Copenhagen. Socin A., 1879, “Alphabetisches Verzeichniss von Ortschaften des Pashlik Jerusalem”, ZDPV 2, pp. 135–163. Spaer. M., 2001, Ancient Glass in the Israel Museum: Beads and Other Small Objects, Israel Museum, Jerusalem. Stacey D., 2004, Excavations at Tiberias, 1973–1974: The Early Islamic Periods (IAA Reports 21), Jerusalem. Stern, E., 1978, Excavations at Tel Mevorakh 1973–1976 (Qedem 9), Jerusalem. Stern, E., 1982, Material Culture of the Land of the Bible in the Persian Period 538–332 B.C., Warminster. Stern, E., 1994, Dor, Ruler of the Seas, Jerusalem. Stern, E., 1995, Local Pottery of the Persian Period, in E. Stern (ed.), Excavations at Dor, Final Report IB: Areas A and C: The Finds (Qedem Reports 2), Jerusalem, pp. 51–92. Stern, E., 1997, Caravansaries, Roads and Inns in Israel, Jerusalem (Hebrew). Stevenson, M.G., 1982, “Toward an Understanding of Site Abandonment Behavior: Evidence from Historic Mining Camps in the Southwest Yukon”, Journal of Anthropological Archaeology 1, pp. 237–265. Stuiver, M., Polach, H.A., 1977, “Discussing Reporting 14C Data”, Radiocarbon 19, pp. 355–363. Sussman, V., 2004, “Distribution of Oil Lamp Workshops in the Land of Israel during the Hellenistic, Roman and Byzantine Period”, Judea and Samaria Research Studies 13, pp. 149–164 (Hebrew).

269

The Excavations of Khirbet er-Rasm: The Changing Faces of the Countryside

Comparative Study”, Geoarchaeology: An International Journal 17 (4), pp. 393–415. Will, E., 1949, “La tour funéraire de la Syrie et les monuments apparentés”, Syria 26, pp. 258–312. Wood, W.R., and Johnson, D.S., 1978, “A Survey of Disturbance Processes in Archaeological Site Formation”, in M.B. Schiffer (ed.), Archaeological Method and Theory 1, New York, pp. 315–381. Woolley, C.L., 1938, “The Excavations at Al Mina, Sudeida”, Journal of Hellenistic Studies 58, pp. 133–170. Yadin, Y., 1966, Masada: Herod’s Fortress and the Zealots’ Last Stand, London. Yeivin, Z., and Edelstein, G., 1970, “Excavations at Tirat Yehuda”, ‘Atiqot (HS) 6, pp. 56–67, 6* (Hebrew). Yizhaq, M., Mintz, G., Cohen, I., Khalaily, H., Weiner, S., Boaretto, E., 2005, “Quality Controlled Radiocarbon Dating of Bones and Charcoal from the Early Pre-Pottery Neolithic B (PPNB) of Motza (Israel)”, Radiocarbon 47, pp. 193–206. Yogev, O., 1982, “Aderet”, Khadashot Arkheologiyot 78–79, pp. 81–82 (Hebrew). Zayadine, F., 1966, “Early Hellenistic Pottery from the Theatre Excavations of Samaria”, Annual of the Department of Antiquities of Jordan 11, pp. 53–64. Zemer, A., 1977, Storage Jars in Ancient Sea Trade, Haifa (Hebrew). Ziadeh-Seely, G., 1999, “Abandonment and Site Formation Processes: An Ethnographic and Archaeological Study”, in T. Kapitan (ed.), Archaeology, History and Culture in Palestine and the Near East, Atlanta, pp. 127–150. Zimhoni, O., 1997, Studies in the Iron Age Pottery of Israel: Typological, Archaeological and Chronological Aspects (Journal of the Institute of Archaeology of Tel Aviv University, Occasional Publications 2), Tel Aviv. Zimhoni, O., 2004, “The Pottery of Levels III and II”, in D. Ussishkin (ed.), The Renewed Archaeological Excavations at Lachish, vol. IV. Tel Aviv, pp. 1789–1899. Zissu, B., 1997, “Horbat Egoz”, ESI 19, pp. 85*. Zissu, B., 1999, “Horbat El Ein”, ESI 109, pp. 87*–88*. Zissu, B., 2000, „`Iyyé Nahash“, ESI 111, p. 80*. Zissu, B., 2001, Rural Settlementsin the Judean Hills and Foothills from the Late Second Temple Period to the Bar Kokhba Revolt, Unpublished Ph.D dissertation, Hebrew University of Jerusalem, Jerusalem (Hebrew). Zissu, B., 2004, “The Hellenistic Fortress at Horvat Tura in the Jerusalem Hills and Identification of Tur Shimon”, Cathedra 112, pp. 5–18 (Hebrew). Zissu, B., 2005, “A Burial Cave with a Greek Inscription and Graffiti at Khirbat el-Ein, Judean Shephelah”, Atiqot 50, pp. 27–36. Zissu, B., and Ganor, A., 2002, “Horvat ‘Etri – the Ruins of a Second Temple Period Jewish Village on the Coastal Plain”, Qadmoniot 123m pp. 18–27 (Hebrew). Zissu, B., and Rokach, S., 1999, “A Hellenistic Columbarium at Ziqim”, Atiqot 38, pp. 65–73. Ziv, Y., 2005, “Emek Ha’ela and the Military History of Eretz-Israel”, in Y. Eshel (ed.), Judaea and Samaria Research Studies: Volume 14, Ariel, pp. 213–221.

Sussman, V., 1982, Ornamented Jewish Oil Lamps, Warminster. Sussman, V., 2007, Oil-lamps in the Holy Land: Saucer Lamps from the Beginning to the Hellenistic Period (Collections of the Israel Antiquities Authority) (British Archaeological Reports International Series 1598), Oxford. Svoronos, I.N., 1904, TA NOMIΣMATA TOY KPATOYΣ TΩN ΠTOΛEMAIΩN, Athens. Syon, D., 2002, “Coins from the Excavations at Khirbet es-Shuhara”, in Z. Gal (ed.), Eretz Zafon: Studies in Galilean Archaeology, Jerusalem, pp. *123–*134. Tal, O., 2006, The Archaeology of Hellenistic Palestine: Between Tradition and Renewal, Jerusalem (Hebrew). Tcherikower, V., 1937, “Palestine under the Ptolemies”, Mizraim 4–5. Tepper, Y., 1986, “The Rise and Fall of Dove-Raising”, in A. Kasher, A. Oppenheimer and U. Rappaport (eds.), Man and Land in Eretz-Israel in Antiquity, Jerusalem, pp. 170–196 (Hebrew). Thareani-Sussely, Y., 2007, “Ancient Caravanserais: An Archaeological View from ‘Aroer”, Levant 39, pp. 123–141. Thompson, H.A., 1934, “Two Centuries of Hellenistic Pottery, The American Excavation in the Athenian Agora, Fifth Report”, Hesperia 3, pp. 311–480. Triebel, L., 2004, Jenseitshoffnung in Wort und Stein, Nefesch und Pyramidales Grabmal als Phänomene antiken jüdischen Bestattungswesens im Kontext der Nachbarkulturen, Leiden and Boston. Tufnell, O., 1953, Lachish Vol. III, The Iron Age, London. Ussishkin, D., 1993, The Village of Silwan, The Necropolis from the Period of the Judean Kingdom, Jerusalem. Vickers, M., and Gill, D., 1994, Artful Crafts: Ancient Greek Silverware and Pottery, Oxford. Vilnai, Z., 1981, Ariel – Land of Israel Encyclopedia, vol. 8, Tel-Aviv (Hebrew). Vincent, L.H., 1954, Jérusalem de l’ancien testament, recherches d’archéologie et d’histoire I: Archéologie de la ville, Paris. Von Gaertringen, F.F.H., and Wilski, P., 1904, Stadtgeschichte von Thera, Berlin. Waldbaum, J.C., 1983, Metalwork from Sardis: The Finds through 1974, London. Waner, M., and Safrai, Z., “A Catalogue of Coin Hoards and the Shelf Life of Coins in Palestine Hoards During the Roman and Byzantine Period”, Liber Annuus LI, pp. 305-336. Weiss, Z., 2005, The Sepphoris Synagogue, Jerusalem. Weksler-Bdolah, S., 2000, “Yad Benyamin”, ESI 112, pp. 98*–100*, 123–125. Weksler Bdolah, S., 2006, “The Old City Wall of Jerusalem: The Northwestern Corner”, ‘Atiqot 54, pp. *95–119 (Hebrew). White, D., 1963, “A Survey of Millstones from Morgantina”, AJA 67, pp. 199-206. Wieder, M., and Adan-Bayewitz, D., 2002, “Soil Parent Materials and the Pottery of Roman Galilee: A

270

Appendix 2.6: pottery database (complete vessels marked in gray) Appendix 2.6: pottery database - storage jars

Storage Jars L100 L101 L102 L103 L104 L105 L106 L107 L108 L109 L110 L111 L112 L113 L114 L115 L116 L117 L118 L119 L120 L121 L122 L123 L124 L125 L126 L127 L128 L129 L130 L131 L132 L133 L134 L135 L136 L137 L138 L139 L140 L141 L142 L143 L144 L145 L146 L147 L148 L149 L150 L151

SJ1A

SJ1B

SJ1C

SJ1D

SJ1E

SJ2

SJ3

SJ4

SJ6

SJ7

1

SJ8

SJ9

SJ10

SJX1

1

4 1

1 1 1

2

1

1 1 . 3

2 1 2 2

1

1 3 1

1

1

.

1 2

2

1

1

1

SJX2

L152 L153 L154 L155 L156 L157 L158 L159 L160 L161 L162 L163 L164 L165 L166 L167 L168 L169 L170 L171 L172 L173 L174 L175 L176 L177 L178 L179 L180 L181 L182 L183 L184 L185 L186 L187 L188 L189 L301 L302 L303 L304 L305 L306 L307 L308 L309 L310 L311 L312 L313 L314 L315 L316 L317 L318 L319

4

1

2

1 1

1 1 1

1 1

1

1 1

1

1 2 1

1 6

1

1

1

2 1

1 1

1 1 1

1 3

4

1 1

1

1

1

1

1 1

1 1

1 1 1

1

1

3 1

1

L320 L321 L322 L323 L324 L325 L326 L327 L328 L329 L330 L331 L332 L333 L501 L502 L503 L504 L505 L506 L507 L508 L509 L510 L511 L512 L513 L514 L515 L516 L517 L518 L519 L520 L521 L522 L523 L524 L525 L526 L527 L601 L602 L603 L604 L605 L606 L607 L700 L701 L702 L703 L704 L705 L706

1

1

1

1 1

1

1 1

1 1

1

1

1

1

1

1 1

1 1

1

3

1

1 1

1

2

1

1

1

1 1

1

1 1

1

1 1

1

1

L801 L802 L803 L804 L805 L806 L807 L808 L809 L810 L811 L812 L813 L814 L815 L816 L817 L818 L819 L820 L821 L822 L823 L824 L825 L826 L990

1 1

1

1 1 1 1 1 1 1

1 1 1 1 1 1 1

1 1 1

1 1 1 1

1

1

1 1 1 1

1 1 2 1 1

1

1

1

1

1

1 1

1 1

1

3 2

1

1

1 1 1 .

Appendix 2.6: pottery database - jugs Jugs L100 L101 L102 L103 L104 L106 L107 L108 L109 L110 L111 L112 L113 L114 L115 L116 L117 L118 L119 L120 L121 L122 L123 L124 L125 L126 L127 L128 L129 L130 L131 L132 L133 L134 L135 L136 L137 L138 L139 L140 L141 L142 L143 L144 L145 L146 L147 L148 L149 L150 L151 L152 L153 L154 L155 L156 L157 L158 L159 L160 L161 L162 L163 L164 L165 L166 L167 L168 L169 L170 L171 L172 L173 L174 L175 L176 L177 L178

R-JG1

R-JG2

R-JG3

R-JG4

R-JG5

R-JG6

R-DT1

R-DT2

1

1

1

1

1 1

1

1

1

1

1

1 1

1

Appendix 2.6: pottery database - jugs Jugs L179 L180 L181 L182 L183 L184 L185 L186 L187 L188 L189 L190 L191 L192 L193 L194 L195 L196 L300 L301 L302 L303 L304 L305 L306 L307 L308 L309 L310 L311 L312 L313 L314 L315 L316 L317 L318 L319 L320 L321 L322 L323 L324 L325 L326 L327 L328 L329 L330 L331

R-JG1

R-JG2

R-JG3

2

R-JG4

1

R-JG5

R-DT1

R-DT2

1

1

1

2

1

L500 L501 L502 L503 L504 L505 L506 L507 L508 L509 L510 L511 L512 L513 L514 L515 L516 L517 L800 L801 L802 L803 L804 L805 L806

R-JG6

1 1

1

Appendix 2.6: pottery database - jugs Jugs L807 L808 L809 L810 L811 L812 L813 L814 L815 L816 L817 L818 L819 L820 L821 L822 L823 L824 L990

R-JG1

R-JG2

R-JG3

R-JG4

R-JG5

R-JG6

R-DT1

R-DT2

1

1

1 1

1 1 1 1 1

Appendix 2.6: pottery database - juglets Juglets L100 L101 L102 L103 L104 L106 L107 L108 L109 L110 L111 L112 L113 L114 L115 L116 L117 L118 L119 L120 L121 L122 L123 L124 L125 L126 L127 L128 L129 L130 L131 L132 L133 L134 L134 L135 L136 L137 L138 L139 L140 L141 L142 L143 L144 L145 L146 L147 L148 L149 L150 L151 L152 L153 L154 L155 L156 L157 L158 L159 L160 L161 L162 L163 L164 L165 L166 L167 L168 L169 L170 L171 L172 L173 L174 L175 L176 L177

R-JT1

R-JT2

R-JT3

R-UN1

R-UN2

1

R-FK1

1

1 1

1

1

1 1 1 1

1

1

1

1 1

1

1 1

Appendix 2.6: pottery database - juglets L178 L179 L180 L181 L182 L183 L184 L185 L186 L187 L188 L189 L190 L191 L192 L193 L194 L195 L196 L300 L301 L302 L303 L304 L305 L306 L307 L308 L309 L310 L311 L312 L313 L314 L315 L316 L317 L318 L319 L320 L321 L322 L323 L324 L325 L326 L327 L328 L329 L330 L131 L333 L500 L501 L502 L503 L504 L505 L506 L507 L508 L509 L510 L511 L512 L513 L514 L515 L516 L517 L523 L800 L801 L802 L803 L804

1

1

1

1 1

1

1 1 1

2

1

1

1 1

Appendix 2.6: pottery database - juglets L805 L806 L807 L808 L809 L810 L811 L812 L813 L814 L815 L816 L817 L818 L819 L820 L821 L822 L823 L824

1 1 2

1 1

1 2 1

1

1 1 1

L100 L101 L102 L103 L104 L106 L107 L108 L109 L110 L111 L112 L113 L114 L115 L116 L117 L118 L119 L120 L121 L122 L123 L124 L125 L126 L127 L128 L129 L130 L131 L132 L133 L134 L135 L136 L137 L138 L139 L140 L141 L142 L143 L144 L145 L146 L147 L148 L149 L150 L151 L152

R-CP1

R-CP2

1

1

R-CP3

1

1

R-CP4

1

R-CP5

1

1

1

1

R-CP6

R-CP7A

1

1

R-CP7B

R-CP7C

1

R-CP8

R-CP9

Appendix 2.6: pottery database - cooking pots R-CP10

R-CS1

R-CS2

R-CS3

L300 L301 L302 L303 L304 L305 L306

L153 L154 L155 L156 L157 L158 L159 L160 L161 L162 L163 L164 L165 L166 L167 L168 L169 L170 L171 L172 L173 L174 L175 L176 L177 L178 L179 L180 L181 L182 L183 L184 L185 L186 L187 L188 L189 L190 L191 L192 L193 L194 L195 L196

R-CP1

1

R-CP2

1

1

1

1

R-CP3

1

1

1

R-CP4

R-CP5

2

R-CP6

R-CP7A

1

R-CP7B

R-CP7C

2

1

1

1

R-CP8

R-CP9

Appendix 2.6: pottery database - cooking pots

1

1

R-CP10

R-CS1

R-CS2

1

R-CS3

L800 L801 L802 L803 L804 L805

L500 L501 L502 L503 L504 L505 L506 L507 L508 L509 L510 L511 L512 L513 L514 L515 L516 L517 L523

L307 L308 L309 L310 L311 L312 L313 L314 L315 L316 L317 L318 L319 L320 L321 L322 L323 L324 L325 L326 L327 L328 L329 L330 L332

R-CP1

1

R-CP2

2

1

1

1

R-CP3

R-CP4

1

R-CP5

R-CP6

R-CP7A

1

1

1

2

1

R-CP7B

R-CP7C

1

2

1

R-CP8

R-CP9

Appendix 2.6: pottery database - cooking pots

2

1

1

1

2

R-CP10

1

R-CS1

R-CS2

R-CS3

L806 L807 L808 L809 L810 L811 L812 L813 L814 L815 L816 L817 L818 L818 L819 L820 L821 L822 L823 L824 L825

R-CP1

1

1

1

R-CP2

2

1 2

1 1

R-CP3

1

1

R-CP4

R-CP5

1

1

1

1 1

R-CP6

1

1

R-CP7A

1 1

‫מ‬

1

1

1

R-CP7B

1

1

R-CP7C

2

1

1

R-CP8

1

R-CP9

Appendix 2.6: pottery database - cooking pots

1

1

R-CP10

R-CS1

1

R-CS2

R-CS3

1

Appendix 2.6: pottery database - plates Plates L100 L101 L102 L103 L104 L106 L107 L108 L109 L110 L111 L112 L113 L114 L115 L116 L117 L118 L119 L120 L121 L122 L123 L124 L125 L126 L127 L128 L129 L129 L130 L131 L132 L133 L134 L135 L136 L137 L138 L139 L140 L141 L142 L143 L144 L145 L146 L147 L148 L149 L150 L150 L151 L152 L153 L154 L155 L156 L157 L158 L159 L160 L161 L162 L163 L164 L165 L166 L167 L168 L169 L170 L171 L172 L173 L174 L175 L176 L177 L178

R-PL1A

R-PL1B

R-PL1C

R-PL2

R-PL3

1

1

1

1

1 1

1

3 1

1

1

2

Appendix 2.6: pottery database - plates Plates L179 L180 L181 L182 L183 L184 L185 L186 L187 L188 L189 L190 L191 L192 L193 L194 L195 L196 L300 L301 L302 L303 L304 L305 L306 L307 L308 L309 L310 L311 L312 L313 L314 L315 L316 L317 L318 L319 L320 L321 L322 L323 L324 L325 L326 L327 L328 L329 L330

R-PL1A

R-PL1B

R-PL1C

R-PL2

R-PL3

1

1 1

1

L500 L501 L502 L503 L504 L505 L506 L507 L508 L509 L510 L511 L512 L513 L514 L515 L516 L517 L800 L801 L802 L803 L804 L805 L806 L807 L808 L809

1

1

1

Appendix 2.6: pottery database - plates Plates L810 L811 L812 L813 L814 L815 L816 L817 L818 L819 L820 L821 L822 L823 L824 L825

R-PL1A

R-PL1B

R-PL1C

R-PL2

R-PL3

1 1 1

1

1

1

1

Appendix 2.6: pottery database - kraters R-KR1 L100 L101 L102 L103 L104 L106 L107 L108 L109 L110 L111 L112 L113 L114 L115 L116 L117 L118 L119 L120 L121 L122 L123 L124 L125 L126 L127 L128 L129 L130 L131 L132 L133 L134 L135 L136 L137 L138 L139 L140 L141 L142 L143 L144 L145 L146 L147 L148 L149 L150 L151 L152 L153 L154 L155 L156 L157 L158 L159 L160 L161 L162 L163 L164 L165 L166 L167 L168 L169 L170 L171 L172 L173 L174 L175 L176 L177 L178

R-KR2

1

R-KRX1

R-KRX2

R-BS1

R-MT1

R-MT2

R-TA1

R-TA2

1

1

1 1

2 1

Appendix 2.6: pottery database - kraters R-KR1

R-KR2

R-KRX1

R-KRX2

R-BS1

R-MT1

R-MT2

L179 L180 L181 L182 L183 L184 L185 L186 L187 L188 L189 L190 L191 L192 L193 L194 L195 L196

L800 L801 L802 L803 L804 L805 L806 L807

R-TA2

1

1

L300 L301 L302 L303 L304 L305 L306 L307 L308 L309 L310 L311 L312 L313 L314 L315 L316 L317 L318 L319 L320 L321 L322 L323 L324 L325 L326 L327 L328 L329 L330 L500 L501 L502 L503 L504 L505 L506 L507 L508 L509 L510 L511 L512 L513 L514 L515 L516 L517

R-TA1

1 1 1

2

1

1

1

1 1

1 1 5

Appendix 2.6: pottery database - kraters R-KR1 L808 L809 L810 L811 L812 L813 L814 L815 L816 L817 L818 L819 L820 L821 L822 L823 L824 L990 L999

R-KR2

R-KRX1

R-KRX2

R-BS1

R-MT1

R-MT2

1

R-TA1

R-TA2

1

1 1 1 2

1

1 (2)

1

1

1 2

1

Bowls L100 L101 L102 L103 L104 L106 L107 L108 L109 L110 L111 L112 L113 L114 L115 L116 L117 L118 L119 L120 L121 L122 L123 L124 L125 L126 L127 L128 L129 L130 L131 L132 L133 L134 L135 L136 L136 L137 L138 L139 L140 L141 L142 L143 L144 L145 L146 L147 L147 L148 L149 L150 L151 L152 L153 L154 L155 L156 L157 L158 L159 L160 L161 L162 L163 L164

R-BL1

1

1

1

1

1

2

2 2

1

1

2 1

2

1

1

1 1

1

2

1

1

R-BL2

1

R-BL1B

1

R-BL1A

R-BL3

1

R-BL4

1

2

R-BL5

R-BL6

R-BL7

R-BL8

R-BLX1

R-BLX2

Appendix 2.6 : pottery database - bowls

1

1

R-BLX3

1

1

1

R-BLX4 R-BLX5 R-BLX6 R-BLX7 R-CUPX1 R-BLbyz1 R-BLByz2 R-BL\KR

L300 L301 L302 L303 L304 L305 L306 L307 L308 L309 L310 L311 L312 L313 L314 L315 L316 L317 L318 L319 L320 L321 L322 L323 L324 L325 L326 L327 L328 L329 L330 L332

Bowls L165 L166 L167 L168 L169 L170 L171 L172 L173 L174 L175 L176 L177 L178 L179 L180 L181 L182 L183 L184 L185 L186 L186 L187 L188 L189 L190 L191 L192 L193 L194 L195 L196

R-BL1

1

2

1

R-BL1A

1

1

4

1

1

5

1

1

1

1

1

1

1

1

1

11

1 2 4

2

2

1

16 1

R-BL1B

R-BL2

5 1

1

3

1

1

2 1

9 2

1

1 1

R-BL3

1

R-BL4

1

1

1 1

R-BL5

R-BL6

R-BL7

R-BL8

R-BLX1

1

R-BLX2

Appendix 2.6 : pottery database - bowls

1

R-BLX3

1

1 1

1

1

1

R-BLX4 R-BLX5 R-BLX6 R-BLX7 R-CUPX1 R-BLbyz1 R-BLByz2 R-BL\KR

1

1

1

1

1

1

L990 L999

L800 L801 L802 L803 L804 L805 L805 L806 L807 L808 L809 L810 L811 L812 L813 L814 L814 L815 L816 L817 L818 L819 L820 L821 L821 L822 L823 L823 L824 L825

3 1

9 1 5

1

1

2 1

3

1

1

1

2

1 1

1 1 1

2

3

1

R-BL1B

L705

R-BL1A

1

R-BL1

L604

L500 L501 L502 L503 L504 L505 L506 L507 L508 L509 L510 L511 L512 L513 L514 L515 L516 L517 L518 L521 L523

Bowls l333 L334 L335

R-BL2

2 2 1 2

1 6

1

2 1

1

1

2 1 1

1

1

1

R-BL3

1

R-BL4

R-BL5

1

R-BL6

1

R-BL7

1

R-BL8

1

R-BLX1

R-BLX2

Appendix 2.6 : pottery database - bowls

1

R-BLX3

1

1 1 1

1

1

R-BLX4 R-BLX5 R-BLX6 R-BLX7 R-CUPX1 R-BLbyz1 R-BLByz2 R-BL\KR

1

Weeds & wild plants Total Unidentified Grand Total

Pulses Total Weeds & wild plants

Fruits Total Pulses

Cereals Total Fruits

Group Cereals

Arnebia (mericarp) Boraginaceae (mericarp) Gramineae (grain) leguminosae (seed) Liliceae/Polygonaceae (seed) Malva (seed) Silene (seed) Thymelaeaceae (nutlet)

Lens (seed) Lens? (seed)

Olea europaea (stone) Vitis vinifera (flower?) Vitis vinifera (pip)

Plant name (organ) Hordeum (grain) Triticum (grain) Triticum? (grain)

1

1

1

3

1 1

2

2

1

1

1

1

1

1

1

1

1

1

1

1

1 1 1

1

1

2

2

2

4

4

4

1

1

1

1

1

1

1

1

1

1 1 19

1

15 1 1 2

15

3

3

3

4

2

1

1

1

1 1

1

1

1

1

17

1 2

1

15

15

1

1

1

2

2

2

2

2

2

30

1 1 4

1 1

25

1 25

1

1

1

1

10

10

10

1

1

1

12

12

12

4

3

1 1 3

1 4

1 1

2

2

6

1

3 1

2 3

2

12

12

12

1

1

1

1

1

1

1

1

1

6

3

1

2

1

2 1

2

4

1

1

2 3

1

2

2

2

2

2

2

5 1 43

1

1 1 1 2

4 2 1 7 25 1 3 29 1

1

1

1

1

1

1

1

1

1

3

3

3

A B C E H Z1 Total B Total C Total E Total H Total Z1 Total Area A Square F49 G48 I48 I52 I53 J51 J52 D51 D54 E49 E54 F53 G53 C2 C4 C8 E49 F49 G48 H48 G48 G55 Lucus 808 815 160 162 164 172 170 184 126 115 116 121 128 320 325 819 321 313 323 328 521 505 523 815 825 807 808 811 822 810 818 824 803 823 810 602 605

Appendix 3.6.1: Number of identified seeds of each taxon represented at Kh. er-Rasm.

4

1

1

3

3

2

2

2

Z2 Total Z2 (blank) 607

2

2

2

4 3 1 8 83 1 3 87 2 1 3 1 1 3 1 1 2 1 1 11 2 111

Total

Appendix 3.6.2: Distribution and density of plant remainsretrieved from Kh. er-Rasm.

Locus 116 121 126 128 160 162 164 170 172 184 313 320 321 323 325 328 505 521 523 602 605 607 803 807

808

810 811 815 818 819 822 823 824 825 Grand Total

Basket No. 1045 1057 1061 1063 1174 1177 1165 1214 1217 1250 3030 3055 3098 3050 3069 3008 3076 5007 5052 5056 6002 6007 6021 8025 8038 8046 8062 8019 8023 8032 8047 8051 8027 8060 8028 8077 8080 8077 8082 8086 8101 8100 8099

Volume (liter soil) 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 344

Quantity (# of seeds) 1 1 2 1 1 1 1 1 1 1 1 3 17 1 1 4 2 10 1 1 1 3 2 2 2 1 3 1 3 1 3 5 1 1 1 3 4 6 1 1 2 4 4 107

Density (seeds/liter) 0.125 0.125 0.25 0.125 0.125 0.125 0.125 0.125 0.125 0.125 0.125 0.375 2.125 0.125 0.125 0.5 0.25 1.25 0.125 0.125 0.125 0.375 0.25 0.25 0.25 0.125 0.375 0.125 0.375 0.125 0.375 0.625 0.125 0.125 0.125 0.375 0.5 0.75 0.125 0.125 0.25 0.5 0.5 0.311

Appendix 3.8.1: Shells uncovered at Kh. er-Rasm

area A A A A A A A A A A A A A A A A B B C C C C C C C C C C C E E E E

square locus basket layer/period I53 170 1201 Hellenistic I53 186 1258 Hellenistic I53 186 1258 Hellenistic I53 186 1261 Hellenistic I53 186 1261 Hellenistic I52 152 1262 Hellenistic I52 152 1262 Hellenistic I52 152 1262 Hellenistic Z53 186 1264 Hellenistic I51 183 1265 topsoil I51 183 1265 topsoil G54 187 1266 topsoil I53 186 1268 Hellenistic I53 186 1270 Hellenistic I53 186 1270 Hellenistic I53 186 1270 Hellenistic G53 327 3057 Hellenistic? G53 324 3079 Hellenistic? C2 518 5050 ? C2 518 5050 ? C2 519 5051 ? C2 521 5052 ? C2 521 5052 ? C2 521 5052 ? C2 521 5052 ? C2 522 5055 ? C2 523 5056 topsoil C2 523 5056 topsoil C8 523 5063 topsoil F49 808 8023 Hellenistic H48 813 8054 Hellenistic H48 813 8071 Hellenistic G48 818 8072 Hellenistic

context wall fall\fill occupation occupation occupation occupation occupation occupation occupation occupation topsoil topsoil topsoil occupation occupation occupation occupation make up ? make up ? natuarl accumulation natuarl accumulation make up ? make up ? make up ? make up ? make up ? natuarl accumulation topsoil topsoil topsoil pit Wall fall Wall fall occupation

species type Levantina Levantina Donax monacha Helix Levantina Monacha Sphincteroc Levantina Helix Levantina Levantina Helix Levantina Donax Gv biv frag of la Gv Calaxis Levantina Helix Trochoidea? Helix Sphincteroc Glycymeris Helix Sphincteroc Donax Donax trunc Buliminus? Helix Helix Levantina

quantity 1980 1982 2400 1980 1981 1980 1980 1981 1981 1980 1980 1980 1980 1980 2400 2451 2991 2450 1980 1981 1981 1980 1981 1981 2451 1981 1981 2400 2400 1980 1981 1980 1980

summaries 3 2 1 4 2 1 1 1 1 1 1 1 2 1 1 1 1 1 4 1 1 1 4 1 1 2 1 2 1 1 1 1 1

5

9

BAR S2187 2011

The Excavations of Khirbet er-Rasm, Israel The changing faces of the countryside

FAUST & ERLICH

Avraham Faust Adi Erlich

THE EXCAVATIONS OF KHIRBET ER-RASM, ISRAEL

B A R

BAR International Series 2187 2011