Stone Tools in the Ancient Near East and Egypt: Ground stone tools, rock-cut installations and stone vessels from Prehistory to Late Antiquity ... Ancient Near Eastern Archaeology) 1789690609, 9781789690606

Citation preview

Stone Tools in the Ancient Near East and Egypt Ground stone tools, rock-cut installations and stone vessels from Prehistory to Late Antiquity edited by

Andrea Squitieri and David Eitam

Archaeopress Ancient Near Eastern Archaeology 4

Stone Tools in the Ancient Near East and Egypt Ground stone tools, rock-cut installations and stone vessels from Prehistory to Late Antiquity

edited by

Andrea Squitieri and David Eitam

Archaeopress Archaeology

Archaeopress Publishing Ltd Summertown Pavilion 18-24 Middle Way Summertown Oxford OX2 7LG www.archaeopress.com

ISBN 978-1-78969-060-6 ISBN 978-1-78969-061-3 (e-Pdf)

© Authors and Archaeopress 2019 Cover illustration: Threshing floor with many rock-cut cupmarks and 4 shallow basins. Tel Bareqet (Israel). Photo by David Eitam.

All rights reserved. No part of this book may be reproduced, or transmitted, in any form or by any means, electronic, mechanical, photocopying or otherwise, without the prior written permission of the copyright owners. Printed in England by Oxuniprint, Oxford This book is available direct from Archaeopress or from our website www.archaeopress.com

To Karen Wright, for establishing a new approach to the study of ground stone tools

Contents

List of Figures and Tables�����������������������������������������������������������������������������������������������������������������������������������iii List of Authors������������������������������������������������������������������������������������������������������������������������������������������������� viii Introduction������������������������������������������������������������������������������������������������������������������������������������������������������ 1 David Eitam and Andrea Squitieri Methodology and Classification The archaeology of discard and abandonment: presence and absence in the ground stone assemblage from Early Neolithic Bestansur, Iraqi Kurdistan�������������������������������������������������������������������������� 9 David Mudd Survey of Rock-Cut Installations at Tel Bareqet (Israel): Food Processor devices in Epipaleolithic, PPNA and the Early Bronze �����������������������������������������������������������������������������������������������������������������������������27 David Eitam Ayn Asil and Elephantine (Egypt): remarks on classification and function of ground stone implements��������������������������������������������������������������������������������������������������������������������������������52 Clara Jeuthe Documentation: Non-Archaeological and Archaeological Sources in Comparison Mill-songs. The soundscape of collective grinding in the Bronze and Iron Age Near East and eastern Mediterranean�����������������������������������������������������������������������������������������������������������������������������71 Luca Bombardieri Rotary Querns and the Presentation of the Past���������������������������������������������������������������������������������������������81 Jennie Ebeling Bourgul in Talmudic and Classical Literature, and Today������������������������������������������������������������������������������93 Rafael Frankel Wine and oil presses in the Roman to Late Antique Near East and Mediterranean: Balancing textual and archaeological evidence����������������������������������������������������������������������������������������������97 Tamara Lewit and Paul Burton Raw Material and Manufacture Tool marks on Old Kingdom limestone vessels from Abusir – production of canopic jars and model vessels������������������������������������������������������������������������������������������������������������������������������������������113 Lucie Jirásková Raw material variety and acquisition of the EB III ground stone assemblage of Tell es-Safi/Gath (Israel)��������������������������������������������������������������������������������������������������������������������������������121 Jeremy A. Beller, Haskel J. Greenfield, Mostafa Fayek, Itzhaq Shai, and Aren M. Maeir Function and Uses The ground stone assemblage from the Early Bronze Age I site Wadi Fidan 4: Gender aspects������������������153 Yael Abadi-Reiss, Mohammad Najjar and Thomas E. Levy i

Cereal processing in stone agri-technological system at late Natufian Huzuq Musa in the Jordan Valley���������������������������������������������������������������������������������������������������������������������������������������162 David Eitam Cuboid-Spheroid Stone Object – an Archaic Scale Weight – Public Weighting-Systems in Iron Age Israel�������������������������������������������������������������������������������������������������������������������������������������������179 David Eitam Groundstone Tools from Site 35 – an Early Iron Age Copper Smelting Site in the Timna Valley (Israel)�������������������������������������������������������������������������������������������������������������������������������189 Aaron Greener and Erez Ben-Yosef The Iron Age stone tool assemblage of Gird-i Bazar, in the Kurdish Autonomous Region of Iraq��������������210 Andrea Squitieri Sites and Tools Macrolithics and the on-going use of stone tools in Qantir-Piramesse and Tell el-Dabʿa-Avaris, Eastern Delta/Egypt���������������������������������������������������������������������������������������������������������������������������������������223 Silvia Prell Millstones, Mortars, and Stone Bowls from Tel Dover and the Southern Levant����������������������������������������234 Refael Frankel Stone Tools of the Iron Age Ein Gev and their Implication. The Japanese Excavation Project �������������������278 David Eitam Selenite (gypsum) from the North Sinai collection: likely function and technology of production������������299 Joan S. Schneider, David Valentine, Avraham Gabay, and Eliezer D. Oren The stone tools and vessels from Tel Miqne-Ekron: a report on the Bronze and Iron Ages������������������������305 Ianir Milevski El-Khirba: Food processing and other ground stone tools from a Roman, Abbasid and Mamluk period site near Nes Ziyyona, Israel������������������������������������������������������������������������������������������������345 Erez Adama, Uzi ‘Ad and Danny Rosenberg

ii

List of Figures and Tables

D. Mudd: The archaeology of discard and abandonment

Figure 1. Plan of the Bestansur site���������������������������������������������������������������������������������������������������������������������������������������������������������������� 12 Figure 2. Composite plan of excavated walls, spaces and features in Trench 10, end of Spring 2017 season������������������������������������ 14 Figure 3. Net sinkers SF0317 in situ���������������������������������������������������������������������������������������������������������������������������������������������������������������� 17 Figure 4. Stone from SF0317 showing worm casing and boring���������������������������������������������������������������������������������������������������������������� 17 Figure 5. Debitage from stoneworking C1752����������������������������������������������������������������������������������������������������������������������������������������������� 18 Figure 6. Plan of Trench 7�������������������������������������������������������������������������������������������������������������������������������������������������������������������������������� 19 Figure 7. Space 16, before excavation, facing northwest, showing gridlines������������������������������������������������������������������������������������������ 19 Figure 8. Space 16 C1243 and C1255, showing ground stone classes and identification numbers������������������������������������������������������� 20 Figure 9. Flat oval river cobbles BF520 and BF522 in situ��������������������������������������������������������������������������������������������������������������������������� 21 Figure 10. Possible stages in the life history of a stone tool���������������������������������������������������������������������������������������������������������������������� 22 Table 1. Comparison of the ground stone artefact numbers in the main classes excavated from Early Neolithic Zagros sites����� 15 Table 2. Number of axes/celts and digging tools at selected Early Neolithic sites�������������������������������������������������������������������������������� 16

D. Eitam: Survey of Rock-Cut Installations at Tel Bareqet (Israel)

Figure 1. General plan of the Tel and excavations areas����������������������������������������������������������������������������������������������������������������������������� 28 Figure 2. General plan of the rock-cut installations������������������������������������������������������������������������������������������������������������������������������������ 30 Figure 3. L. 3020: EB winepress with pressing surface drained to collecting vat������������������������������������������������������������������������������������ 31 Figure 4. L. 3020: collecting vat of EB winepress������������������������������������������������������������������������������������������������������������������������������������������ 31 Figure 5. L. 3030: round large basin cut into oblong bedrock, a cupmark, and a small round basin��������������������������������������������������� 31 Figure 6. L. 3032 and 3033: cupmarks on flat rock face������������������������������������������������������������������������������������������������������������������������������� 31 Figure 7. Plan of L. 3039: PPNA threshing floor with cupmarks cut on an unlevelled oblong bedrock exposure����������������������������� 32 Figure 8. L. 3039: PPNA threshing floor with many rock-cut cupmarks and 4 shallow basins������������������������������������������������������������� 32 Figure 9. General plan of L. 3041: complex RCI with Late Natufian narrow conical mortar����������������������������������������������������������������� 33 Figure 10. L. 3041.20: Natufian narrow conical mortar halted by a hard stone at bottom�������������������������������������������������������������������� 33 Figure 11. L. 3042.21: Natufian narrow conical mortar, reused and cut into an EB oval deep basin��������������������������������������������������� 33 Figure 12. L. 3941.22: EB deep concave basin������������������������������������������������������������������������������������������������������������������������������������������������ 34 Figure 13. L. 3942.24: Natufian small concave conical mortar������������������������������������������������������������������������������������������������������������������� 34 Figure 14. L. 3042: General view of large bedrock with 31 cupmarks, and EB round deep basin��������������������������������������������������������� 34 Figure 15. L. 3042.1: deep concave EB basin, with reused Natufian narrow conical mortar����������������������������������������������������������������� 34 Figure 16. Plans of L. 3042.1: with EB round deep basin; L. 3063.1: oval deep EB basin, 2, 3 and 5: small deep basins��������������������� 35 Figure 17. L. 3053: cupmarks cut on top of a rock���������������������������������������������������������������������������������������������������������������������������������������� 35 Figure 18. L. 3063.1-3: oval deep EB basin, and 2 small deep basins��������������������������������������������������������������������������������������������������������� 36 Figure 19. L. 3948: oval deep basin cut by a shallow small bowl���������������������������������������������������������������������������������������������������������������� 36 Figure 20. L. 3063: EBII oval basin which reuses a Natufian narrow conical mortar as collection vat������������������������������������������������ 37 Table 1. Types of rock-cut installations��������������������������������������������������������������������������������������������������������������������������������������������������������� 38 Appendix A. Rock-cut installations at Tel Bareqet�������������������������������������������������������������������������������������������������������������������������������������� 42

C. Jeuthe: Ayn Asil and Elephantine (Egypt)

Figure 1. Overview Balat and research areas������������������������������������������������������������������������������������������������������������������������������������������������ 53 Figure 2. Overview Elephantine Island and research areas������������������������������������������������������������������������������������������������������������������������ 54 Figure 3. Tools of the first category from Ayn Asil�������������������������������������������������������������������������������������������������������������������������������������� 58 Figure 4. Tools of the second category from Ayn Asil��������������������������������������������������������������������������������������������������������������������������������� 59 Figure 5. Tools of the third category from Ayn Asil������������������������������������������������������������������������������������������������������������������������������������ 60 Figure 6. Tools of the fourth category from Ayn Asil���������������������������������������������������������������������������������������������������������������������������������� 62 Table 1. Ayn Asil: raw materials of stone implements attested and their frequency rates������������������������������������������������������������������ 63 Table 2. Elephantine: raw materials of stone implements attested and their frequency rates����������������������������������������������������������� 64 Table 3. Balat/Sheikh Muftah: raw materials of stone implements attested and their frequency rates�������������������������������������������� 65 Table 4. Frequency rates of functional categories within the sites and individual research areas����������������������������������������������������� 66

L. Bombardieri: Mill-songs

Figure 1. Susa. Uruk period. Clay sealing with grinding scene������������������������������������������������������������������������������������������������������������������ 71 Figure 2. Balawat/Imgur-Enlil. Neo-Assyrian period. Scene incised on decorated bronze bands, with two soldiers in the Assyrian camp of king Shalmaneser III������������������������������������������������������������������������������������������������������������������������������������������� 72 Figure 3. Episkopi (?). Cypro-Archaic period. Terracotta depicting a grinding scene��������������������������������������������������������������������������� 72 Figure 4. Ninive. SW Palace. Room XXXIII. Particular of the wall relief with two Babylonian prisoners forced to grind their father’s bones������������������������������������������������������������������������������������������������������������������������������������������������������������������������������������ 73 Figure 5. Provenace unknown. RP amphora with modelled complex scenic composition������������������������������������������������������������������� 74 Figure 6. Provenance unknown. Red Slip terracotta����������������������������������������������������������������������������������������������������������������������������������� 75 Figure 7. Provenance unknown. Red Slip terracotta����������������������������������������������������������������������������������������������������������������������������������� 76 Figure 8. Thebes, Boeotia. Terracotta������������������������������������������������������������������������������������������������������������������������������������������������������������ 76

iii

J. Ebeling: Rotary Querns and the Presentation of the Past

Figure 1. ‘Two women at the mill.’����������������������������������������������������������������������������������������������������������������������������������������������������������������� 82 Figure 2. A rotary quern still in use in northern Jordan����������������������������������������������������������������������������������������������������������������������������� 83 Figure 3. National Heritage Museum at the University of Jordan, Amman, Jordan������������������������������������������������������������������������������� 84 Figure 4. Joe Alon Museum of Bedouin Culture, Lahav, Israel������������������������������������������������������������������������������������������������������������������� 84 Figure 5. Museum of Jordanian Heritage at Yarmouk University, Irbid, Jordan������������������������������������������������������������������������������������� 85 Figure 6. Madaba Archaeological and Folklore Museum, Madaba, Jordan���������������������������������������������������������������������������������������������� 86 Figure 7. Palestinian Heritage Center, Bethlehem, Palestine��������������������������������������������������������������������������������������������������������������������� 86 Figure 8. Image from the collection of the Palestinian Heritage Center, Bethlehem, Palestine���������������������������������������������������������� 87 Figure 9. Bedouin Heritage Center, Shibli, Mt. Tabor, Israel���������������������������������������������������������������������������������������������������������������������� 87 Figure 10. Rotary quern from the Golan Heights����������������������������������������������������������������������������������������������������������������������������������������� 88 Figure 11. Haret Jdoudna Restaurant, Madaba, Jordan������������������������������������������������������������������������������������������������������������������������������� 89 Figure 12. Beit Al-Turath Handicraft Company, Amman, Jordan�������������������������������������������������������������������������������������������������������������� 89 Figure 13. A private garden in Amman, Jordan�������������������������������������������������������������������������������������������������������������������������������������������� 90 Figure 14. Embroidered artwork of a woman with a rotary quern����������������������������������������������������������������������������������������������������������� 90

R. Frankel: Bourgul in Talmudic and Classical Literature, and Today

Figure 1. Crushing the grain, Yarka, Western Galilee���������������������������������������������������������������������������������������������������������������������������������� 94 Figure 2. The four sieves of the Salman abu Yusef family from Yarka, Western Galilee����������������������������������������������������������������������� 95

T. Lewit and P. Burton: Wine and oil presses in the Roman to Late Antique Near East and Mediterranean

Figure 1. Lever and drum press, reconstructed according to Cato’s description in 1996��������������������������������������������������������������������� 98 Figure 2. Traditional lever and screw press, with a mobile weight, similar to that used in ancient Northern Syria and elsewhere in the Levant����������������������������������������������������������������������������������������������������������������������������������������������������������������������������������������������� 98 Figure 3. Traditional direct screw press, similar to that used in ancient southern Levant and elsewhere��������������������������������������� 98 Figure 4. Reconstruction of the lever and weights press at Oilery FVIII, Chhîm, Lebanon������������������������������������������������������������������ 99 Figure 5. Typical limestone weight from a Levantine style lever and weights press, probably Byzantine�������������������������������������� 100 Figure 6. View of late 5th-6th century lever and weights press at Oilery FVIII, Chhîm, Lebanon���������������������������������������������������� 100

L. Jirásková: Tool marks on Old Kingdom limestone vessels from Abusir

Figure 1. Base of the canopic jar 24/AS37/2007 with copper chisel cut marks������������������������������������������������������������������������������������ 114 Figure 2. Cross-section of the canopic jar 5/AS37/2007��������������������������������������������������������������������������������������������������������������������������� 115 Figure 3a. Cross section of the canopic jar 24/AS37/2007, which was completely gouged out using a copper chisel������������������� 115 Figure 3b. Detail of the interior of 24/AS37/2007 with unsmoothed cut marks���������������������������������������������������������������������������������� 116 Figure 4. Wavy interior of the canopic jar 388_1/AS68/2014, which was modelled by a figure-of-eight stone drill��������������������� 116 Figure 5. Roughly shaped unsmoothed exterior of the bowl 19_9/AS37/2007 with planes left after the use of a copper adze����� 117 Figure 6a. Depression of the bowl 6_27/AS67/2012 with triangular boring traces������������������������������������������������������������������������������ 117 Figure 6b. Depression of the bowl 16_30/AS67/2012 with circular boring traces�������������������������������������������������������������������������������� 117 Figure 7. Depression of the bowl 19_25/AS37/2007 which was not bored but gouged out using a copper chisel�������������������������� 118 Figure 8. The symbolic shallow interior of the jar 383_16/AS68/2014 was worked with a copper chisel or a pick������������������������ 118

J. A. Beller et al.: Raw material variety and acquisition of the EB III ground stone assemblage

Figure 1. Notable sites of the EB II-III southern Levant���������������������������������������������������������������������������������������������������������������������������� 122 Figure 2. Excavation areas of Tell es-Safi/Gath������������������������������������������������������������������������������������������������������������������������������������������ 124 Figure 3. Typology of ground stone assemblage (by count)�������������������������������������������������������������������������������������������������������������������� 125 Figure 4. Typology of ground stone assemblage (by % frequency)��������������������������������������������������������������������������������������������������������� 125 Figure 5. Lower grinding stone (Basket #16E94A011)������������������������������������������������������������������������������������������������������������������������������� 129 Figure 6. Upper grinding stone (Basket #16E84A007)������������������������������������������������������������������������������������������������������������������������������� 130 Figure 7. Perforated stone weight (Basket #16E83C087)��������������������������������������������������������������������������������������������������������������������������� 130 Figure 8. Vessel (Basket #1143010)��������������������������������������������������������������������������������������������������������������������������������������������������������������� 130 Figure 9. Pounder (Basket #748124)������������������������������������������������������������������������������������������������������������������������������������������������������������� 130 Figure 10. Mortar/socket (Basket #748122)������������������������������������������������������������������������������������������������������������������������������������������������ 131 Figure 11. Material of ground stone objects����������������������������������������������������������������������������������������������������������������������������������������������� 131 Figure 12. XRD results of limestone artefact (Basket #845008)��������������������������������������������������������������������������������������������������������������� 132 Figure 13. Breakdown of material and select ground stone types���������������������������������������������������������������������������������������������������������� 133 Figure 14. Selected basalt sources of Near East������������������������������������������������������������������������������������������������������������������������������������������ 136 Figure 15. Total alkali-silica diagram of samples for sources������������������������������������������������������������������������������������������������������������������ 138 Figure 16. Total alkali-silica diagram of source averages������������������������������������������������������������������������������������������������������������������������� 138 Figure 17. SiO2 vs. Na2O+K2O scatter plot of sources and artifacts samples������������������������������������������������������������������������������������������� 140 Figure 18. SiO2 vs. Na2O+K2O plot of source averages and artifact samples along with baskets associated with the discussed groups������������������������������������������������������������������������������������������������������������������������������������������������������������������������������� 141 Figure 19a. TiO2 vs. SiO2 plot of selected source averages and artifact samples (baskets). 19b. TiO2 vs. Fe2O3(t) plot of selected source averages and Basket #1042042.������������������������������������������������������������������������������������������������������������������������������ 142 Figure 20. Movement of ground stone commodities to Tell es-Safi/Gath��������������������������������������������������������������������������������������������� 144 Table 1. Selected characteristics of EB III Tell es-Safi/Gath ground stone catalogue�������������������������������������������������������������������������� 126 Table 2. Averaged geochemical profile for Near Eastern and Egyptian basalt sources����������������������������������������������������������������������� 139

iv

Table 3. ED-XRF data for EB III ground stone tools from Area E of Tell es-Safi/Gath�������������������������������������������������������������������������� 140 Table 4. Association between Near East basalt sources and Tell es-Safi/Gath artifacts���������������������������������������������������������������������� 143

Y. Abadi-Reiss et al.: The ground stone assemblage from the Early Bronze Age I site Wadi Fidan 4

Figure 1. Site location map���������������������������������������������������������������������������������������������������������������������������������������������������������������������������� 154 Figure 2. Grinding stones: A, B, Lower grinding slabs. C, D, Upper grinding stone������������������������������������������������������������������������������ 155 Figure 3. Selected hammer stones���������������������������������������������������������������������������������������������������������������������������������������������������������������� 156 Figure 4. The mortars������������������������������������������������������������������������������������������������������������������������������������������������������������������������������������� 157 Figure 5. The pestles��������������������������������������������������������������������������������������������������������������������������������������������������������������������������������������� 157 Figure 6. Perforated stones: A, B, and the axes: C, D��������������������������������������������������������������������������������������������������������������������������������� 158 Table 1. Ground stone assemblage from WFD4������������������������������������������������������������������������������������������������������������������������������������������ 154 Table 2. WFD4 upper grinding stone characteristics�������������������������������������������������������������������������������������������������������������������������������� 155 Table 3. Measurements of the complete upper grinding stones from WFD4���������������������������������������������������������������������������������������� 155 Table 4. Hammerstone subtypes from WFD4��������������������������������������������������������������������������������������������������������������������������������������������� 156

D. Eitam: Cereal processing in stone agri-technological system at late Natufian Huzuq Musa

Figure 1. Huzuq Musa, view to the south���������������������������������������������������������������������������������������������������������������������������������������������������� 162 Figure 2. Map of Natufian sites with rock-cut installations in the Southern Levant, including Hruq Musa����������������������������������� 163 Figure 3. Map of Huzuq Musa with surface architectural remains and rock-cut installations���������������������������������������������������������� 164 Figure 4. Plan and sections of the northern dwelling area with huts (A-I, O)��������������������������������������������������������������������������������������� 165 Figure 5. Plan and sections of central dwelling huts area IV (N, P-T), part of the large structure (K, L) and terrace wall������������ 166 Figure 6. Legend of ground stones and rock-cut installations with fabrication marks, usewear and striations���������������������������� 169 Figure 7. Boulder pierced-bottom narrow conical mortar����������������������������������������������������������������������������������������������������������������������� 170 Figure 8. Huzuq Musa, selection of flint tools collected on surface������������������������������������������������������������������������������������������������������� 172 Figure 9. Rock-cut installations of Huzuq Musa����������������������������������������������������������������������������������������������������������������������������������������� 173 Figure 10. Threshing floors at complex VI�������������������������������������������������������������������������������������������������������������������������������������������������� 174 Figure 11. Ground stone tools found on the surface of the site�������������������������������������������������������������������������������������������������������������� 175 Table 1. Types of rock-cut installations in Huzuq Musa��������������������������������������������������������������������������������������������������������������������������� 168

D. Eitam: Cuboid-Spheroid Stone Object – an Archaic Scale Weight

Figure 1. Five Cuboid-Spheroid Objects������������������������������������������������������������������������������������������������������������������������������������������������������� 180 Figure 2. Cuboid-Spheroid Objects�������������������������������������������������������������������������������������������������������������������������������������������������������������������������������������� 181 Table 1. Cubic-Spheroid Stone Objects�������������������������������������������������������������������������������������������������������������������������������������������������������� 182

A. Greener and E. Ben-Yosef: Groundstone Tools from Site 35 – an Early Iron Age Copper Smelting Site

Figure 1. Map of major copper production sites in the Southern Levant���������������������������������������������������������������������������������������������� 190 Figure 2. Archaeological sites in Timna Valley and the location of Site 35������������������������������������������������������������������������������������������� 191 Figure 3. Site 35 and surrounding areas (facing SE)���������������������������������������������������������������������������������������������������������������������������������� 192 Figure 4. Slag Scatter at Site 35��������������������������������������������������������������������������������������������������������������������������������������������������������������������� 192 Figure 5. Site 35 excavation areas, major architectural elements and slag scatter������������������������������������������������������������������������������ 193 Figure 6. Aerial view of Area A���������������������������������������������������������������������������������������������������������������������������������������������������������������������� 194 Figure 7. Area B4, storage pit������������������������������������������������������������������������������������������������������������������������������������������������������������������������ 194 Figure 8. Timna Valley Geological Map with the location of the sites studied as part of the current research������������������������������ 195 Figure 9. Area A grinding stones, anvils and abraders������������������������������������������������������������������������������������������������������������������������������ 196 Figure 10. Area A pounders��������������������������������������������������������������������������������������������������������������������������������������������������������������������������� 197 Figure 11. Areas B and C pounders��������������������������������������������������������������������������������������������������������������������������������������������������������������� 198 Figure 12 & 13. Ground stone and rock type distributions at the different excavation areas of Site 35������������������������������������������ 199 Figure 14. Area B4 pit during the excavations�������������������������������������������������������������������������������������������������������������������������������������������� 199 Figure 15. Area B4 (pit) anvils and grinding stones����������������������������������������������������������������������������������������������������������������������������������� 200 Figure 16. Area B4 (pit) pounders����������������������������������������������������������������������������������������������������������������������������������������������������������������� 201 Figure 17. The occurrence of ‘ad hoc’ tools in Areas A and B4����������������������������������������������������������������������������������������������������������������� 201 Figure 18. Ground stone tools collected during the survey��������������������������������������������������������������������������������������������������������������������� 202 Figure 19. Pounders collected during the survey�������������������������������������������������������������������������������������������������������������������������������������� 203 Figure 20. Ground stone and rock type distributions of the Site 35 survey collection������������������������������������������������������������������������ 203 Figure 21. Modern experiments using grinding stones used to grind the local copper ore��������������������������������������������������������������� 204 Figure 22. Grinding slab with traces of malachite from the Chalcolithic site of Tall Hujayrat al-Ghuzlan near Aqaba����������������� 205 Figure 23. Crushed slag fragments from Timna����������������������������������������������������������������������������������������������������������������������������������������� 205 Figure 24. The small dimpled pounders (situated by us) on top of an anvil with multiple cup marks�������������������������������������������� 205 Figure 25. Ground stone and rock type distributions at the different areas of Site 34������������������������������������������������������������������������ 206 Figure 26. The occurrence of ‘ad hoc’ tools at Site 34�������������������������������������������������������������������������������������������������������������������������������� 206 Figure 27. Ground stone and rock type distributions at the Site 34 survey������������������������������������������������������������������������������������������ 206

A. Squitieri: The Iron Age stone tool assemblage of Gird-i Bazar

Figure 1. The location of Gitrd-i Bzaar in the Peshdar Plain. Inset: the location of the Peshdar Plain in Iraq�������������������������������� 211 Figure 2. Drone image showing the excavations conducted by the Peshdar Plain Project in the Bora Plain���������������������������������� 212 Figure 3. Drone image by ICONEM��������������������������������������������������������������������������������������������������������������������������������������������������������������� 213

v

Figure 4. A. The western part of the excavated area at Gird-i Bazar. B: Room 46 where the pivoted stone for the potter’s slow-wheel was found; C: Courtyard 18 where a large amount of smashed pottery vessels and stone tools was found.������� 214 Figure 5. Pebble mortars from Gird-i Bazar������������������������������������������������������������������������������������������������������������������������������������������������ 215 Figure 6. Pounder from Gird-i Bazar������������������������������������������������������������������������������������������������������������������������������������������������������������ 216 Figure 7. Polisher from Gird-i Bazar������������������������������������������������������������������������������������������������������������������������������������������������������������� 216 Figure 8. A pounder/polisher from Gird-i Bazar���������������������������������������������������������������������������������������������������������������������������������������� 216 Figure 9. A weight from Gird-i Bazar����������������������������������������������������������������������������������������������������������������������������������������������������������� 217 Figure 10. A perforated stone from Gird-i Bazar���������������������������������������������������������������������������������������������������������������������������������������� 217 Figure 11. The pivoted stone found at Gird-i Bazar used in combination with a socketed stone for a potter’s slow-wheel�������� 218 Figure 12. Distribution map of the small finds from western part of Gird-i Bazar������������������������������������������������������������������������������� 219

S. Prell: Macrolithics and the on-going use of stone tools in Qantir-Piramesse and Tell el-Dabʿa-Avaris

Figure 1. Location of Qantir-Piramesse and Tell el-Dabʿa – Avaris in the Eastern Delta��������������������������������������������������������������������� 224 Figure 2. Location of site Q I and Q IV south of the modern village of Qantir�������������������������������������������������������������������������������������� 225 Figure 3. Overall distribution of stone tools in site Q I����������������������������������������������������������������������������������������������������������������������������� 226 Figure 4. Examples for hammerstones and pounders from site Q I�������������������������������������������������������������������������������������������������������� 226 Figure 5. Examples for abrasive stones from site Q I��������������������������������������������������������������������������������������������������������������������������������� 227 Figure 6. Examples for polishing stones from site Q I������������������������������������������������������������������������������������������������������������������������������� 227 Figure 7. Examples for different shapes of querns from site Q I together with a limestone stool discovered in the fields surrounding Qantir���������������������������������������������������������������������������������������������������������������������������������������������������������������������������������� 228 Figure 8. Tools of comparable shape are used for: a) Embossing metal vessels; b) Smoothing wood; c) Embossing metal sheets���������������������������������������������������������������������������������������������������������������������������������������������������������������������� 228 Figure 9. Polishing tool for bone pins/arrowheads made from phyllite����������������������������������������������������������������������������������������������� 229 Figure 10. Polishing tool for bone pins/arrowheads made from steatite���������������������������������������������������������������������������������������������� 229 Figure 11. Pressure stone for wooden bow drill����������������������������������������������������������������������������������������������������������������������������������������� 229 Figure 12. Location of site R/III, R/IV and ʿEzbet Helmi in the fields surrounding Tell el-Dabʿa and ʿEzbet Rushdi��������������������� 229 Figure 13. Overall distribution of stone tools in site R/III������������������������������������������������������������������������������������������������������������������������ 230 Figure 14. Quern from site R/III with concave grinding surface and diagonal grooves���������������������������������������������������������������������� 230 Figure 15. Examples for grinding equipment from site R/III������������������������������������������������������������������������������������������������������������������ 231 Figure 16. Combined hammerstone from site Q I with abrading surfaces of differing graining������������������������������������������������������� 232 Figure 17. Combined hammerstone made from gneiss deriving from earlier layers in site R/III����������������������������������������������������� 232

R. Frankel: Millstones, Mortars, and Stone Bowls from Tel Dover and the Southern Levant

Figure 1. Saddle Querns Nos. 1-8������������������������������������������������������������������������������������������������������������������������������������������������������������������ 236 Figure 2. Mortars and Bowls Nos. 10-19������������������������������������������������������������������������������������������������������������������������������������������������������ 238 Figure 3. Mortars and Bowls Nos. 20-24������������������������������������������������������������������������������������������������������������������������������������������������������ 238 Figure 4. Mortars and Bowls Nos. 25-28������������������������������������������������������������������������������������������������������������������������������������������������������ 239 Figure 5. Footed Stone Bowls Nos. 30-32����������������������������������������������������������������������������������������������������������������������������������������������������� 239 Figure 6. Footed Stone Bowls Nos. 33-35����������������������������������������������������������������������������������������������������������������������������������������������������� 240 Figure 7. Olynthus Mulls Nos. 36-37������������������������������������������������������������������������������������������������������������������������������������������������������������� 241 Figure 8. Olynthus Mills Nos. 39-43�������������������������������������������������������������������������������������������������������������������������������������������������������������� 242 Figure 9. Rotary Hand Mills Nos. 45-47�������������������������������������������������������������������������������������������������������������������������������������������������������� 248 Figure 10. Rotary Hand Mills Nos. 48-52������������������������������������������������������������������������������������������������������������������������������������������������������ 249 Figure 11. Rotary Hand Mills Nos. 53-58������������������������������������������������������������������������������������������������������������������������������������������������������ 250 Figure 12. Rotary Hand Mills Nos. 59-63������������������������������������������������������������������������������������������������������������������������������������������������������ 251 Figure 13. Pompeian Donkey Mills Lower Stones Nos. 64-66������������������������������������������������������������������������������������������������������������������� 255 Figure 14. Pompeian Donkey Mills Lower Stones Nos. 67, 68������������������������������������������������������������������������������������������������������������������� 255 Figure 15. Pompeian Donkey Mills Lower Stones Nos. 70-72������������������������������������������������������������������������������������������������������������������� 256 Figure 16. Pompeian Donkey Mills Lower Stones Nos. 73-76������������������������������������������������������������������������������������������������������������������� 257 Figure 17. Pompeian Donkey Mills Upper Stones Nos. 81-83������������������������������������������������������������������������������������������������������������������� 258 Figure 18. Pompeian Donkey Mills Upper Stones Nos. 84-87������������������������������������������������������������������������������������������������������������������� 259 Figure 19. Pompeian Donkey Mills Lower Stones Nos. 88-90������������������������������������������������������������������������������������������������������������������� 260 Figure 20. Pompeian Donkey Mill Upper Stones Nos. 91-92�������������������������������������������������������������������������������������������������������������������� 260 Table 1. Stone tools from Tel Dover������������������������������������������������������������������������������������������������������������������������������������������������������������� 264 Appendix A������������������������������������������������������������������������������������������������������������������������������������������������������������������������������������������������������� 272

D. Eitam: Stone Tools of the Iron Age Ein Gev and their Implication

Plate 1. 1, 2, 4: fine bowls (basalt); 3: fragment of horizontal and diagonal bars bowl (basalt); 5, 7: massive bowls (basalt); 6: Platter; 7: cupmarks (limestone)���������������������������������������������������������������������������������������������������������� 282 Plate 2. 1-3: cupmarks; 4: Nuddle; 5: Ovoid; 6: Abrader & anvil; 7: Polisher; 8, 9: Rubbing stone; 10: Possibly scale weight; 11-13: Ovoid and spheroid pounders���������������������������������������������������������������������������������������������������������������������������������������������������� 283 Plate 3. 1: Rectangular grinding slab; 2: Grinding slab, slab; 3: Loaf grinding slab, symmetric; 4: Oval handstone, symmetric; 5: Handstone, reused fragment of symmetric loaf handstone; 6: plausibly basalt��������������������������������������������������� 284 Plate 4. 1: Loaf handstone, symmetric, trapezoid? 2: Oval handstone, symmetric; 3: Small oval handstone or palette (bifacial); 4: Rectangular, trapezoid hanstone, high & heavy; 5: Small rectangular handstone, trapezoid; 6: Handstone, reuse fragment of loaf handstone, trapezoid������������������������������������������������������������������������������������������������������������� 285

vi

Appendix A. Typology list of Stone Tools of IA Ein Gev���������������������������������������������������������������������������������������������������������������������������� 289 Appendix B. Catalog of Stone Tools of IA Ein Gev�������������������������������������������������������������������������������������������������������������������������������������� 293

J. S. Schneider et al.: Selenite (gypsum) from the North Sinai collection

Figure 1. Selenite fragment with straight sawed edge������������������������������������������������������������������������������������������������������������������������������ 299 Figure 2. Fragment of a selenite panel; this was the approximate shape and size made at the workshop�������������������������������������� 299 Figure 3. Waste fragment of selenite showing the precipitate cortex��������������������������������������������������������������������������������������������������� 299 Figure 4. Magnified view of saw marks on the cut edge of an archaeological selenite fragment������������������������������������������������������ 302 Figure 5. Resulting cut marks from experimental attempts to cut selenite with string and wire���������������������������������������������������� 302 Figure 6a. Saw used to replicate selenite panel production��������������������������������������������������������������������������������������������������������������������� 302 Figure 6b. Piece of replicated selenite panel production������������������������������������������������������������������������������������������������������������������������� 302 Figure 7. Experimentally split panels of selenite��������������������������������������������������������������������������������������������������������������������������������������� 302 Table 1. Selenite (Gypsum) Fragments from the North Sinai Survey Collection���������������������������������������������������������������������������������� 300

I. Milevski: The stone tools and vessels from Tel Miqne-Ekron

Figure 1. Location map with Miqne and main sites mentioned in the text������������������������������������������������������������������������������������������� 306 Figure 2. Tel Miqne – Ekron, top plan with fields of excavation������������������������������������������������������������������������������������������������������������� 307 Figure 3. Stone tools and vessels from the Chalcolithic and Bronze Ages��������������������������������������������������������������������������������������������� 312 Figure 4. Stone tools from the Iron I, Strata VII-IV. Lower and Upper Grinding Stones and a Basin������������������������������������������������ 315 Figure 5. Stone tools from the Iron I, Strata VII-IV. Mortars, Pestle, Hammer Stone and Rubbing Stone��������������������������������������� 317 Figure 6. Stone tools from the Iron I, Strata VII-IV. Rings, Pounders, Multiuse and Recycled Stones���������������������������������������������� 319 Figure 7. Stone tools from the Iron I, Strata VII-IV. Vessels��������������������������������������������������������������������������������������������������������������������� 320 Figure 8. Stone tools from the Iron II, Stratum IB. Lower Grinding Stone�������������������������������������������������������������������������������������������� 322 Figure 9. Stone tools from the Iron II, Stratum IB. Lower Grinding Stones������������������������������������������������������������������������������������������� 323 Figure 10. Stone tools from the Iron II, Stratum IB. Lower and Upper Grinding Stones��������������������������������������������������������������������� 324 Figure 11. Stone tools from the Iron II Stratum IB. Mortars�������������������������������������������������������������������������������������������������������������������� 326 Figure 12. Stone tools from the Iron II, Strata IB-C IA. Pestles, Rubbing and Hammer Stones, a Pounder and Pierced Stones������ 327 Figure 13. Stone tools from the Iron II, Strata IB-C. Potter’s Wheels and Rollers�������������������������������������������������������������������������������� 330 Figure 14. Stone objects from the Iron II, Strata III, IB-C. Pommels, Palette and Lid��������������������������������������������������������������������������� 331 Figure 15. Stone vessels from the Iron II, Strata IB-C. Bowls������������������������������������������������������������������������������������������������������������������� 333 Figure 16. Stone vessels from the Iron II, Stratum IB. Basins������������������������������������������������������������������������������������������������������������������ 334 Figure 17. Ground stone types frequency by period at Tel Miqne – Ekron from the LB II and Iron I����������������������������������������������� 339 Figure 18. Ground stone types frequency by period at Tel Miqne – Ekron from the Iron II��������������������������������������������������������������� 339 Figure 19. Ground stone raw materials frequency by period at Tel Miqne – Ekron from the LB II and Iron I��������������������������������� 340 Figure 20. Ground stone raw materials frequency by period at Tel Miqne – Ekron from the Iron II������������������������������������������������ 341 Table 1. Stratigraphy of Tel Miqne, Middle Bronze to Iron Age�������������������������������������������������������������������������������������������������������������� 307 Table 2. Distribution of stone tools and vessels from fields I, II, III, IV, V, VII, IX and X according to strata����������������������������������� 308 Table 3. Fields I NE, I NW, I SE and I SW������������������������������������������������������������������������������������������������������������������������������������������������������� 336 Table 4. Field II SW������������������������������������������������������������������������������������������������������������������������������������������������������������������������������������������ 336 Table 5. Fields III NE, III SE, and IV SE���������������������������������������������������������������������������������������������������������������������������������������������������������� 337 Table 6. Field IV NE upper, IV NW upper and V SW���������������������������������������������������������������������������������������������������������������������������������� 337 Table 7. Fields IV NE lower and IV NW lower��������������������������������������������������������������������������������������������������������������������������������������������� 338 Table 8. Field X NW����������������������������������������������������������������������������������������������������������������������������������������������������������������������������������������� 338

E. Adama et al.: El-Khirba: Food processing and other ground stone tools

Figure 1. Lower grinding stone from stratum VII�������������������������������������������������������������������������������������������������������������������������������������� 347 Figure 2. Upper grinding stones from stratum VII������������������������������������������������������������������������������������������������������������������������������������ 347 Figure 3. Lower Olynthian mills from stratum VII������������������������������������������������������������������������������������������������������������������������������������ 348 Figure 4. Upper Olynthian mills from stratum VII������������������������������������������������������������������������������������������������������������������������������������ 348 Figure 5. Vessels from stratum VII��������������������������������������������������������������������������������������������������������������������������������������������������������������� 350 Figure 6. Mortars from stratum VII�������������������������������������������������������������������������������������������������������������������������������������������������������������� 351 Figure 7. Pestle from stratum VII����������������������������������������������������������������������������������������������������������������������������������������������������������������� 351 Figure 8. Vessels from stratum VI���������������������������������������������������������������������������������������������������������������������������������������������������������������� 352 Figure 9. Lower rotary hand mill from stratum IV������������������������������������������������������������������������������������������������������������������������������������ 354 Figure 10. Striation marks on lower rotary hand mill (EK38)������������������������������������������������������������������������������������������������������������������ 354 Figure 11. Upper rotary hand mills from stratum IV�������������������������������������������������������������������������������������������������������������������������������� 355 Figure 12. Vessels from stratum IV�������������������������������������������������������������������������������������������������������������������������������������������������������������� 356 Figure 13. Varia object from stratum IV������������������������������������������������������������������������������������������������������������������������������������������������������ 356 Table 1. Distribution of tools types and periods in the el-Khirba ground stone tools assemblage��������������������������������������������������� 346 Table 2. Distribution of tool types and raw materials periods in the el-Khirba ground stone tools assemblage���������������������������� 346 Table 3. Distribution of tool types and raw materials in the Stratum VII (Roman period) assemblage������������������������������������������� 346 Table 4. Distribution of tool types and raw materials in the Stratum VI (Byzantine period) assemblage��������������������������������������� 352 Table 5. Distribution of tool types and raw materials in the Stratum IV (Abbasid Period) assemblage������������������������������������������� 353 Table 6. Distribution of tool types and raw materials in the Stratum III (Mamluk period) assemblage������������������������������������������� 357 Table 7. Distribution of tool types and raw materials in the indistinct contexts assemblage������������������������������������������������������������ 357

vii

List of Authors Yael Abadi-Reiss Israel Antiquity Authority [email protected]

Aaron Greener The Laboratory for Ground Stone Tool Research, Zinman Institute of Archaeology, University of Haifa; The W.F. Albright Insitute of Archaeological Research [email protected]

Uzi ‘Ad Israel Antiquities Authority [email protected]

Haskel J. Greenfield University of Manitoba, Department of Anthropology and St. Paul’s College, Winnipeg MB; R3T 2N2, Canada [email protected]

Erez Adama Laboratory for Ground Stone Tools Research, Zinman Institute of Archaeology, University of Haifa [email protected]

Clara Jeuthe German Archaeological Institute, Cairo [email protected]

Jeremy A. Beller Department of Anthropology, University of Victoria; Victoria BC, V8P 5C2, Canada [email protected]

Lucie Jirásková Czech Institute of Egyptology, Prague [email protected]

Erez Ben-Yosef The Jacob M. Alkow Department of Archaeology and Ancient Near Eastern Cultures, Tel Aviv University [email protected]

Thomas E. Levy Department of Anthropology, UC San Diego [email protected] Tamara Lewit The University of Melbourne [email protected]

Luca Bombardieri Department of Humanities, University of Torino, Italy [email protected] Paul Burton The Australian National University [email protected]

Aren M. Maeir Bar-Ilan University, Institute of Archaeology, the Martin (Szusz) Department of Land of Israel Studies and Archaeology, Ramat Gan 52900, Israel [email protected]

Jennie Ebeling University of Evansville [email protected]

Ianir Milevski Israel Antiquities Authority [email protected]

David Eitam Independent researcher, Hararit, 20182 Israel [email protected]

David Mudd Department of Archaeology, University of Reading [email protected]

Mostafa Fayek University of Manitoba, Department of Geological Sciences, Winnipeg MB; R3T 2N2, Canada [email protected]

Mohammad Najjar Department of Antiquities of Jordan [email protected] Eliezer D. Oren Ben Gurion University of the Negev, Professor Emeritus

Rafael Frankel University of Haifa [email protected]

Silvia Prell Austrian Academy of Sciences [email protected]

Avraham Gabay Karev, Israel viii

Danny Rosenberg Laboratory for Ground Stone Tools Research, Zinman Institute of Archaeology, University of Haifa, Israel [email protected]

Andrea Squitieri Ludwig-Maximilian University of Munich (LMU) [email protected] David Valentine Idaho Power Company, Boise, ID, USA

Joan S. Schneider California State Parks, Colorado Desert District (retired) [email protected] Itzhaq Shai The Department of Land of Israel Studies and Archaeology, Ariel University, P.O.B. 3, Ariel 40700 Israel [email protected]

ix

x

Introduction David Eitam and Andrea Squitieri

1. Aims and scope of the book

Thanks to recent and more accurate methodologies, stone tools have also proved to be in fact in constant change both diachronically and cross-culturally, and to be a valid source of information to better understand the changing political and socio-economic landscapes of the Near East and Egypt, especially in critical moments during the prehistory and history of these regions, such as the transition from hunter-gatherer to agricultural societies, the first urbanisation process at the beginning of the Bronze Age, the collapse of the Late Bronze Age civilisations, the takeover of large multi-national empires in the course of the Iron Age and the Persian period, the transition from the classical era to the Byzantine periods, and from the early Arab period until the beginning of the Ottoman empire.

This book deals with groundstone tools, rock-cut installations and stone vessels in the Ancient Near East and Egypt from the Prehistory to the Late Antiquity, with the aim of exploring various aspects of these objects, from raw material procurement to uses and their socio-economic and cultural meanings. In the first decades of the archaeological research in the Middle East, these categories of objects have often been overlooked (with some exception, e.g., Macalister 1912; Petrie 1917, 1937). They were rarely published in the excavation reports, or, when published, the information provided about their morphology, raw material, tool marks and contexts was in most cases incomplete or even missing. The main reason for neglecting this material was that it was considered self-explanatory with trivial chronological meaning, providing the archaeologists with little information about historical and cultural aspects, two of the aspects which were the main focuses of the archaeological research in that time (see Albright 1938: 84 vs Childe 1943: 19).

Naturally, much remains to do in the field of stone tool research. Some key issues are still seldom tackled also in the most recent studies of stone tools. Likewise, the methodology to effectively study at least some stone tool categories (i.e., rock-cut installations) is still not fully established. The aim of this book is to shed light on these issues and pave the way to a fruitful debate on them. Six key issues were selected to deal with in the current volume:

As the archaeological research extended beyond historical and cultural aspects towards broader issues on anthropological and socio-economic aspects of past societies, a new era for stone tool research has opened up since the 90s, especially after the pioneering works by K. Wright, R. Frankel, D. Eitam and others (1992; 1999, 1979; 1996 respectively).

1. The essence of any stone tool is determined by its usage. Consequently, a systematic study focused on the tools’ functional aspect is essential. Generally, this issue is answered by random suggestions, like ritual uses of stone tools (Dietrich et al. 2012: 687689; Hayden et al. 2013; Nadel and Lengyel 2009), or tackled solely through experimental studies of usewear (Dubreuil 2004); or through residue analysis applied on tool’s working surface(s); but generally, not through a technical analysis. Since stone devices are in fact ‘machines’ that operated according to physical laws studied by material science and engineering, investigating the uses of stone tools requires some degree of technical knowledge. Hence, the collaboration with industrial and mechanical engineers or physicists is especially necessary while conducting experimental studies (e.g., Eitam 1979; Eitam et al. 2015). For a better understanding of the many factors involved in the study of stone tools and for avoiding blunders (e.g., Grosman and Goren-Inbar 2007), a research project of stone tools may also include expert/s in other fields who can provide essential insights to

Their works and the many studies that have followed have shown that stone tools can be used effectively to tackle questions regarding the organisation of everyday activities such as food preparation and industrial activities (e.g., Ebeling and Rowan 2004; BenYosef 2012); patterns of access to raw materials (e.g., Rosenberg et al. 2014); regional and interregional trade in both raw materials and finished goods (e.g., Milevski 2005; Squitieri 2017); gender issues (e.g., Abadi Reiss et al. in this volume); wealth distribution (Wright 2014); discard behaviour (e.g., Mudd in this volume), and rituals of communal meals (e.g., Wright 2000). Being involved in most of the key everyday activities of past societies, stone tools can represent the embodiment of socio-economic and anthropological aspects that informed those activities and that can be decoded by means of effective methods of analysis. 1

Stone Tools in the Ancient Near East and Egypt fully understand the uses of stone tools, like geomorphologist and in other cases, agronomist, food specialist, or metallurgists.

only based on wheat and barley phytoliths found on a ‘grinding stone’ without examining the stone tool, compare Piperno et al. 2004 to Eitam 2009: Fig. 9 stone tool type I.E1 vs. type I.E2; see also Dubreuil and Nadel 2015; and see the limitations of usewear analysis, Dubreuil et al. 2015; 146-147).

2. A second key issue is the need for a classification system of stone tools. Wright’s classification system (1992) of ground stones from the Epipaleolithic to the Chalcolithic, including a precise terminology and typology classes, criteria and morphological terms, has been well accepted for 26 years by archaeologists studying prehistory, as well as by those investigating ancient history. Eitam has added to Wright’s ground stone classification system the installations cut in bedrock and chiselled in stone blocks, as many of the rock-cut installations are the same devices as the ground stones (such as cupmarks, grinding devices, mortars, conical mortars and basins, see Eitam 2009). Eitam also added some stone objects forming separated typological systems for different cultures (e.g., Natufian and PPNA, and Iron Age stone tools, Eitam 2013, and see Eitam in this volume). Sparks (2007) and Squitieri (2017) established the stone vessel typology for the Bronze and Iron Ages respectively; while Frankel has built up a classification system of improved oil and wine rock-cut installations (1999) and mills for producing flour (see Frankel in this volume).

In our opinion, stone tools, as other ancient findings (such as lithics, pottery sherds, metal or even artistic objects), should all be studied in a classification system as this can help comparative studies both within and across regions as well as periods. Such a classification system should be defined according to morphological (that is raw material, shape and manufacturing marks) as well as technological features (such as usewear, although sometimes it is hard or even impossible to distinguish between usewear and manufacturing marks). To this aim, employing a consistent terminology and typology, which avoids some long-established errors and confusions, is essential for advancing in the field of stone tool studies. 3. The third key issue concerns the geographical scope of stone tool research. Stone tools do generally show strong similarities across faraway areas such as the Near East, Egypt, Europe, and the Americas, but comparative studies among faraway regions are often problematic. Two cases may reinforce our point: A. Grinding of maize in Central America is different from grinding wheats in the Near East and Egypt because of many factors involved, such as the raw material of the device, its shape and the ways of use (compare adding water while grinding corn, see Hayden 1987, vs dry grinding of wheats). Therefore, the suggestion that loaf-shaped handstones in Iron Age Israel were used for grinding on both the front and the bottom surfaces, and sometimes also on left and right edges (Liebowitz 2008) should be taken with caution. While this assumption finds some support in how maize tortillas in Central America are prepared (Liebowitz 2008), local ethnographical parallels of cereal processing are preferable, given the conservative nature of food preparation methods. B. It has been suggested that acorns were part of the Natufian diet (Nishiaki 1998; Olszewski 2004), while Rosenberg also proposed that the pounding and processing of acorns were done in the numerous Natufian bedrock mortars (2008). Acorns consumption is a well recorded in archaeological findings and in historical documents in both Europe and the Americas. A rich variety of oak species have been growing in the wild for tens of thousands of years in the New World, Europe and Turkey (253 species in Mexico, 84 in North America, and 32 in Europe), in contrast to only five species of oak grow in the Near East and Western Asia. While examining this suggestion, one may stress that the

Some researchers have recently challenged the need for a stone tool typology, suggesting drastic alterations (e.g., by determining the types of Natufian rock-cut installations based on stone assemblages of two sites; e.g. Nadel et al. 2009). It seems that, for some scholars, typology is not a strong means for the investigation of stone tools (e.g., Dubreuil 2002), and in particular rock-cut installations (e.g., Nadel and Rosenberg 2010). Other researchers establishing ground stones typology solely on morphology avoiding usewear (Rosenberg 2011). This claim, however, seems to be contradicted by the fact that certain morphological features of stone tools are in fact the result of their specific intensive use (for example, the narrow cylindrical shaft at the end of some conical mortars; the deep concave shape of grinding stones called ‘saddle querns’; and the pierced bottom of some cupmarks, mortars and querns). Therefore, specific morphological patterns constituting a typological system can be found among stone tools and be connected to their functions. Also, the trend of defining the usage of a stone tool exclusively by means of residue analysis, without the background of comparative morphological studies, does not always achieve conclusive results (see e.g., the claim that bread was prepared 32,000 years ago 2

D. Eitam and A. Squitieri: Introduction Southern Levant oak acorns are woody and hardly nourishing, unlike in other parts of the world where oak species produce nutritious acorns that are widely eaten (e.g., Mason and Nesbitt 2012). The nutritional values of local acorns vary among species (265/172 kcal in 100 g of Common Oak acorns) and they are rarely found in large amounts in ancient south Mediterranean sites (e.g., Meson 1996). Only in Early Epipalaeolithic Ohalo II, thousands of acorns were found (Weiss 2002), which they were possibly used as fuel and in tanning animal hides (the Tabor oak has an especially high ratio of nonedible acid, Avitsur 1975, 1976; Araf 1975). Second, acorns consumption in pre-Israel Palestine was limited to famines and other harsh times (Avitsur 1976), unlike in European North America where the practice of eating acorns, also said balanophagy, is well documented. Third, by far most of the Natufian mortars have pointed inner bottom and therefore they do not seem to be suitable for pounding acorns as the mash would have been difficult to extract from the narrow bottom. It seems, therefore, that in the Late Epipalaeolithic Levant, acorns were probably only marginally consumed. Hence, despite being in many cases very insightful, cross-cultural comparisons should always be used with cautions.

devices, for example, (parallel to the lower grinding slabs), are first to be seen in the Late Natufian, and occasionally re-appeared across later periods until the Roman period. Cupmarks are another example, which first appeared in the Late Palaeolithic, and continued to exist consecutively until the Early Arab period. Both devices emerged in the Southern Levant in a large quantity in a particular period, becoming there a landmark of that culture: grinding devices in the late Chalcolithic-Early Bronze I (Eitam 2008; Van Den Brink 2008), and cupmarks, in various forms, in the PPNA (Noy 1979; Rosenberg and Gopher 2010). Their appearing in large quantities bear great significant: the clusters of numerous grinding devices is evidence for joint or communal making of bread, while the standard simple food processor, that is the cupmark with pestle, points towards a returning to cereal groats and porridge meals vs the previous Natufian bread making. Furthermore, while carefully examining the same type of stone tool along the periods, one would notice variations in measures, ratio and style. The development of an archetypal device through a long timespan in a specific geographical area is another characteristic of stone tools. A good example of this is a special olive oil device typical of the Central Hills of Israel, probably invented in the Chalcolithic period. Initially, this was a small deep, round basin; subsequently, in the Iron Age I a wide shallow basin was added, surrounding a similar small deep basin, thus enabling crashing the olives in more efficient way prior to the extraction of the oil into the deep basin (Eitam 2003). The common Iron Age II lever and weights oil-press with a central collecting vat was probably a development of the prototype devices. Later on, the improved large presses with a central collecting vat became in the Roman time exclusive to Judea and were transferred to the Golan and the Galilee (as a lever and screw and direct screw presses) during the deportee of the Jewish population from Judea (Frankel 1999).

4. Another issue concerning stone tool research is the dearth of detailed published data, especially concerning some regions within the Near East and Egypt, and periods. The publications of some of the major sites offer reports about ground stone tools in the form of a catalogue accompanied by illustrations of only selected tools (e.g., Megiddo, Lachish and Hazor: Sass 2000, 2004; Sass and Cinamon 2006; Sass and Ussishkin 2004; Ebeling 2012; Rosenberg 2013, respectively, but see Yadin et al. 1958, 1960, 1961). Moreover, these reports are rarely followed by subsequent more in-depth studies on the stone tools (as, for example, in Ebeling and Rosenberg 2015). The present volume, therefore, hopes to stimulate both the publication of more detailed reports about stone tools from sites, and more of in-depth studies on stone tools by publishing stone tool assemblages from several sites.

6. Finally, and most importantly, there is the socioeconomic and anthropological implications of stone tools. Stone tools became key measures for processing plant staple food from the Late Epipalaeolithic onwards, and a major means of production of large scale manufacture from the Iron Age II period onward. Consequently, the study of stone tools can reveal or clarify the subsistence economy of early societies as well as socio-economic and cultural aspects of historical communities. Two examples may clarify this point. A survey of about 1000 Natufian rock-cut installations revealed an agro-technological system for processing wild

5. Typological dating of ground stones and of rockcut installations is still not fully accepted by many archaeologists, as it is for pottery sherds and lithic assemblages. Nonetheless, during the more recent research projects of stone tools some types have revealed to be exclusively related to one period or culture (such as the 8th-9th centuries BCE oil press, Eitam 1979; the Late Natufian narrow conical mortar, Eitam 2009; and the Early Bronze four-handled basalt vessel, Rosenberg and Chasan 2017). On the other hand, many stone tools existed through thousands of years. The rock-cut grinding 3

Stone Tools in the Ancient Near East and Egypt cereals and producing cereal-food, including barley bread, which implies that the Natufian became a food producing society (Eitam et al. 2015; Eitam in this volume). Another example is the study of the Iron Age rock-cut installations, which revealed to be numerous same type of improved oil-presses (previously defined by Albright as dye vats, Albirght 1938). This technological improvement enabled a large-scale production of olive oil in the 9th-8th centuries BCE by farmers, also organised, possibly as royal large-scale production, concentrated in small fortified industrial villages (Eitam 1987, 1997), and in the 7th century BCE the establishing of a vast enterprise of oil production, possibly initiated by the Assyrians, in the Kingdom of Ekron (Eitam 1996).

Childe, V. G. 1943. Rotary Querns on the Continent and in the Mediterranean Basin. Antiquity 17: 1926. Dietrich, O., Heun, M., Notroff, J. and Schmidt, K. 2012. ‘The Role of Cult and Feasting in the Emergence of Neolithic Communities. New Evidence from Göbekli Tepe, South-Eastern Turkey’. Antiquity 86, 333: 674695. Dubreil, L., Savage, D., Delgado-Raack, S., Lesson, H., Stephenson, B. and de la Torre, I. 2015. ‘Current Analytical Frameworks for Studies of Use-wear on Ground Stone Tools’ in J. M. Marreiros, J. F. Gibaja Bao and N. F. Bicho (eds) Use-wear and Residue Analysis in Archaeology. Cham; New York. Springer: 105-165. Dubreuil, L. 2002. Etude fonctionnelle des outils de broyage natoufiens: nouvelles perspectives sur l’émergence de l’agriculture au Proche-Orient. PhD dissertation. University of Bordeaux. Talence. Dubreuil, L. 2004. ‘Long-Term Trends in Natufian Subsistence: A Use-Wear Analysis of Ground Stone Tools’. Journal of Archaeological Science 31: 1613-29. Ebeling, J. R. and Rosenberg, D. 2015. ‘A Basalt Vessel Workshop and its Products at Iron Age Hazor, Israel’. Journal of Field Archaeology 40: 665-674. Ebeling, J. R. and Rowan, Y. M. 2004. ‘The Archaeology of the Daily Grind: Ground Stone Tools and Food Production in the Southern Levant’. Near Eastern Archaeology 67 (2): 108-117. Ebeling, J. R. 2012. ‘Ground Stone Artifacts’ In A. BenTor, D. Ben-Ami, and D. Sandhaus (eds) Hazor VI: The 1990-2009 Excavations, The Iron Age. Jerusalem: Israel Exploration Society and the Institute of Archaeology, Hebrew University of Jerusalem: 542-58. Eitam, D. 1979. ‘Olive Presses of the Israelite Period’. Tel Aviv 6: 146-55. Eitam, D. 1987. ‘Olive Oil Production during the Biblical Period’ in M. Heltzer and D. Eitam (eds) Olive Oil in Antiquity, Israel and Neighboring Countries, from the Neolithic to the Early Arab Period, Conference. University of Haifa, Israel Oil Industry Museum, Dagon Museum: 16-36. Eitam, D. 1990. ‘Royal Industry in Ancient Israel during the Iron Age Period’ The Town as a Regional Economic Center in the Ancient Near East. Tenth International Economic History Congress, Leuven: 56-73. Eitam, D. 1996. ‘The Olive Oil Industry at Tell Miqne – Ekron in the Late Iron Age’ in M. Heltzer and D. Eitam (eds) Olive Oil in Antiquity, Israel and Neighboring Countries, from the Neolithic to the Early Arab Period: 167-196. Studies VII, Sargon srl: 166-196. Eitam, D. 1997. ‘Khirbet Khadash – An Industrial Village for the Production of Olive Oil in the Kingdom of Israel’, Il Dono e La Quiete il Mare Verde Dell’Olio, Homo Edens V: 56-73. Eitam, D. 2009. ‘Late Epipalaeolithic rock-cut installations and groundstone tools in the Southern Levant-methodology and classification system’. Paléorient 35: 77-104.

2. Structure of the book The book is divided into six sections. The first section ‘Methodology and classification’ tackles some broader issues concerning methodological aspects of the study of stone tools based on case studies; the following section ‘Documentation: non-archaeological and archaeological sources in comparison’ show some case studies in which a diverse range of sources, namely written, ethnographic and archaeological sources, is used to shed light on various aspects on stone tools’ functions and their anthropological meanings; the section ‘Raw material and manufacture’ focuses on the very first steps of the stone tools’ life-cycles that is the raw material procurement and the manufacture process; ‘Function and uses’ present several case studies showing the diverse uses of stone tools, from food production to industrial and economic activities; finally, the section ‘Sites and tools’ offers detailed reports of so-far unpublished material from some key sites of the Near East and Egypt. Bibliography Araf, S. 1975. ‘The use of forest trees in the Arab village’ in Ginsburg (ed.), The Tree and Man. G. Kefar Etzion Field School (in Hebrew). Albright, W. F. 1938. The Excavation of Tell Beit Mirsim, Vol. 2: The Bronze Age. Annual of the American Schools of Oriental Research. 17. New Haven, CT: American Schools of Oriental Research. Avitzur, S. 1975. ‘Uses of forest plants’ in G. Ginsburg (ed.), The Tree and Man. Kefar Etzion Field School (Hebrew): 72-75. Avitzur, S. 1976. Daily life in Eretz Israel in the XIX Century. Tel Aviv: Am Hasefer (Hebrew). Ben-Yosef, E. 2012. ‘Environmental Constraints on Ancient Copper Production in the Aravah Valley: Implications of the Newly Discovered Site of Khirbet Mana’iyah in Southern Jordan’. Tel Aviv 39: 58-74. 4

D. Eitam and A. Squitieri: Introduction Eitam, D., Kislev, M. E., Karty, A. and Bar-Yosef, O. 2015. ‘Experimental barley flour production in 12,500-year-old rock-cut mortars in Southwestern Asia’. PLoS ONE 10(7): e0133306. Frankel, R. 1999. Wine and Oil Production in Antiquity in Israel and other Mediterranean Countries. JSOT/ASOR Monograph Series 10, Sheffield: Sheffield Academic Press. Grosman, L. and N. Goren-Inbar 2007. ʻ“Taming” Rocks and Changing Landscapes. A New Interpretation of Neolithic Cupmarks’. Current Anthropology 48, 5. Hayden, B. 1987. ‘Traditional Metate Manufacturing in Guatemala Using Chipped Stone Tools’ in B. Hayden (ed.) Lithic Studies Among the Contemporary Highland Maya, Tucson: University of Arizona: 8-119. Hayden, B., Canuel, N. and Shanse, J. 2013. ‘What was brewing in the Natufian? an archaeological assessment of brewing technology in the Epipaleolithic’. Journal of Archaeological Method Theory 20: 102-150. Kislev M. E., Nadel, D. and Carmi, I. 1992. ‘Epipalaeolithic (19,000 BP) cereal and fruit diet at Ohalo II, Sea of Galilee, Israel’. Review of Palaeobotany and Palynology 73: 161-166. Liebowitz, H. A. 2008. ‘Wear Patterns on Ground Stone Implements from Tel Yin’am’ in Y. M. Rowan and J. R. Ebeling (eds) New Approaches to Old Stones: Recent Studies of Ground Stone Artifacts, London and Oakville: Equinox: 182-95. Macalister, R. A. S. 1912. Excavation of Gezer, 1902-1905 and 1907-1909, Vol. 1. London: Murray. Mason, S. 1996. ‘Acornutopia? Determining the role of acorns in past human subsistence’ in R. J. Wilkins, D. Harvey and C. M. Dobson (eds) Food in Antiquity Exeter: University of Exeter Press: 12-24. Meson, S. and M. K. Nesbitte 2012. ‘Acorn as food in southeast Turkey: implications for the prehistory subsistence of Southwest Asia’ in S. Andrew and E. Weiss. (eds) From Foragers to Farmers: Papers in Honor of Gordon C. Hillman. Fairbairn, Oxford: Oxbow Books: 71-85. Milevski, I. I. 2005. Local Exchange in Bronze Age Canaan. Unpublished Thesis. Tel Aviv University. Nadel, D. and Lengyel, G. 2009. ‘Human-made bedrock holes (mortars and cupmarks) as a Late Natufian social phenomenon. Archaeology’, Ethnology and Anthropology of Eurasia 37(2): 37-48. Nadel, D., Rosenberg, D. and Yeshurun, R. 2009. ‘The Deep and the Shallow: The Role of Natufian Bedrock Features at Rosh Zin, Central Negev, Israel’. Bulletin of the American Schools of Oriental Research 355: 129. Nadel, D. and D. Rosenberg 2010. ‘New insights into Late Natufian bedrock features (mortars and cupmarks)’. Eurasian Prehistory 7(1): 65-87. Nishiaki, O. Y. 1998. ‘The Paleolithic and Neolithic of Syria: an overview with reference to Jordanian prehistory’ in D. O. Henry (ed.) The Prehistoric

Archaeology of Jordan. BAR International Series 705. Oxford: Archaeopress: 195-207. Noy, T. 1979. ‘Stone Cup-Holes and Querns from Gilgal-I: A Pre-Pottery Neolithic-A Site in Israel’. Paléorient 5: 233-238. Olszewski, Deborah I. 2004. ‘Plant food subsistence issues and scientific inquiry in the Early Natufian’ in Delage, C. (ed.) The Last Hunter-Gatherer Societies in the Near East. BAR International Series 1320. Oxford: John and Erica Hedges: 189-209. Petrie, W. M. F. 1917. Tools and Weapons. London: British School of Archaeology in Egypt. Piperno, D. R., E. Wiess, I. Holst and D. Nadel 2004. ‘Processing of wild cereal grains in the Upper Palaeolithic Revealed by Starch Grain Analysis’. Nature 430: 670-673. Rosenberg, D. 2008. ‘The possible use of acorns in past economies of the Southern Levant: a staple food or a negligible food source?’ Levant 40(2): 167-175. Rosenberg, D. 2011. Development, Continuity and Change: The Stone Industries of the Early Ceramic Bearing Cultures of the Southern Levant. Ph.D. dissertation, University of Haifa (Hebrew). Rosenberg, D. 2013. ‘The Groundstone Assemblage’ in I. Finkelstein, D. Ussishkin, and E. H. Cline (eds) Megiddo V. The 2004-2008 Seasons, Vol. III. Monograph Series of the Sonia and Marco Nadler Institute of Archaeology 31. Tel Aviv: Emery and Claire Yass Publications in Archaeology, Institute of Archaeology, Tel Aviv University: 930-76. Rosenberg, D. and Gopher, A. 2010. ‘Food processing tools and other Groundstone Implements from Gilgal I and Gilgal III’ in O. Bar-Yosef, A. N. GoringMorris and A. Gopher (ed) Gilgal: Early Neolithic Occupations in the Lower Jordan Valley: The Excavations of Tamar Noy. American School of Prehistoric Research Monograph 4. Brill, Winona Lake: 139-176. Rosenberg, D. and Chasan, R. 2017. ‘The characteristics and significance of prestige goods during the Early Bronze Age period of the southern Levant: The particular case of the four-handled basalt vessels phenomenon’, Quaternary International, http:// dx.doi.org/10.1016/j.quaint. Rosenberg, D., Shimelmitz, R., Gluhak, T. M., Assaf, A. 2014. ‘The Geochemistry of Basalt Handaxes from the Lower Palaeolithic Site of Ma’Ayan Baruch, Israel—A Perspective on Raw Material Selection’. Archaeometry 57(S1): 1-19. Sass, B. 2000. ‘The Small Finds’ in I. Finkelstein, D. Ussishkin, and B. Halpern (eds) Megiddo III: The 1992-1996 Seasons, Vol. 2. Monograph Series of the Sonia and Marco Nadler Institute of Archaeology 18. Tel Aviv: Emery and Claire Yass Publications in Archaeology, Institute of Archaeology, Tel Aviv University: 349-428. Sass, B. 2004. ‘Pre-Bronze Age and Bronze Age Artefacts. Section A: Vessels, Tools, Personal Objects, Figurative Art and Varia’ in D. Ussishkin (ed.) The Renewed 5

Stone Tools in the Ancient Near East and Egypt Archaeological Excavations at Lachish (1973-1994), Vol. 3, Monograph Series of the Sonia and Marco Nadler Institute of Archaeology 22. Tel Aviv: Emery and Claire Yass Publications in Archaeology, Institute of Archaeology, Tel Aviv University: 1405-1524. Sass, B. and Ussishkin, D. 2004. ‘Iron Age and Post-Iron Age Artefacts. Section A: Vessels, Tools, Personal Objects, Figurative Art and Varia’ in D. Ussishkin (ed.) The Renewed Archaeological Excavations at Lachish (1973-1994), Vol. 4: Monograph Series of the Sonia and Marco Nadler Institute of Archaeology 22. Tel Aviv: Emery and Claire Yass Publications in Archaeology, Institute of Archaeology, Tel Aviv University: 19832057. Sass, B. and Cinamon, G. 2006. ‘The Small Finds’ in I. Finkelstein, D. Ussishkin, and B. Halpern (eds) Megiddo IV: The 1998-2002 Seasons, Vol. 1. Monograph Series of the Sonia and Marco Nadler Institute of Archaeology 24. Tel Aviv: Emery and Claire Yass Publications in Archaeology, Institute of Archaeology, Tel Aviv University: 353-436. Sparks, R. T. 2007. Stone Vessels in the Levant. Palestine Exploration Fund Annual II. Leeds: Maney. Squitieri, A. 2017. Stone Vessels in the Near East during the Iron Age and the Persian Period (c. 1200-330 BCE). Oxford: Archaeopress. Van Den Brink, E. C. M. 2008. ‘A New Fossile Directur of the Chalcolithic Landscape in the Shephelah and Samarian and Judean Countries: Stationary Grinding Facilities in Bedrock’. Israel Exploration Journal 58 1:1-23.

Weiss, E. 2002. Reconstruction the Human Economy and Society of the Epipalaeolithic Site Ohalo II from Macrofossil Botanical Remains. Unpublished PhD dissertation, Bar Ilan University, Ramat-Gan. Wiess, E., M. E. Kislev, O. Simchoni, D. ‘Nadel and H. Tschauner 2008. Plant-food preparation area on an Upper Paleolithic brush hut floor at Ohalo II, Israel’. Journal of Archaeological Science 35: 2400-2414. Wright, K. 1992. ‘A classification system for ground stone tools from prehistoric Levant’. Paléorient 18(2): 53-81. Wright, K. 2000. ‘The Social Origins of Cooking and Dining in Early Villages of Western Asia’. Proceedings of Prehistoric Society 66: 89-121. Wright, K. 2014. ‘Domestication and Inequality? Households, Corporate groups and Food Processing Tools at Neolithic Çatalhöyük’. Journal of Anthropological Archaeology 33: 1-33. Yadin, Y., Aharoni, A., Amiran, R., Dothan, T., Dunayevsky, I. and Perrot, J. 1958. Hazor I: An Account of the First Season of Excavations, 1955. Jerusalem: Magnes. Yadin, Y., Aharoni, Y., Amiran, R., Dothan, T., Dunayevsky, I. and Perrot, J. 1960. Hazor II: An Account of the Second Season of Excavations, 1956. Jerusalem: Magnes. Yadin, Y., Aharoni, Y., Amiran, R., Dothan, T., Dunayevsky, I. and Perrot, J. 1961. Hazor III‒IV: An Account of the Third and Fourth Seasons of Excavations, 1957-1958, Plates. Jerusalem: Magnes.

6

Methodology and Classification

7

8

The archaeology of discard and abandonment: presence and absence in the ground stone assemblage from Early Neolithic Bestansur, Iraqi Kurdistan David Mudd Keywords: ground stone, abandonment, Neolithic, Zagros, discarding behaviour

1. Abandonment, discard, and the archaeological record

that does survive. He argued that all communities continuously destroy and renew themselves, but that at some point, renewal stops and destruction continues. The site moves from an ‘inhabited phase’ to a ‘ghost phase’, and then to an archaeological phase when it is exposed and investigated by the archaeologist. Disorganisation (entropy) increases, and evidence can be lost or missed in the transition through these phases. He argued for more rigour in the use of ethnographic analogy to interpret archaeological data, particularly when it assumes a comparable mode of subsistence in the past and modern communities. Site formation theory was developed further by Schiffer (1972, 1983, 1987). He argued that loss of data was not the main problem, pointing out the importance of qualitative changes to the archaeological record due to variable discard practices in past cultural systems and post-depositional factors. He warned against the assumption by many archaeologists that the elements and spatial patterning of archaeological assemblages corresponded to their actual locations of use in past activities. He proposed a model of the processes by which durable artefacts moved through the ‘systemic context’ (Ascher’s ‘inhabited phase’) – in a life-cycle of procurement, manufacture, use, maintenance, and discard (Schiffer 1972: 157-160). Each of these processes took place in a specifiable location, and if the material was discarded at its place of use, archaeological investigation of locations and their contents should enable the relationship between location and past activity to be determined. In his later work, he gave greater weight to post-depositional environmental changes (Schiffer 1987).

What happens when people leave places? The term ‘abandonment’ implies disuse and an end to claims of ownership or interest. It can however include a wide range of circumstances affecting the decision to abandon, and of behaviour after the decision. Abandonment may be planned or unplanned, sudden or gradual, temporary, episodic, long-term, or permanent. Abandonment may occur at the level of the region, the settlement, a building, or a room or activity area. After abandonment, a building or settlement may be allowed to decay, or be destroyed, maintained for storage or reoccupation, or rebuilt and reoccupied. Possessions may be lost, discarded, removed for reuse, scavenged later, or deliberately left behind. Each of these possibilities is likely to have different implications for the archaeological record and its interpretation. The archaeological assemblage itself may be a palimpsest, with materials having been deposited or removed in a sequence of events rather than a single act of discard or abandonment. This paper explores how these ideas may be used to interpret aspects of the ground stone assemblage from Bestansur, an Early Neolithic settlement in Iraqi Kurdistan, excavated by the Central Zagros Archaeological Project (CZAP) from 2012 onwards. In many ways, the assemblage is similar to those found at other sites of this period in the Zagros, but there are some puzzling examples of spatial patterning. Some types of artefact which might be expected to be present are absent, or present only in low numbers.

Schiffer proposed several categories of refuse. Primary refuse refers to items discarded at the end of their use-life, in the location where they had been used. Secondary refuse is the discard of items in locations other than where they were used. Their removal will thus leave little or no trace in the place where they were used. The location of secondary refuse may itself be temporary, until it is taken to its final place of deposition. Provisional discard is the practice of storing

2. Site formation theory The detailed and scientific study of site formation processes and the development of assemblage formation theory underwent a step change with the advent of processual archaeology in the 1970s. Ascher (1961, 1968) identified two problems in archaeology: missing evidence, and interpreting the evidence 9

Stone Tools in the Ancient Near East and Egypt items with potential practical value, or large hindrance value, in out-of-the-way places, either for possible reuse or later discard. Provisional discard may also be linked to the nostalgic value of items which played a part in people’s earlier lives. These practices could be associated with the continuing use of a location, but he linked two practices specifically with abandonment (of activity areas, structures or settlements, and not necessarily permanently) (Schiffer 1987: 89-98). De facto refuse consists of ‘tools, facilities, structures and other cultural materials that, although still usable (or reusable), are left behind when an activity area is abandoned’. Curate behaviour refers to the process of removing and transporting still-usable or repairable items from the abandoned activity area to another location for continued use. De facto refuse is thus an accretion behaviour, whereas curate behaviour results in depletion (LaMotta and Schiffer 1999). Assuming a ‘least effort’ model of abandonment behaviour, curated items are likely to be portable, still usable, and have a high replacement cost. Objects left behind as de facto refuse, on the other hand, are likely to be difficult to transport, easy to replace, and have little residual utility.

opposing states, and the transition from the first to the second is a single final event. This may not have been the case in Early Neolithic settlements: seasonal activities such as hunting and herding may have meant that part or all of the settlement was not permanently occupied. Asouti and Fuller (2013) note that the literature on Early Neolithic settlements in South-West Asia focuses on villages and the built environment as the focus for social life and cultural practices, with ‘the landscape’ seen simply as a source of subsistence and raw materials. They point out that there are actually relatively few Early Pre-Pottery Neolithic (PPN) habitation sites in South-West Asia, and there are other sites interpreted as transient hunter-foraging camps and activity areas (Asouti and Fuller 2013: 311314). They suggest that early PPN sites might better be seen as ‘temporally variable nodes of habitation and co-residence, rather than as permanent, year-round, sedentary villages with fixed agricultural and foraging territories’. Second, the processual literature, building on Schiffer’s work, tends to describe examples which represent a utilitarian view of discard and abandonment behaviour, based on an implied model of decisions based on economic efficiency, tidiness and cleanliness (Deal 1985; Halstead et al. 1978; Hardy-Smith and Edwards 2004; Hayden and Cannon 1983; Tani 1995; Wilson 1994). Materials are consumed until they are exhausted, and tools used until they wear out, and if they have an alternative potential function are then recycled or scavenged. If not, they are discarded as valueless (the term ‘trash’ is widely used in the American literature). Discarded refuse, seen from this perspective, is a potential problem (hazardous, untidy, unclean), ‘a nuisance, needing to be removed from places where people do things’ (Martin and Russell 2000: 57). This model may be true of modern Western societies; it may not reflect values and practices in early farmer-forager settlements. An alternative perspective challenges the idea of discarded refuse having no value. Some materials which no longer have a mundane or utilitarian function may have had a continuing meaning to people in past societies. Examples include burials and grave goods, hoards, votive offerings, and midden accumulations from feasting. So the concept of rubbish is culturally specific, and the meaning of discard can only be considered in the context in which it takes place (Hodder 1987). The concept of ‘structured deposition’ was developed by Richards and Thomas (1984), looking at possible ritual functions for the Late Neolithic henges of Wessex. They suggested ritual deposits could be distinguished from straightforward deposits of unwanted ‘domestic’ waste deposits if they showed repeated patterns of association (or avoidance) between materials. The act of deposition itself may have been a socially meaningful ceremonial practice. LaMotta and Schiffer point out that the ritual deposition of objects

Schiffer’s framework of discard and abandonment processes has been adopted and developed by other researchers. A range of causes and types of abandonment have been identified, including punctuated or episodic mobility such as pastoralism or seasonal agricultural labour and settlement instability due to depletion of natural resources (Baker 1975; Deal 1985; Graham 1993; Horne 1993; Joyce and Johannessen 1993; Tomka 1993). Changes and innovations in cultivation strategies such as choice of crops, storage technology, ways of dealing with weeds, pests and diseases, managing soil fertility and drainage, may also lead to settlement shifts (Stone 1993). Ethnographic data, particularly from North and Central America, has been used as evidence of discard behaviour (papers in Cameron and Tomka 1993; Deal 1985; LaMotta and Schiffer 1999; Lange and Rydberg 1971; Metcalfe and Heath 1990; O’Connell 1987; Papadopoulos 2013; Shahack-Gross et al. 2004; papers in Staski and Sutro 1991; Stevenson 1982). Much of this literature focuses on abandonment at the settlement level, rather than the smaller scale of individual activity areas or buildings. 3. Abandonment and discard in Early Neolithic South-West Asia The literature on discard and abandonment processes has limitations when applied to the Early Neolithic of South-West Asia. Some research (particularly cases based on ethnographic examples of economic abandonment caused by depletion of mineral resources such as gold and coal) works on the implicit assumption that use/discard and occupation/abandonment are 10

D. Mudd: The archaeology of discard and abandonment can result in enriched floor assemblages, which could be confused with abundant de facto refuse (LaMotta and Schiffer 1999: 23). The concept of structured deposition has been widely influential in the study of British and European prehistoric archaeology, although Garrow has argued that a tendency may have developed for structured deposits to be uncritically labelled as ‘ritual’ when there may be ‘everyday’ reasons for their patterning (Garrow 2012).

area of the mound, the floor was clean and had been replastered, but there was a small dirty area with ash, animal bone, pot fragments and other household waste. This was interpreted as a temporary accumulation of oven rakeout, deposited before it was taken out through the roof to be dumped in a midden between buildings. At some stage in the life of the building, its southern half burnt down, was walled off and subsequently uninhabited. The deposits and artefacts inside it, including a grindstone, were interpreted not as discard, but as floor assemblages and items built into the walls and ceiling of this space at the time it burnt down. In the South area of the mound, a third type of context was excavated. Open areas and disused buildings were infilled in the intervals between occupation phases and used as the foundation for rebuilding. The infill was used for human burial, perhaps as foundation deposits for the new buildings. In one area, the infill was household waste, including worn-out stone tools; in another area, Unit 2006, the infill was a relatively sterile deposit of silty clay, with only sparse inclusions of charcoal and plaster fragments and plant remains. Its possible source was recycled building materials from mudbricks, mortar and roof debris.

Less attention has been given to discard and abandonment in the study of Epipalaeolithic and Neolithic South-West Asia, although several studies have examined processes of refuse discard in the habitation phase of sites. At the Early Natufian site of Wadi Hammeh 27, Hardy-Smith and Edwards found a range of refuse management practices (2004: 281-2). Most refuse consisted of the remains of food processing and artefact production sequences, and was discarded on internal hut floors with only a single secondary deposit of refuse cleared away from the residential dwellings. Exterior spaces remained relatively clear of refuse. De facto refuse and larger items were placed near the walls, with the central space freer of debris. These findings were consistent with interpretations of other Natufian sites (e.g. Goring-Morris 1988), and the authors concluded that at this early stage of sedentary life, Natufian people ‘had not yet tailored their indifferent household refuse disposal practices to the long-term requirements of sedentary living.’ In contrast, the authors cited examples from the succeeding PPNA and PPNB, when houses at sites such as Netiv Hagdud, Qermez Dere, Beidha and Dja’ade were kept clean, with refuse swept off their plastered floors and deposited in external middens.

The authors conclude that the removal and reuse of household waste and building materials in these different ways had both practical and symbolic functions. Following these practices enabled houses to be kept clean, and the useful material in middens to be recycled. At the same time, the reuse of earlier material in new building and burial of the dead may have symbolised continuity with previous buildings and the people who lived in them. Following these rules and practices gave structure to communal life. Martin and Russell’s interpretation of discard management in the PPNB suggests that more nuanced strategies had developed since the Natufian practices described by Hardy-Smith and Edwards. They illustrate the varied ways in which activity spaces and buildings can be abandoned and reoccupied.

Hardy-Smith and Edwards refer to the Wadi Hammeh buildings as ‘residential dwellings’, and the PPNA/B examples as ‘houses’, with an implicit assumption that the functions of the buildings and the intensity of their occupation in these periods were comparable. In two Natufian buildings at Eynan, Samuelian found stone and faunal floor assemblages similar to those at Wadi Hammeh (Samuelian 2013: 181). He takes a different slant on the functions of the buildings. He asks the reader ‘to envisage a group of artisans seated on stone blocks around the central hearth at the edge of the covered area. There, they could have remained in partial shelter but also have taken advantage of good light.’ This has a different connotation from a ‘residential dwelling’.

The association between ground stone tools and discard management at the PPNB site of Ayn Abū Nukhayla in southern Jordan has been examined by Kadowaki (2010). It was occupied over a short timespan of 180260 years. Its architecture consisted of sub-round stone structures with shared walls, no intervening gaps, and no doorways. Like the Çatalhöyük buildings, they were presumably accessed through their roofs. He examined spatial patterning in the 309 items of ground stone found in 11 loci. He observed several common features in their refuse structure – the high density of refuse, size-sorted patterns of distribution, a high proportion of complete ground stone tools, and functional associations of tools, such as toolkits for milling and pigment-processing. In some areas, the patterning was distorted by post-depositional scavenging, cleaning

Refuse disposal practices in several buildings at PPNB Çatalhöyük have been examined by Martin and Russell (Martin and Russell 2000), who identified several types of refuse context. Access to the houses was through the roof, and refuse could not therefore have been simply swept out of the door. In Building 1 in the North 11

Stone Tools in the Ancient Near East and Egypt activities before the building was abandoned, and the dumping of refuse such as broken tools into abandoned rooms. Some abandoned structures were reoccupied, possibly as outdoor activity areas, and their ground stone floor assemblages were less dense and less varied than previously. Kadowaki gives two alternative interpretations. Either this was a distinct reoccupation phase for the whole site, with new inhabitants changing it from a relatively sedentary settlement to a more ephemeral camp, or the abandoned buildings were reoccupied by the existing inhabitants during the same habitation phase whilst other houses were still occupied. It is one of the very few studies to give ground stone a major role in interpreting the history of a site.

Early (Pre-Pottery) Neolithic occupation and later Neo-Assyrian and Sasanian levels, with an 8 m high central mound. Much of the mound itself is from later periods. The total area of the ancient settlement (from geophysical survey) is ca. 7,000 m2, and an area of about 450 m2 has been excavated. Intact Neolithic deposits and architecture survive at a depth of ca. 30-50 cm below the modern topsoil, and at similar absolute heights in the base of the mound and in the surrounding fields. The fields have been subject to modern ploughing. During seven field seasons 201217, 13 widely distributed trenches have been excavated (and a 14th to uncover a Sasanian building) (Fig. 1). These have uncovered open areas used for butchery, craft production and cooking. No pottery was found in the Neolithic levels. There are groups of rectilinear buildings constructed of mud-brick and tauf. The buildings are oriented NW-SE, suggesting communitywide planning and collaboration. The radiocarbon dates presently available give an occupation period of ca. 500 years, though not necessarily continuous, between 7720-7580 cal BC and 7175-7055 cal BC.

How can these ideas about discard and abandonment help to interpret aspects of the ground stone assemblage from Bestansur? 4. Early Neolithic Bestansur Bestansur is a tell site in Iraqi Kurdistan, excavated by CZAP investigating the origins of sedentism, domestication, and agriculture of the Zagros region (Matthews and Matthews in prep.). The site is in the western Zagros foothills, at 550 metres above sea level, close to a perennial spring thought to have been present in the Early Holocene. The site has extensive

The site is located close to the boundary of several ecozones – flat steppe, river and marshlands, and the foothills and high mountains of the Zagros. This would have enabled the inhabitants of Bestansur to exploit a wide range of animal, vegetable and mineral resources.

Figure 1. Plan of the Bestansur site. Image – CZAP.

12

D. Mudd: The archaeology of discard and abandonment The abundance of female goat bones and the use of dung as a fuel, suggest herd management, on or close to the site (Bendrey 2014). There are larger mammals such as gazelle, red deer and roe deer, as well as small animals such as fox and hare. Dietary resources included water birds, crab and fish, reptiles such as tortoise and snake, and large quantities of molluscs. The bones of sheep, pig, cattle and red deer are more common in open butchery and discard areas further from the centre of the mound. Fish bones, on the other hand, are more common towards the centre, as are commensals such as mice and other small mammals. Preservation of archaeobotanical material was not good, although small amounts of cereals (einkorn wheat) and pulses were recovered (Williams 2014). Phytoliths from grasses and reed leaves and stems are abundant, indicating exploitation of grasslands and wetlands close to the site. These plants were used as fuel, matting, basketry and fodder. There was virtually no charcoal, unlike comparable Zagros sites.

time, with a probable lifespan of a few decades at most (Cameron 1991; Frankel and Webb 2006; Joyce and Johannessen 1993; LaMotta and Schiffer 1999; Lange and Rydberg 1971; Lightfoot 1993; Metcalfe and Heath 1990; Montgomery 1993; Sánchez-Polo and BlancoGonzález 2014; Watts 2012; Webb 1995, 1998, 2000). The abandonment and reconstruction of decaying rooms and buildings must have been routine for the Neolithic inhabitants.

In the Early Holocene, the plains and the Zagros foothills were steppe with very scattered tree stands (van Zeist and Bottema 1991: 122-3, figs. 42, 43). Without dense tree cover, wild einkorn forms dense stands, with yields equivalent to those of traditionally cultivated crops. For humans, the spread of wild cereals and annual grasses would have given a much-increased dietary yield from the land, and a corresponding increase in its carrying capacity. Food storage outside, and later inside, buildings would have enabled people to benefit from an extended annual period of reliance on wild foods.

Trench 9 (6 m x 6 m) also contained traces of Neolithic activity. Ash deposits, areas of burning, and animal bone and mollusc deposits, with stone pounding and grinding tools, all suggested the processing and preparation of food in an external area.

The most significant archaeological remains were excavated from Trenches 7, 9, 10, and 12/13 (Matthews et al. 2012, 2013a, 2013b, 2014a, 2014b, 2016). Trench 7 (6 m x 6 m) contained rectilinear architecture, with a room (Space 16) 2.5 m x 2.4 m. In this room, over 90 ground stone items were found, with some chipped stone. In the external area there were deposits of mollusc shell, more ground stone items, fire installations, and a male and female burial. The ground stone in Trench 7 is discussed in more detail below.

Trench 10 is the largest trench. Excavations exposed a rectilinear structure Building 5, ca. 12.5 m x 7.75 m, with small internal room spaces and an enclosed courtyard (Fig. 2). This building overlay the footprint of an earlier structure with painted walls (Building 8). The packing deposits of the large room Space 50 contained multiple deposits of sub-floor disarticulated human burials (MNI > 55), many of whom were children and adolescents. The floor packing of Building 5 is archaeologically sterile, reminiscent of Martin and Russell’s Unit 2006 at Çatalhöyük. The large number of burials suggests a supra-household social network. Some skulls are foundation burials for Building 5; others were cut into the floors, foundation packing, and underlying walls of Building 8. The excavation and interpretation of this space and the burials are continuing, and it would be premature to comment at this stage on the significance of the ground stone in these buildings.

The material culture includes chipped stone tools made from Anatolian obsidian from sources some 800-1200 km distant (Richardson 2013), evidence of spheres of interaction, probably through a set of exchange networks which do not necessarily imply long-distance travel. There is no archaeological evidence (e.g. of fire, inter- or intra-communal violence) to explain the termination of continuous occupation of the settlement as a whole. The site was abandoned for a long time at some stage after the Pre-Pottery Neolithic, with occupation resuming only in the Iron Age. The existence of Chalcolithic and Bronze Age sites in the vicinity of Bestansur suggests that the abandonment was local and not part of a regional event. The population may have been drawn to the development of the massive tell site of Yasin Tepe, 19 metres high and covering 10 hectares, only 2 km from Bestansur (Mühl 2012).

5. The ground stone assemblage 5.1 The assemblage The assemblage comprises 424 artefacts which are recognisable tool types and pieces of worked stone or pigment, using Wright’s classification system for Çatalhöyük (Wright et al. 2013). A further 412 are pieces or groups of manufacturing debris and unworked stone. The total weight of ground stone recovered was 511 kg. Of the 424 artefacts, 256 (60%) were found in Neolithic

There is good evidence of the decay and reconstruction of individual buildings, sometimes on the footprints of earlier structures. The architecture is mud-brick and tauf/pisé. Even with regular maintenance, buildings made from these materials degrade over 13

Stone Tools in the Ancient Near East and Egypt

Figure 2. Composite plan of excavated walls, spaces and features in Trench 10, end of Spring 2017 season. Walls of Building 5 follow the alignment and plan of earlier Building 8. Image – CZAP.

levels, and a further 154 (36%) from post-Neolithic levels. 14 (3%) were surface or topsoil finds. All the artefacts (except a small number of spherical slingshots) were assigned to the Neolithic period on the basis of morphology. Stone artefacts may be displaced laterally or vertically by human activity such as trampling, ploughing and earth-moving (Schiffer 1977; 1987: 121-

9) and biological processes such as faunal burrowing, disturbance by tree roots, and alternate wetting and drying (Villa and Courtin 1983). However, the artefacts in higher levels do not seem to be a random selection of tool types: the number of handstones and querns is high. The tools are more likely to have been taken selectively from earlier levels, reused and redeposited. 14

D. Mudd: The archaeology of discard and abandonment 5.3 Site comparisons

This implies that for later inhabitants of the site, ground stone tools had an extended practical use.

Table 1 compares the number of ground stone tools in the main classes excavated from Early Neolithic Zagros sites, based on the classification of Wright et al., 2013. Where the site is multi-period, and where the original author has identified items from Early Neolithic phases, these numbers have been used. In order to simplify comparison, in some cases, tool classes have been translated from those of the original author, using their published descriptions and illustrations. Beads, pendants, and figurines are excluded.

5.2 Raw materials There is a low limestone outcrop above the modern village, some 300-500 m from the mound and extending several kilometres northwards. This hill shows extensive quarrying, of unknown period, but unmechanised. This outcrop and the river bed are very likely to be the source of limestone in the archaeological assemblage (Mudd 2016). There are sources of sandstone and pigments (ochre, gypsum and a green pigment, celadonite) at the base of low foothills within 4-5 km of Bestansur. 89% of the ground stone artefacts are limestone or sandstone. The limestone itself is fairly hard (Mohs 3-5), tough and fine-grained. The sandstone is harder (Mohs 6-7), and more resistant to erosion. Sandstone of different granularities was selected for different types of abrading tool. The raw material for some of the artefacts (igneous and metamorphic rock such as granite, basalt, gabbro, and marble) comes from sources higher in the Zagros some 30-40 km away from Bestansur. Access to some of these non-local resources of stone, animal, and perhaps plant resources may have required periodic absence from the settlement.

Site

Jarmo Shimshara

The total number from Bestansur is large enough to make numerical comparisons, and the range of tool classes is similar to other sites. The commonest tool classes at most sites are pounding, coarse grinding, and fine abrading tools, and Bestansur follows this pattern. Its polishing tools appear to have been used particularly for polishing and burnishing plaster floor and wall surfaces. The proportion of fragmented tools is broadly similar across the sites, ranging from 2% at Nemrik 9 to 21% at Hajji Firuz Tepe. Bestansur, at 11%, is in the middle of the range. The exceptions are Abdul Hosein, where the

Zawi Chemi Shanidar

Tepe Guran

Ali Kosh

Hajji Firuz Tepe

Tepe Abdul Hosein

315

281

626

36

30

2982

836

155

597

26

14

2647

378

Nemrik Bestansur 9

N analysed

2319

45

N complete tools

414

37

N tool fragments

1903

8

30

29

7

16

54

46

Fragments as % of all tools

82%

18%

16%

5%

21%

53%

2%

11%

A Pounding tool

160

2

70

29

195

0

13

1006

119

B Coarse grinding tools

136

4

72

75

112

12

5

390

78

C Fine abrading tools

81

1

36

5

8

7

1

96

89

D Polishing tools

421

6

26

31

67

6

0

125

38

E Grooved tools

0

0

13

0

0

0

0

7

1

F Cutting tools

64

6

76

1

169

3

0

104

1

G Perforated tools

0

0

0

0

0

0

0

0

0

H Vessels

1323

22

1

34

37

0

7

21

6

W Weights

81

3

0

10

35

2

2

788

72

X Miscellaneous

28

1

31

0

3

3

2

164

20

2317

45

325

185

626

33

30

2701

424

101

205

2802

836

Tools Y Debitage Z Unworked All

2 2319

207 45

325

185

626

33

30

Table 1. Comparison of the ground stone artefact numbers in the main classes excavated from Early Neolithic Zagros sites, based on the classification of Wright et al. 2013. Beads, pendants and figurines are excluded. Where the site is multi-period, and where the original author has identified items from Early Neolithic phases, these numbers have been used. In order to simplify comparison, in some cases, tool classes have been translated from those of the original author, using their published descriptions and illustrations. (Adams 1983; Hole et al. 1969; Mazurowski 1997; Moholy-Nagy 1983; Mortensen 1970; Mortensen 2014; Pullar 1990; Solecki 1981; Voigt 1983).

15

Stone Tools in the Ancient Near East and Egypt overall numbers are relatively small, and Jarmo, where fragments (mostly vessel fragments) constitute 82% of the total. The average size of the Bestansur fragmented artefacts was 63 mm (longest dimension), and their average weight 266 g, so these were not the very small fragments which might have been missed by prehistoric inhabitants during cleaning. 55% were recovered from contexts identified as external areas, and a further 36% from topsoil. Only 3% were from building interiors. The majority of the external area contexts also contained debitage from chipped and ground stone, animal bone fragments, and fire installations. These locations represent in situ food preparation, stone and bone working. The prehistoric inhabitants did not leave broken stone tools inside their buildings, even when they abandoned them.

6. Grooved tools: ‘shaft-straighteners’

The Bestansur stone tools were not used until they wore out: the working surfaces of querns and grinding slabs show very little concavity due to usewear.

7. Axes/celts

One type of grooved stone is an oval pebble with a longitudinal groove on one face, often known as a ‘shaft straightener’, and is thought to have been used to straighten wood for use as arrowshafts (Belfer-Cohen and Bar-Yosef 1981; Cosner 1951; Solecki and Solecki 1970; Usacheva 2013). They are known widely across the Fertile Crescent from the 10th to the 8th millennia cal BC (Kozlowski and Aurenche 2005). Bestansur has only one (doubtful) example. There are however, numerous chert arrowheads, so arrows were used. It is possible that people used reeds as arrowshafts, rather than wood which was not plentiful locally. Reeds would not have required straightening, hence the absence of shaft-straighteners.

The class ‘Cutting tools’ includes axes and celts. These heavy chopping tools were usually pecked and ground to give a smooth working face, and their cutting edge maintained by regrinding. A likely function would have been chopping and splitting wood. Bestansur has only one fragmented example, far fewer than

There are some tool classes which are over-represented at Bestansur, and others which are absent, or present only in low numbers. These classes are discussed in the following sections. Axe/celt/ chopper

Adze/hoe/ pick/ digging tool

Jericho

2

0

(Crowfoot Payne 1983; Dorrell 1983)

Sha’ar Hagolan

8

0

(Rosenberg and Garfinkel 2014)

Gilgal I

19

0

(Rosenberg and Gopher 2010)

Netiv Hagdud

21

0

(Gopher 1997)

Ayn Abu Nukhayla

3

0

(Kadowaki 2014)

Abu Hureyra

22

12

Nea Nikomedeia

193

57

(Pyke 1994)

Kholetria-Ortos

5

0

(Simmons 1994)

Çayönü

265

13

(Davis 1982)

Çatalhöyük

213

1

(Wright et al. 2013)

Region/site

Reference

Levant

N Syria (Moore et al. 2000)

Eastern Med

Anatolia

Zagros/Eastern Fertile Crescent Bestansur

0

0

Jarmo

77

0

Tell Sabi Abyad 1

12

0

(Akkermans 1987)

Tell Sabi Abyad 2

7

0

(Collet and Spoor 1996; Huigens et al. 2014)

Karim Shahir

42

0

(Mazurowski 1997)

Nemrik 9

43

21

(Mazurowski 1997) (Mazurowski 1997)

(Moholy-Nagy 1983)

M’lefaat

10

0

Ali Kosh

230

28

(Hole et al. 1969)

Chogha Golan

0

0

(Conard and Zeidi 2013; Zeidi and Conard 2013)

Z. C. Shanidar

45

76

(Solecki 1981)

Table 2. Number of axes/celts and digging tools at selected Early Neolithic sites (or EN phases of multi-period sites). Numbers include both ground and chipped (chert/flint) stone items.

16

D. Mudd: The archaeology of discard and abandonment 9. Weights

assemblages of a comparable size in the Zagros and more widely in the Early Neolithic of SW Asia and the eastern Mediterranean (Table 2). Although trees were sparse in the Zagros foothills, heavy cutting tools have been found in large numbers at other sites in similar ecotones. It seems unlikely that they were not present at Bestansur, and this suggests that they were removed before locations were abandoned, or were scavenged afterwards. Their manufacture and maintenance required a significant investment of time and skill, and their relatively high replacement cost may explain why they were not abandoned at Bestansur. The reason for this contrast with other sites is unclear.

Most researchers use the classification ‘Perforated tools’ to cover items with a diverse range of shapes and functions, having only their perforation in common. I have reassigned them to other classes, defined by function rather than shape, mostly to ‘Weights’. The relatively high number at Bestansur is explained by the presence of two groups (n=34 and 11) of small flat river pebbles (Fig. 3). They were snapped to give an even size and weight, and drilled near the edge for stringing. Many have boring and casings from a freshwater worm, Polychaeta, showing that they were immersed for long periods (Fig. 4). They are interpreted as net sinkers to

8. Digging tools ‘Cutting tools’ also includes digging tools (adze, hoe, pick). When I began my doctoral research, I had assumed that these would be common tools in early societies practising agriculture (or horticulture). Neolithic and traditional South-West Asian agricultural practices have been discussed by, among others, White and Wolff (2012). These tool types would have been used to clear unwanted plants from the soil, to till (loosen) the soil before planting seeds or small plants and protect them, and self-sown seeds, from predators, to remove unwanted plants between crops as they grew, and to harvest subsoil vegetables and tubers. All these techniques would have increased plant yields in fields which were planted every year. Stone tools used for digging, with evidence of hafting, have been recorded at Çayönü and Ali Kosh (Davis 1982; Hole et al. 1969). However, I found no examples of these tool types at Bestansur, and they are only recorded at a few Early Neolithic sites in South-West Asia and the eastern Mediterranean, and then only in low numbers (Table 2). There appear to be two possible explanations. The first is that digging tools were made from other materials such as wood or animal scapulae, which have not survived. But people would have needed tough tools to work soil which for much of the year would have been baked hard by the sun, and stone seems an obvious choice of material.

Figure 3. Net sinkers SF0317 in situ. Image – CZAP.

The second possibility is excavation bias. Presumably, people would have left agricultural tools in the fields where they were used, rather than taking them back to their houses. At Bestansur, the best location to grow plants would have been in the fields between the tell and river. This is an area of alluvial soil, with a high water table, and soil fertilised by silt from periodic floods. This area has not been excavated, as the project has concentrated on the buildings of the settlement. The same is true of many other excavations, and this may be why digging tools have been recovered at so few sites.

Figure 4. Stone from SF0317 showing worm casing and boring. Image – author.

17

Stone Tools in the Ancient Near East and Egypt 12. Space 16: a storage area?

catch fish: fish bones are plentiful at the site. They are a different shape to those from other Early Neolithic sites, which are usually biconical and waisted.

Excavation of the 6 m x 6 m Trench 7 in Summer 2012 revealed a large open area (Spaces 11, 17, 18, 22) containing groups of small firestones and ground stone tools, broken chert and obsidian tools, and scatters of fragmented animal bones and mollusc shells. Such large quantities of land snail shells would only have been available in the springtime. These traces were found on sequences of trampled surfaces more than 1 m in depth. Episodic cooking and eating activities were taking place here, perhaps seasonally, over a long period.

They were deposited in a group, possibly attached to the net, and damaged by later trampling. 10. Vessels The six vessels at Bestansur were all fragments (with no refits), so we know that this type of artefact was used. Jarmo is the only other Zagros site where no complete vessels, bowls or trays were recovered. Hollow artefacts such as bowls and vessels are difficult to manufacture from stone blocks, compared to flat items such as querns. Again, this suggests that vessels may have had a high replacement cost, and were therefore removed rather than discarded. They were made from high quality non-local alabaster or marble. It is also possible that these artefacts, used in quotidian meals and feasting, were kept because of their importance as mementoes of significant social events.

Space 16 was a large rectilinear room 2.2 m x 2.4 m, contained by pisé walls (Fig. 6). It held an occupation deposit with 98 ground stone items, in discrete clusters on a surface. The room had no doorway, so may have been entered from the roof or at a higher level in its walls. For excavation, the area was gridded into nine squares A-I, in order to detect patterns in the clustering of artefacts (Fig. 7). The locations of 69 stones were recorded (Fig. 8). The stones included hammerstones, anvil, abraders, handstones, polishers, grinding slabs, pestles, a quern, unworked blocks, and some debris. All were limestone or sandstone – there was no unusual or non-local raw material. Most of the stones had been deposited in small groups (sometimes piles) of 5-15 stones, each in an area ca. 50-100 cm across. These groups appeared to be sets of stones – most had one or more percussion tools (hammerstone, pestle, anvil/slab), and a coarse grinding stone. Only three groups had fine abrasion tools. Each group had one or more blanks or preforms for recognisable tool types, and unworked stone. Chert tools were found in direct association with the ground stone. Unworked stone was mostly in a east-west line in Squares D, E, and F. Debitage was found mostly in Squares G, H and I. The stones are likely to represent sets of tools and materials. Square E had fewer ground stone pieces than the others.

11. Debitage A number of contexts contained debitage. In general this debris was recovered as heavy residue from flotation: small pieces of gravel are not easily identified as debitage during excavation. The debitage enables stoneworking locations to be identified. Analysis of flaking debitage showed small quantities of harder sandstone flakes amongst large volumes of limestone debris. This suggests that sandstone was used for knapping, pecking and abrading the softer limestone. The sandstone flakes represent fragments which broke off the percussion tool, or which were deliberately removed to resharpen its edge (Fig. 5). The absence of debitage can also be informative. 30 artefacts were made from non-local igneous and metamorphic rock. Hardly any debitage was found from this rock, showing that the artefacts were imported as finished goods.

Figure 5. Debitage from stoneworking C1752. Note the smaller quantities of harder sandstone (darker colour) from the tools used to work the softer limestone (lighter). The opinion of the excavator should be ignored. Image – author.

18

D. Mudd: The archaeology of discard and abandonment

Figure 6. Plan of Trench 7. Image – CZAP.

Figure 7. Space 16, before excavation, facing northwest, showing gridlines. Image – CZAP.

A separate small room 70 cm x 70 cm, (shown as Space 12/9 in Fig. 6) contained 30 stones, all complete or fragmented flat oval river cobbles (Fig. 9). None was a recognisable tool, and none had usewear. They may have been stored as blanks, or to be used as cooking stones or floor paving.

range of ground stone tools, although they were not grouped by type. There were flint blades, and piles of butchered animal bones. Fragments of mat and basket impressions were also found. The houses contained sub-floor human burials. The area was interpreted as a dump or working area after the abandonment of the houses. The second example is a stone scatter outside House 1 at Nemrik 9 (Mazurowski 1997: 71; Pl. LXIV). This group contained large elongated polishing stones, flint balls and grinders, and Mazurowski interprets the area as a stone workshop. Because of the number of stones (unspecified, but possibly ca. 200) he concludes that it must have functioned for a relatively long period.

What are the stones in Space 16? There are two comparable stone groups from published PPN SouthWest Asian sites. The first is Zone B2 at Tepe Ali Kosh (Hole et al. 1969: 42-5; fig. 10), in a PPN level. In an area outside, between, and inside two buildings were dozens of grinding slabs, handstones, ‘sash weights’, pounders, pestles, abraders and others – the complete 19

Stone Tools in the Ancient Near East and Egypt

Figure 8. Space 16 C1243 and C1255, showing ground stone classes and identification numbers. Some very small fragments omitted. Image – CZAP.

The activities associated with Bestansur’s Space 16 are less easy to interpret, although some can be ruled out. The clusters of tools might represent workstations, but as an activity area the space would have been quite cramped. There were no large quarried blocks, so this was not preliminary stone reduction. There are many pieces of unworked stone, though most are quite small. Levels of micro-artefacts from flotation are low, and

there is very little ground stone microdebris, so we are not seeing the manufacture of ground stone tools. The pounding and grinding tools had all been used, but not so much that they required repair. There were few polishers, so this was not hide or bead processing. It is possible that the area may have been used for processing a material which has not been preserved, such as bark, wood or animal skin, but this would be 20

D. Mudd: The archaeology of discard and abandonment

Figure 9. Flat oval river cobbles BF520 and BF522 in situ. Image – CZAP.

unlikely to involve the use of a quern or pestle. In contrast to the external Spaces 17 and 18, the interior of the building was not a craft activity area. It seems most likely that the stone tools were being stored in this space, rather than being used here. It also seems likely that the roof and walls of the room had collapsed when it was used to store the stone tools: stone does not need to be stored indoors. They could have been put here in order to leave a clear space somewhere else. Such a large number of tools is more than would be needed for use by one person. Although the stones were lying on a single surface, their deposition may have been a sequence of events, not necessarily a single episode.

abandoned. In the Bestansur ground stone assemblage we can see examples of a range of discard and abandonment processes. People kept house interiors clean, removing refuse and fragments to the outside, as secondary refuse. In the external areas next to houses, they left primary refuse from food processing and craft activities, with no obvious effort to clean or tidy these areas. These practices are associated with the occupation phase of the buildings. Occupation and abandonment may be temporary or episodic, and leaving artefacts in situ or in a storage area may indicate anticipated return. When buildings decayed and were abandoned, they might be reused as storage areas and perhaps then rebuilt. When people left a building, they took with them tools with a high replacement cost or nostalgia value such as axes and vessels, and the domestic assemblage was depleted (or increased at the new location). After abandonment, useful tools such as querns and grinding slabs were scavenged for reuse and redeposited later. The final abandonment process of the settlement is unclear. It was gradual rather than sudden, but many useful artefacts were left behind. These included artefacts made of scarce or exotic material materials such as obsidian and unusual stones from distant sources. Many of these are small and lightweight: they were not simply left behind because they were too difficult to transport. Archaeology generally looks at the material which is recovered from excavation, but the absence of particular artefact types can also be useful in interpreting site formation processes: the archaeology of things that aren’t there.

Non-local raw material is almost entirely found in the upper third of Neolithic levels at Bestansur, indicating that it was used and deposited later in the occupation sequence of the settlement, or that these tools were reused later, or for a longer period than those in local raw material. All the artefacts in Space 16 are local limestone and sandstone, so they are earlier, and were not deposited at the time when the settlement itself was abandoned. Why were they still here? This large cache of raw material and tools of various types in good condition would have had a continuing potential uselife, and one might expect such items to have been retrieved or scavenged when, or after, the building was abandoned. Perhaps they belonged to a person or people who left the settlement temporarily but never returned. Perhaps they were left behind when people left this area because the stones were simply too many and too heavy to be taken away.

The life history of a stone tool includes several phases, and a simple model is shown at Fig. 10. The maker starts with unworked raw material, reducing it to an unfinished blank or roughout, and generating debitage. When manufacture is complete, the finished

13. Discussion Ultimately, most of the artefacts recovered in archaeology are there because they were discarded or 21

Stone Tools in the Ancient Near East and Egypt

Figure 10. Possible stages in the life history of a stone tool. The artefact could enter the archaeological record from any of these stages.

tool may be in active use, or may be out of active use and stored (or lost), or may be stored while awaiting repair and maintenance or recycling. After its utilitarian function is over, it may be discarded with no anticipated reuse. Discard may not be due to damage, wear, or obsolescence. Even though reuse was not anticipated, the artefact might be retrieved and put to use once more. It could also be assigned a symbolic or ritual use. People may move the artefact between different locations at these different stages, and the stage in the artefact’s life may not correspond with stages in the life of the building in which it is located. The stone tool may enter the archaeological record from any of these stages, and the nature of its location may not of itself explain the artefact’s ‘final’ state in the model. Co-location in a midden with forms of refuse such as oven rakeout and animal bone does not necessarily mean that the uselife of the stone tool was over.

I would argue that precision in terminology is required. ‘Discard’ is an appropriate term when the artefact has been deposited with no further anticipated use. For other states, ‘deposition’ is a more neutral term, and ‘storage’ may be more accurate for items which, for the time being, were not in active use, even if they remained where they were used. The abandonment of buildings is not necessarily final, and does not necessarily coincide with the abandonment of their contents. ‘Vacated’ may be a better term: a vacant building has not necessarily been abandoned. Decayed buildings were used for storage and later rebuilt. There is not a simple dichotomy between ‘used’ and ‘discarded’ artefacts, and neither is there a simple contrast between ‘occupied’ and ‘abandoned’ buildings. The lives and practices of the people of Early Neolithic Bestansur were more complex and nuanced than this, and their use of ground stone is a valuable indicator of this complexity.

14. Conclusions

Bibliography

I have tried to use inter-site variation in the frequency of ground stone tool types in the South-West Asian Early Neolithic as a way of prompting questions about the social and economic practices of their inhabitants. Inevitably, some of this variation will be simply due to excavation bias, but archaeologists must work with what they have got. In some cases, what they haven’t got may be just as informative. Site formation processes associated with discard and abandonment are a helpful starting point for interpreting aspects of the Bestansur ground stone assemblage.

Adams, R. M. 1983. The Jarmo Stone and Pottery Vessel industries, in L. S. Braidwood, R. J. Braidwood, B. Howe, C. A. Reed and P. J. Watson (eds) Prehistoric Archeology Along the Zagros Flanks: 209-232. Chicago: Oriental Institute of the University of Chicago. Akkermans, P. M. M. G. 1987. A Late Neolithic and Early Halaf Village at Sabi Abyad, Northern Syria. Paléorient, 13, (1): 23-40. Ascher, R. 1961. Analogy in archaeological interpretation. Southwestern Journal of Anthropology, 17, (4): 317-325. 22

D. Mudd: The archaeology of discard and abandonment Ascher, R. 1968. Time’s Arrow and the Archaeology of a Contemporary Community, in K. C. Change (ed.) Settlement Archaeology: 43-52. Palo Alto: National Press Books. Asouti, E. and Fuller, D. Q. 2013. A contextual approach to the emergence of agriculture in Southwest Asia. Current Anthropology, 54, (3): 299-345. Baker, C. M. 1975. Site Abandonment and the Archeological Record: An Empirical Case for Anticipated Return. Arkansas Academy of Science Proceedings, XXIX: 10-11. Belfer-Cohen, A. and Bar-Yosef, O. 1981. The Aurignacian at Hayonim Cave. Paléorient, 7, (2): 19-42. Bendrey, R. 2014. Preliminary Assessment of the Zooarchaeological Assemblage, in: R. J. Matthews, W. Matthews and K. R. Raheem (eds) CZAP Archive Report. Excavations at Bestansur and Shimshara, Sulaimaniyah Province, Kurdistan Regional Government, Republic of Iraq 15th August – 2nd October 2013. Not published. Cameron, C. M. 1991. Structure Abandonment in Villages. Archaeological Method and Theory, 3: 155-194. Cameron, C. M. and Tomka, S. A. (eds) 1993. Abandonment of settlements and regions: ethnoarchaeological and archaeological approaches. Cambridge: Cambridge University Press. Collet, P. and Spoor, R. H. 1996. The Ground Stone Industry, in P. M. M. G. Akkermans (ed.) Tell Sabi Abyad: the Late Neolithic Settlement. Vol II: 415438. Istanbul; Leiden: Nederlands HistorischArchaeologisch Institut. Conard, N. J. and Zeidi, M. 2013. The ground stone tools from the aceramic Neolithic site of Chogha Golan, Ilam province, western Iran, in F. Borrell, J. J. Ibáñez and M. Molist (eds) Stone Tools in Transition: From Hunter-Gatherers to Farming Societies in the Near East: 365-375. Barcelona: Universitat Autònoma de Barcelona. Cosner, A. J. 1951. Arrowshaft-straightening with a grooved stone. American Antiquity, 17, (2): 147-148. Crowfoot Payne, J. 1983. Appendix C: The Flint Industries of Jericho, in K. M. Kenyon and T. A. Holland (eds) Excavations at Jericho Volume 5. The Pottery Phases of the Tell and Other Finds: 622-759. London: British School of Archaeology in Jerusalem. Davis, M. 1982. The Çayönü ground stone, in L. S. Braidwood and R. J. Braidwood (eds) Prehistoric village archaeology in south-eastern Turkey: the eighth millennium B.C. site at Çayönü, its chipped and ground stone industries and faunal remains. British Archaeological Reports (International Series) 138: 73-174. Oxford: British Archaeological Reports. Deal, M. 1985. Household pottery disposal in the Maya Highlands: an ethnoarchaeological interpretation. Journal of Anthropological Archaeology, 4, (4): 243-291. Dorrell, P. G. 1983. Appendix A: Stone Vessels, Tools, and Objects, in K. M. Kenyon and T. A. Holland (eds) Excavations at Jericho Volume 5. The Pottery Phases of the

Tell and Other Finds: 485-575. London: British School of Archaeology in Jerusalem. Frankel, D. and Webb, J. M. 2006. Neighbours: Negotiating space in a prehistoric village. Antiquity, 80, (308): 287-302. Garrow, D. 2012. Odd deposits and average practice. A critical history of structured deposition. Archaeological Dialogues, 19, (02): 85-115. Gopher, A. 1997. Groundstone Tools and Other Stone Items from Netiv Hagdud, in O. Bar-Yosef and A. Gopher (eds) An Early Neolithic Village in the Jordan Valley. Part 1: The Archaeology of Netiv Hagdud: 151-176. Cambridge, Mass.: Peabody Museum of Archaeology and Ethnology, Harvard University. Goring-Morris, A. N. 1988. Trends in the spatial organization of Terminal Pleistocene huntergatherer occupations as viewed from the Negev and Sinai. Paléorient, 14, (2): 231-244. Graham, M. 1993. Settlement organization and residential variability among Raramuri, in C. M. Cameron and S. A. Tomka (eds) Abandonment of settlements and regions. Ethnoarchaeological and archaeological approaches: 54-73. Cambridge: Cambridge University Press. Halstead, P., Hodder, I. and Jones, G. 1978. Behavioural archaeology and refuse patterns: a case study. Norwegian Archaeological Review, 11, (2): 118-131. Hardy-Smith, T. and Edwards, P. C. 2004. The Garbage Crisis in History: artefact discard patterns at the Early Natufian site of Wadi Hammeh 27 and the origins of household refuse disposal strategies. Journal of Anthropological Archaeology, 23: 253-289. Hayden, B. and Cannon, A. 1983. Where the Garbage Goes: Refuse Disposal in the Maya Highlands. Journal of Anthropological Archaeology, 2: 117-163. Hodder, I. 1987. The Meaning of Discard: Ash and Domestic Space in Baringo, in S. Kent (ed.) Method and Theory for Activity Area Research: An Ethnoarchaeological Approach: 424-448. New York: Columbia University Press. Hole, F., Flannery, K. V. and Neely, J. A. 1969. Prehistory and human ecology of the Deh Luran Plain: an early village sequence from Khuzistan, Iran. Ann Arbor: University of Michigan. Horne, L. 1993. Occupational and locational instability in arid land settlement, in C. Cameron and S. Tomka (eds) Abandonment of settlement and regions. Ethnoarchaeological and archaeological approaches: 4373. Cambridge: Cambridge University Press. Huigens, H. O., van Exel, V., Flohr, P., Kelder, J. M., Koek, E. and Rooijackers, T. 2014. The Ground Stone Industry, in P. M. M. G. Akkermans, O. Nieuwenhuyse, H. Huigens and M. Brüning (eds) Excavations at Late Neolithic Tell Sabi Abyad, Syria: 135-163. Turnhout: Brepols. Joyce, A. A. and Johannessen, S. 1993. Abandonment and the production of archaeological variability at domestic sites, in C. M. Cameron and S. A. 23

Stone Tools in the Ancient Near East and Egypt Tomka (eds) Abandonment of settlements and regions: ethnoarchaeological and archaeological approaches: 138153. Cambridge: Cambridge University Press. Kadowaki, S. 2010. Ground Stone Tools, Refuse Structure, and the Life Histories of Residential Buildings at Ayn Abū Nukhayla, Southern Jordan, in Y. M. Rowan and J. R. Ebeling (eds) New Approaches to Old Stones: Recent Studies of Ground Stone Artifacts: 230-257. London: Equinox. Kadowaki, S. 2014. Groundstones and Grinding Technology, in D. O. Henry and J. M. Beaver (eds) The Sands of Time: The Desert Neolithic Settlement at Ayn Abū Nukhayla: 259-286. Berlin: ex oriente. Kozlowski, S. K. and Aurenche, O. 2005. Territories, Boundaries and Cultures in the Neolithic Near East. British Archaeological Reports (International Series) 1362. Oxford: Archaeopress. LaMotta, V. M. and Schiffer, M. B. 1999. Formation processes of house floor assemblages, in P. M. Allison (ed.) The Archaeology of Household Activities: 19-29. Abingdon: Routledge. Lange, F. W. and Rydberg, C. R. 1971. Abandonment and Post-Abandonment Behavior at a Rural Central American House-site. American Antiquity, 37, (3): 419-432. Lightfoot, R. R. 1993. Abandonment processes in prehistoric pueblos, in C. M. Cameron and S. A. Tomka (eds) Abandonment of Settlements and Regions: Ethnoarchaeological and Archaeological Approaches: 165-177. Cambridge: Cambridge University Press. Martin, L. and Russell, N. 2000. Trashing Rubbish, in I. Hodder (ed.) Towards reflexive method in archaeology: the example at Çatalhöyük: 57-69. Cambridge: McDonald Institute Monographs. Matthews, R. and Matthews, W. (eds) in prep. CZAP Vol. 2 Sedentism and resource management in the Neolithic of the Central Zagros. Matthews, R. J., Matthews, W. and Raheem, K. R. 2012. CZAP Archive Report. Excavations at Bestansur, Sulaimaniyah Province, Kurdistan Regional Government, Republic of Iraq 17th March – 24th April 2012. Not published. Matthews, R. J., Matthews, W. and Raheem, K. R. 2013a. CZAP Archive Report. Excavations at Bestansur, Sulaimaniyah Province, Kurdistan Regional Government, Republic of Iraq 21st March – 24th April 2013. Not published. Matthews, R. J., Matthews, W. and Raheem, K. R. 2013b. CZAP Archive Report. Excavations at Bestansur, Sulaimaniyah Province, Kurdistan Regional Government, Republic of Iraq, 18th August – 27th September 2012. Survey in Zarzi Region, January 2013. Not published. Matthews, R. J., Matthews, W. and Raheem, K. R. 2014a. CZAP Archive Report. Excavations at Bestansur and Shimshara, Sulaimaniyah Province, Kurdistan Regional Government, Republic of Iraq 15th August – 2nd October 2013. Not published.

Matthews, R. J., Matthews, W. and Raheem, K. R. 2014b. CZAP Archive Report. Excavations at Bestansur, Sulaimaniyah Province, Kurdistan Regional Government, Republic of Iraq 22nd March – 27th April 2014. Not published. Matthews, R. J., Matthews, W. and Raheem, K. R. 2016. CZAP Archive Report. Excavations at Bestansur, Sulaimaniyah Province, Kurdistan Regional Government, Republic of Iraq 26th March – 15th April 2016. Not published. Mazurowski, R. 1997. Nemrik 9, Pre-pottery Neolithic Site in Iraq Vol 3: Ground and pecked stone industry in the Pre-pottery Neolithic of Northern Iraq. Warszawa: Wydawnictwa Uniwersytetu Warszawskiego. Metcalfe, D. and Heath, K. M. 1990. Microrefuse and Site Structure: the Hearths and Floors of the Heartbreak Hotel. American Antiquity, 55, (4): 781-796. Moholy-Nagy, H. 1983. Jarmo Artifacts of Pecked and Ground Stone and Shell, in L. S. Braidwood, R. J. Braidwood, B. Howe, C. A. Reed and P. J. Watson (eds) Prehistoric Archeology Along the Zagros Flanks: 289346. Chicago: Oriental Institute of the University of Chicago. Montgomery, B. K. 1993. Ceramic analysis as a tool for discovering processes of pueblo abandonment, in C. M. Cameron and S. A. Tomka (eds) Abandonment of Settlements and Regions: Ethnoarchaeological and Archaeological Approaches: 157-164. Cambridge: Cambridge University Press. Moore, A. M. T., Hillman, G. C. and Legge, A. J. 2000. Village on the Euphrates: from foraging to farming at Abu Hureyra. Oxford: Oxford University Press. Mortensen, P. 1970. Tell Shimshara: the Hassuna Period. Copenhagen: Royal Danish Academy of Sciences and Letters. Mortensen, P. 2014. Excavations at Tepe Guran: The Neolithic Period. Leeuven: Peeters. Mudd, D. 2016. People and ground stone tools in the Zagros Neolithic – economic and social interpretations of the assemblage from Bestansur, Iraqi Kurdistan. Ph.D, University of Reading. Mühl, S. 2012. Human Landscape – Site (Trans-) Formation in the Transtigris Area, in R. Hofmann, F.-K. Moetz and J. Müller (eds) Tells: Social and Environmental Space. Proceedings of the International Workshop ‘Socio-Environmental Dynamics over the Last 12,000 Years: The Creation of Landscapes II (14th-18th March 2011’ in Kiel. Vol 3: 79-92. Bonn: Verlag Dr. Rudolf Habelt GmbH. O’Connell, J. F. 1987. Alyawara Site Structure and Its Archaeological Implications. American Antiquity, 52: 1. Papadopoulos, C. 2013. An Evaluation of Human Intervention in Abandonment and Post-abandonment Processes in a Deserted Cretan Village. Journal of Mediterranean Archaeology, 26, (1): 27-50. Pullar, J. 1990. Tepe Abdul Hosein: a Neolithic site in Western Iran: Excavations 1978. British Archaeological 24

D. Mudd: The archaeology of discard and abandonment Reports (International Series) 563. Oxford: British Archaeological Reports. Pyke, G. 1994. The stratigraphy, structures and small finds of Nea Nikomedeia, Northern Greece. PhD, University of Birmingham. Richards, C. and Thomas, J. 1984. Ritual activity and structured deposition in later Neolithic Wessex, in R. Bradley and J. Gardiner (eds) Neolithic Studies: a Review of some Current Research. (British Archaeological Reports, British Series 133): 189-218. Oxford: Britsh Archaeological Reports. Richardson, A. 2013. pXRF of Obsidian and Chert, in: R. J. Matthews, W. Matthews and K. R. Raheem (eds) CZAP Archive Report. Excavations at Bestansur, Sulaimaniyah Province, Kurdistan Regional Government, Republic of Iraq 21st March – 24th April 2013. Not published. Rosenberg, D. and Garfinkel, Y. 2014. Sha’ar Hagolan 4. The Ground Stone Industry at the Dawn of Pottery Production in the Southern Levant. Jerusalem: Israel Exploration Society and The Hebrew University of Jerusalem. Rosenberg, D. and Gopher, A. 2010. Food Processing Tools And Other Groundstone Implements From Gilgal I And Gilgal III, in O. Bar-Yosef, A. N. GoringMorris and A. Gopher (eds) Gilgal. Early Neolithic Occupations in the Lower Jordan Valley: The Excavations of Tamar Noy: 139-175. Oxford: Oxbow Books. Samuelian, N. 2013. A Study of two Natufian Residential Complexes at Eynan (Ain Mallaha), Israel, in O. BarYosef and F. R. Valla (eds) Natufian Foragers in the Levant. Terminal Pleistocene Social Changes in Western Asia: 172-184. Ann Arbor: International Monographs in Prehistory. Sánchez-Polo, A. and Blanco-González, A. 2014. Death, Relics, and the Demise of Huts: Patterns of Planned Abandonment in Middle BA Central Iberia. European Journal of Archaeology, 17, (1): 4-26. Schiffer, M. B. 1972. Archaeological context and systemic context. American Antiquity, 37, (2): 156-165. Schiffer, M. B. 1977. Toward a unified science of the cultural past, in S. South (ed.) Research strategies in historical archaeology: 13-50. San Diego: Academic Press. Schiffer, M. B. 1983. Towards the Identification of Formation Processes. American Antiquity, 48, (4): 675706. Schiffer, M. B. 1987. Formation Processes of the Archaeological Record. Albuquerque: University of New Mexico Press. Shahack-Gross, R., Marshall, F., Ryan, K. and Weiner, S. 2004. Reconstruction of spatial organization in abandoned Maasai settlements: implications for site structure in the Pastoral Neolithic of East Africa. Journal of Archaeological Science, 31, (10): 1395-1411. Simmons, A. H. 1994. Early Neolithic Settlement in Western Cyprus: Preliminary Report on the 1992-3 Test Excavations at Kholetria Ortos. Bulletin of the American Schools of Oriental Research, 295: 1-14.

Solecki, R. L. 1981. An early village site at Zawi Chemi Shanidar. Malibu: Undena Publications. Solecki, R. L. and Solecki, R. S. 1970. Grooved Stones from Zawi Chemi Shanidar, a Protoneolithic Site in Northern Iraq. American Anthropologist, 72, (4): 831841. Staski, E. and Sutro, L. D. (eds) 1991. The Ethnoarchaeology of Refuse Disposal. Tucson: Arizona State University. 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, (3): 237-265. Stone, G. D. 1993. Agricultural abandonment: a comparative study in historical ecology, in C. M. Cameron and S. A. Tomka (eds) Abandonment of Settlements and Regions: Ethnoarchaeological and Archaeological Approaches, Cambridge University Press, Cambridge: 74-81. Cambridge: Cambridge University Press. Tani, M. 1995. Beyond the Identification of Formation Processes. Journal of Archaeological Method and Theory, 2, (3): 231-252. Tomka, S. A. 1993. Site abandonment behavior among transhumant agro-pastoralists: the effects of delayed curation on assemblage composition, in C. M. Cameron and S. A. Tomka (eds) Abandonment of settlements and regions: Ethnoarchaeological and archaeological approaches: 11-24. Cambridge: Cambridge University Press. Usacheva, I. 2013. On the Function of ‘Grooved Stones’. Archaeology, Ethnology and Anthropology of Eurasia, 41, (4): 58-64. van Zeist, W. and Bottema, S. 1991. Late Quaternary Vegetation of the Near East. Wiesbaden: Reichert. Villa, P. and Courtin, J. 1983. The interpretation of stratified sites: a view from underground. Journal of Archaeological Science, 10, (3): 267-281. Voigt, M. M. 1983. Hajji Firuz Tepe, Iran: the Neolithic settlement. Philadelphia: University Museum, University of Pennsylvania. Watts, S. 2012. The Structured Deposition of Querns: The Contexts of Use and Deposition of Querns in the SouthWest of England from the Neolithic to the Iron Age. PhD, Exeter. Webb, J. M. 1995. Abandonment processes and curate/ discard strategies at Marki-Alonia, Cyprus. The Artefact, 18: 64-70. Webb, J. M. 1998. Lithic technology and discard at Marki, Cyprus: consumer behaviour and site formation in the prehistoric Bronze Age. Antiquity, 72, (278): 796805. Webb, J. M. 2000. Curation, expediency and discard: The ground stone industry at Marki-Alonia, in G. K. Ioannides and S. Hadjistellis (eds) Praktika tou Tritou Diethnous Kyprologikou Synedriou (Lefkosia, 1620 Apriliou 1996): 261-279. Nicosia: Society of Cypriot Studies. 25

Stone Tools in the Ancient Near East and Egypt White, C. E. and Wolff, N. P. 2012. Beyond the house and Into the fields: cultivation practices in the Late PPNB, in B. J. Parker and C. P. Foster (eds) New Perspectives on Household Archaeology: 267-288. Winona Lake, Indiana: Eisenbrauns. Williams, A. 2014. CZAP Archive Report. Excavations at Bestansur, Sulaimaniyah Province, Kurdistan Regional Government, Republic of Iraq 22nd March – 27th April 2014, in: R. J. Matthews, W. Matthews and K. R. Raheem (eds) CZAP Archive Report. Excavations at Bestansur, Sulaimaniyah Province, Kurdistan Regional Government, Republic of Iraq 22nd March – 27th April 2014. Not published. Wilson, D. C. 1994. Identification and assessment of secondary refuse aggregates. Journal of Archaeological Method and Theory, 1, (1): 41-68.

Wright, K., Tsoraki-Chan, C. and Siddall, R. 2013. The Ground Stone Technologies of Çatalhöyük, in I. Hodder (ed.) Substantive Technologies at Çatalhöyük: Reports from the 2000-2008 Seasons. Çatalhöyük Research Project Vol. 9: 365-416. London and Los Angeles: British Institute at Ankara and Cotsen Institute of Archaeology UCLA. Zeidi, M. and Conard, N. J. 2013. Chipped stone artifacts from the aceramic Neolithic site of Chogha Golan, Ilam province, western Iran, in F. Borrell, J. J. Ibáñez and M. Molist (eds) Stone Tools in Transition: From Hunter-Gatherers to Farming Societies in the Near East: 315-327. Barcelona: Universitat Autònoma de Barcelona.

26

Survey of Rock-Cut Installations at Tel Bareqet (Israel): Food Processor devices in Epipaleolithic, PPNA and the Early Bronze David Eitam Keywords: rock-cut installations, narrow conical mortars, PPNA, Early Bronze Age, Tel Bareqet

are co-directed by Y. Paz and S. Paz on behalf of the Institute of Archaeology at Tel Aviv University). In 2004, an extensive salvage excavation conducted in the lower part of the site uncovered an urban settlement surrounded by a massive fortification system, with residential quarters, streets, and public facilities, all dating to the EB II (ca. 3050-2800 BC). The rich finds in the buildings included complete pottery vessels, metal tools and weapons, large quantities of flint artifacts, and jewelry (Fig. 1; Paz and Paz 2007). Another excavation, that was conducted at the same time by the Israel Antiquity Authority, on the eastern slopes of the lower terrace, revealed a 2.5 ha settlement dated to the Hellenistic and Roman periods (Amit et al. 2008; Torgë 2012). After the salvage excavation, the lower terrace was completely destroyed, making way for a modern industrial zone. In 2006 and 2008, excavations of the upper mound were conducted. The vast majority of the architectural remains and other finds belong to the EBII town. This included fortifications, domestic and nonresidential structures (with well-preserved ceramic vessels on the floors), and some ground stones and RCIs (written by Y. Paz and S. Paz).

1. Introduction Bedrock features, or rock-cut installations (henceforth RCIs), constitute a central part of the archaeological record of the Southern Levant, appearing as early as the Epipaleolithic period and continuing throughout the historical periods. These RCIs illustrate how early cultures harnessed the rocky landscape to their advantage, making it a tool for a variety of activities (e.g. Eitam 1987; Peterson 1999). The focus of this article is on a highly dense cluster of RCIs found at Tel Bareqet, in central Israel. Some of these features are typologically dated to several periods, from the Late Epipaleolithic to the Early Bronze Age (EB) and appear to have served some specific functions. The site surface contains a large number of cupholes or cupmarks clusters, which are well-familiar from several Pre-Pottery Neolithic A (PPNA) sites throughout the Southern Levant (e.g., Gopher 1997; Rosenberg and Gopher 2010) as well as Natufian bedrock mortars and a variety of RCIs typological dated to the Late Epipaleolithic and EB. Each of these is defined by an explicitly different character of the bedrock feature as well by the correlation that combines the various features into a complex of installations. The presence of these devices in their specific locations (in some cases overlapping with each other), provides a fruitful ground to shed new light on their possible function and their significance to the social-economy to both the site and beyond. The current paper presents the RCIs from Tel Bareqet and describes their morphological and technological characteristics. As well, we will investigate their potential function and propose a dating for the various installations based on typological characteristics. A survey of these rock-cut installations was conducted by the author during the 2008 excavation season.

While the EB II period, which is the most prominent at Tel Bareqet, is well represented in numerous aspects of the material culture, the Epipaleolithic and PPNA occupations left none or fewer visible remains. The lithics from Area DS, located near a large complex of RCIs (L.3042), include 911 items of which 62 are tools (mostly retouched flakes, denticulate/notches and burins) and the rest are chips and chunks. As opposed to the majority of EB lithics in the excavated areas, here the only artifact clearly related to the EB is a Canaanite type sickle blade, while a single, small, broken axe was found in the assemblage. Most of the Area DS tools can be related to the Natufian and the PPNA, but a clear assignment of the material to the Natufian or the PPNA is not possible, and most likely it represents a mixture of both. Nevertheless, the PPNA presence is more pronounced, as exemplified by the small bifacial tools with the tranchet-style removal and by the Beit Ta’amir Knife (written by R. Shimelmitz).

Tel Bareqet (Khirbet Burnat, grid 14610/15775) is located on the southeastern flanks of the Ayalon basin, ca. 8 km. north-east of the town of Lod, in central Israel. The site is ca. 5 hectares in size, comprising a western upper mound (ca. 1.2 ha) and a lower terrace located on the eastern side (excavations at the site 27

Stone Tools in the Ancient Near East and Egypt

Figure 1. General plan of the Tel and excavations areas.

The bedrock exposures and elevated rocks were located on the surface at the flat summit of the site, as well as along its slopes. Most of the RCIs were cut into bedrock or into the elevated rocks at the northwest part of the summit – the upper EB city (Fig. 2). Few RCIs were located on the eastern edge or on the northern slope of the site (Loci 3057, 3058, 3060 and other installations, excavated by the Israel Antiquity Authority, date to the Roman and Byzantine periods, and are not dealt with here, see, Torgë 2012).

cupmark, basin and grinding implements chiseled in stone (ground stones) share identical functions as those cut in rock or designed as vessels. The metric data of RCIs focuses on length, width and depth and/or diameter (and depth in the case of symmetrical features). In installations with uneven or irregular bottoms, the lower diameter was measured one centimeter above the base. Special attention was paid to recording the working surfaces of the features. Many non-metric parameters and the description of the artifacts found in their illustrations were also included (see Fig. 16: 1 upper). Our system of classifying RCIs and ground stones primarily follows Wright’s detailed terminology and typology criteria (1992) with some necessary changes. We have done this in order to retain a unified terminological and typological system, essential for the facilitation of further studies of stone tools. The published typological list (Eitam 2010) includes those ground stone groups and types that are relevant to the study of Natufian and PPNA rock-cut installations, but it can easily be adjusted to suit later periods (Eitam and Squitieri in preparation).

2. Methodology Various methodologies for the study of stone tools (ground stones and RCIs) are used in the Levant; most examine late prehistoric features, in particular ground stone artifacts, through usewear tests (e.g., Dubreuil 2024) or by analytical starch test of phytoliths and other material (e.g., Buonasera 2005; Power et al. 2014). The methods for studying RCIs used in the Tel Bareqet were the result of long-term field research projects covering Epipaleolithic to the EBA and Iron Age, as well as later periods (e.g., Eitam 2005, 2010, and see, Miller et al. 2014; Rosenberg and Nadel 2014). Our method focused on tool use, as the essence of any stone tool is determined by its function. Consequently, a comprehensive methodology and comparative classification system, combining ground stones and rock-cut installations from various historical and prehistoric periods, was established, as mortar,

The morphological classification systems of ground stones are generally based on sequences of reduction (Wright 1992). When possible, we have distinguished between fabrication (the production of the tool) and use-wear marks. Stages and methods of fabrication can 28

D. Eitam: Survey of Rock-Cut Installations at Tel Bareqet (Israel) sometimes be traced in RCIs when the manufacturing process was halted or hampered by some nonhomogeneous structure in the rock formation (for example, if it contained a hard limestone pebble or a flint core). The method of cutting the rock around the hard pebble, deepening the mortar, is probably similar to the usual method of manufacture.

pressing floor, assisting the flow of grape juice into the collection vat. The round, bell-shaped collection vat (L. 3021: upper diameter 73-75 cm, middle diameter 8085 cm, lower diameter 54-58 cm, depth 75 cm) ended with a small sunken vat in the bottom (measuring 22 cm upper diameter, 5 cm lower diameter, 8 cm depth). This collection vat was found damaged: a side door had been opened through it at a later period. That the side vat (irregular in shape and cross sections) is part of the winepress is evidenced by the curved side of the pressing floor which embraces the vat (L. 3022, measuring 45 x 27 x 47 cm). The irregular shape is most likely due to a combination of more recent reuse and karstic erosion (Figs. 3-4). Nine concave and shallow cupmarks (L. 2023-3029) were cut nearby, to the north of the wine press; one of them is cut into the pressing floor and can be considered a part of the original winepress.

The function of stone tools is studied through usewear analysis and by observing its changing uses throughout the history of the tool, according to ethnographic parallels, as well as by experimental operation (Eitam et al. 2015). Other methods of examining the manufacture of RCIs include tests of measurable residues, such as phytoliths (Power et al. 2014). Typological dating is done through the comparative study of variations between stone tool types. Indeed, some stone tools were typical of, or even exclusive to, a specific culture or period (Eitam 2008, 2009). Ground stones found on the surface were collected during the study of the site, in addition to the detailed recording of the RCIs.

L. 3030 (Fig. 5): A deep, concave basin cut into a small bedrock, near a massive structure (area HS), at the edge of the mound. The rock surface around the basin is slightly slanted toward the basin. The shape of the basin varies slightly from the other EB basins.

The bedrock features found close to each other, gathered in groups of a few or up to as many as dozens, were defined as a ‘complex of installations’. This definition refers to the close spatial distribution of the rock-cut installations (on suitable bedrock) and does not necessarily indicate an equivalent dating between the rock-cut installations and their coexistent use. In some cases, one RCI cuts into another one, either by accident or through recycling a prior installation by adding a new one to it. When a correlation between two (and up to dozens) of rock-cut features can be clearly established, they are referred to as ‘composite installations’.

L. 3032 (Fig. 6): A complex of 18 cupmarks of different types, cut into the rock surface on the north side, close to a low heap of collected fieldstones (and partly covered by it). A large square boulder, possibly connected to the heap of fieldstones, had been placed on the bedrock, partially covering some of these cupmarks. L. 3033 (Fig. 6): A complex of eight cupmarks cut into a rock surface west of the heap of fieldstones, partly covered by it and sharing the same bedrock as L. 2032. A few of the cupmarks are cut into fragments of rock face under the east side of the fieldstone heap.

3. Data – the rock-cut installation complexes

L. 3034: A complex of five small, shallow basins and cup-marks cut into four elevated rocks, 7 m east of L. 2033.

In all, 45 complexes of individual RCIs and composite installations were found on the upper tell, comprising altogether 985 rock-cut specimens. Approximately two dozen single installations (mainly cupmarks) were cut into small or medium elevated rocks, scattered between the installation complexes (Fig. 2). The following is a description of the different installation complexes, supplemented with detailed descriptions of the installations, including measurements, in Appendix A (supplementary data).

L. 3035: A composite installation of seven shallow cupmarks arranged in a circle (one cupmark is positioned in the centre) cut into a flat rock measuring 2 x 2 m. Similar installations have been found in other PPNA sites (e.g., Netiv Hagdud, Eitam 2013: Fig. 26:4). L. 3036: A complex composed of a medium deep basin, one regular cup, and three shallow cupmarks, all cut into bedrock. It is located 2 m east of L. 3034.

L. 3020 (Figs. 3-4): Wine press cut into wide bedrock, located near the north edge of the tell, consisting of a pressing floor, a collection vat, and an additional vat. The kidney-shaped pressing floor is surrounded by curved sides (10 cm height) which wrap around both the collection vat and the additional vat. The sloped (6O) pressing floor (L. 3020) measures 175 x 182 cm. A shallow, short groove has been cut into the end of the

L. 3037: A complex of four shallow cupmarks (three with small cups at the bottom). L. 3038: A complex of one small basin with cup at bottom and three shallow cupmarks (2 with small cups at their bottoms). 29

Stone Tools in the Ancient Near East and Egypt

Figure 2. General plan of the rock-cut installations.

L. 3039 (Figs 7, 8): A composite installation of sixty-four cupmarks and three small shallow basins with cups at their bottoms, cut into unlevelled, oblong bedrock measuring 6 x 1.5 m. This installation is a typical PNNA feature (see discussion).

L. 3040: A medium basin with a flat bottom and a cup cut into the north portion of bedrock. L. 3041 (Figs. 9-13): A complex of 30 installations located in the centre of the western part of the upper 30

D. Eitam: Survey of Rock-Cut Installations at Tel Bareqet (Israel)

Figure 3. L. 3020: EB winepress with pressing surface drained to collecting vat, adjacent vat on the west of the press, and 2 cupmarks on the east. Figure 5. L. 3030: round large basin cut into oblong bedrock (possibly EB), a cupmark, and a small round basin. Figure 4. L. 3020: collecting vat of EB winepress (side of vat was cut in recent time).

mound, to the west of L. 3065 and comprising part of L. 3040 on its south side. The complex contains 26 cupmarks cut into a large bedrock surface measuring approximately 5 x 3 m, of which 19 are clustered on the south edge of the bedrock (see discussion on PPNA RCI); two narrow conical mortars (henceforth NCM), one of which was reused as a collection vat with an oval basin; a large round basin; and a small concave mortar cut on wide bedrock. Most of this rock surface was covered with a layer of earth containing additional lithic finds (see above). The two narrow conical mortars are Late Natufian installations (one with hard stone

Figure 6. L. 3032 and 3033: cupmarks on flat rock face. Covered partly by tumulus or a heap of small stones.

31

Stone Tools in the Ancient Near East and Egypt

Figure 7. Plan of L. 3039: PPNA threshing floor with cupmarks cut on an unlevelled oblong bedrock exposure.

at the bottom, part of the rock formation), while the concave basins may possibly date to the EB (see discussion). L. 3042 (Figs. 14-16): A complex consisting of a large basin with a flat bottom and a collection vat (Late Natufian NCM in secondary use during the EB), one small mortar eroded on the bottom, and 31 cupmarks cut into a wide bedrock (measured 7.7 x 5 m). This complex was located 5 m east of the heap of fieldstones. Seven cupmarks were assembled in a circle, with one cupmark in the centre. Five cupmarks made another circle, with one cupmark in the centre (see composite installation L. 3035). L. 3043: A complex of 13 shallow concave and conical cupmarks, cut into bedrock (7 x 2 m) located east of L. 3042. This composite installation is possibly a small specific PPNA RCI (see discussion). An additional nine cupmarks were cut into four elevated rocks situated around the bedrock.

Figure 8. L. 3039: PPNA threshing floor with many rock-cut cupmarks and 4 shallow basins.

32

D. Eitam: Survey of Rock-Cut Installations at Tel Bareqet (Israel) L. 3046: A complex of two small, shallow basins and a few cupmarks cut into an oblong rocky step (11 x 4 m). An additional basin is cut on the north end of the rock step, surrounded by eight cupmarks. L. 3047: Four small rock surfaces with 13 cupmarks, located about 15 m west of excavation area LS and 5 m south of L. 3046. This composite installation is possibly a specific PPNA RCI (see discussion). L. 3048: A complex of 11 installations on oblong bedrock (5 x 12 m), situated 7 m east of a large Roman quarry. It contained one large, one medium, and four small basins, a Natufian narrow conical mortar (reused as the collection vat for a small basin), a small wide conical mortar (most likely Natufian), and three cupmarks. Figure 9. General plan of L. 3041: complex RCI with Late Natufian narrow conical mortar, reused during the EB as a collection vat for a large basin, Natufian small wide conical mortar, and cupmarks.

L. 3049: A large round basin with a flat bottom, cut close to a smaller flat-base basin.

L. 3050: A complex of 18 cupmarks on a rock surface, located 3 m west of L. 3049. Five cupmarks compose an open circle with one cupmark near the centre. Six cupmarks are assembled into another open circle with one cupmark near centre (see L. 3035).

L. 3044: A complex of seven shallow cupmarks cut into bedrock, located north-east of L. 3038. L. 3045: A composite installation of 39 concave conical, and a few shallow cupmarks, cut into unlevelled bedrock (3.5 x 2.5 m) 6 m south of L. 3043. This is a specific PPNA RCI (see discussion). Four additional cupmarks were cut into two rocks beside it.

L. 3051: A complex composed of a small concave basin and three cupmarks, cut into an elevated rock on the

Figure 11. L. 3042.21: Natufian narrow conical mortar, reused and cut into an EB oval deep basin.

Figure 10. L. 3041.20: Natufian narrow conical mortar halted by a hard stone at bottom (part of rock formation).

33

Stone Tools in the Ancient Near East and Egypt

Figure 15. L. 3042.1: deep concave EB basin, with reused Natufian narrow conical mortar.

Figure 12. L. 3941.22: EB deep concave basin.

south-west corner of the upper mound (an area of ca. 300 m2). L. 3052: Ten installations cut into four scattered elevated rocks, including four small basins (partially covered) and 6 cupmarks. L. 3053 (Fig. 17): A few single cupmarks cut into a rock surface, and a concentration of five cupmarks located 10 m southeast of L. 3052. L. 3054: A concentration of elevated rocks with 44 cupmarks: seven on a square rock; 19 (mainly shallow) cupmarks cut into a rock beside it; 10 shallow cupmarks located on a rock 3 m south of the latter; and eight cupmarks (six concave and two shallow) located 5 m further north. This composite installation is a specific PPNA RCI (see discussion).

Figure 13. L. 3942.24: Natufian small concave conical mortar.

Figure 14. L. 3042: General view of large bedrock with 31 cupmarks, and EB round deep basin, reused Natufian narrow conical mortar.

34

D. Eitam: Survey of Rock-Cut Installations at Tel Bareqet (Israel) L. 3058: A complex consisting of one small basin with five cupmarks beside it (2 of the cups are connected) cut into a small portion of bedrock, located about 20 m north of L. 3057. L. 3059: A large round basin (partly covered) cut into a small rock face with a few cupmarks cut into the rocks beside it. The basin may also be typologically dated to the EB. L. 3060: A complex of six cupmarks cut into a small section of slightly slanted bedrock, located 10 m south of the southern fortified wall. A limestone pounder was found beside the complex.

Figure 16. Plans of L. 3042.1: with EB round deep basin, reused narrow conical mortar, and small mortar eroded in bottom; L. 3063.1: oval deep EB basin, reused Natufian narrow conical mortar, 2, 3 and 5: small deep basins with cup at bottom.

L. 3062: A complex of 5 cupmarks cut into a small rock surface, situated 1.5 m west of the pebbled floor courtyard, in excavation area LS, 6 m east of L. 3056. L. 3063: This complex contains a medium oval basin (EBII) with a collection vat (reused Late Natufian narrow conical mortar), three small basins with cups, and one cupmark (Figs. 16, 18). Two shallow basins are cut farther north. These installations are cut into a wide stretch of bedrock, located ca.10 m south of L. 3042. In addition, nine scattered cupmarks cut into a long rock surface, 2 m north-east to L. 3062.

Figure 17. L. 3053: cupmarks cut on top of a rock.

L. 3064: A complex of nine shallow cupmarks cut into an elevated rock, located 1 m west of L. 3040.

L. 3055: A complex of five concave cupmarks cut into a small rock surface, and six shallow cupmarks cut into a flat rock beside them, located 5 m west of L. 3054.

L. 3065: A PPNA specific composite RCI of 12 cupmarks and possibly a small basin (partly covered), cut into the centre of a wide rock step, located 5 m east of L. 3041 (southern part of L. 3046 bedrock, see discussion).

L. 3056: A composite installation of a specific PPNA RCI (see discussion) composed of 21 shallow and concave cupmarks cut into a slanted surface of small and oblong rock, located 7 m west of excavation area LS.

L. 3066: 14 cupmarks cut into a small rock surface step, located east of excavation area SD. This includes 11 cut in a row on a narrow oblong rock; two on a round rock face; and one on a small rock.

L. 3057: Two sections of bedrock, one of which one contained four cupmarks and a small oval basin, and the other of which contained eight conical cupmarks. This complex is located on the east side of the mound about 20 m west of the EB fortification line.

L. 3067: A PPNA composite installation, a specific type of RCI (see discussion), of 24 cupmarks cut near each 35

Stone Tools in the Ancient Near East and Egypt

Figure 18. L. 3063.1-3: oval deep EB basin, reused Natufian narrow conical mortar, and 2 small deep basins with cup in bottom.

Figure 19. L. 3948: oval deep basin cut by a shallow small bowl (both possibly EB).

other, of which six are arranged in a circle with one cupmark in the centre, located 3 m south of L. 3049.

L. 3071: A complex of six concave and conical cupmarks, cut in two rows into an oblong and narrow rock step (measured 6 x 0.7 m).

L. 3068: A complex of two small shallow basins and three single cupmarks, cut into bedrock near the Roman quarry (see Fig. 2: western slope of the upper mound).

4. Discussion A variety of RCIs were found in 45 complexes located in a relatively small area of the upper northwest part of Tel Bareqet. The types of RCI found at Tel Bareqet are as follows (see Table 1):

L. 3069 (Fig. 19): A large concave basin cut near a smaller shallow bowl on a rock step, beside the collapsed Roman quarry. The basin may be dated to EB. L. 3070: A complex of six cupmarks cut into small rock face. A basalt pounder was found beside this complex.

1. 36

Cupmarks of different sections (shallow and flat, conical, concave).

D. Eitam: Survey of Rock-Cut Installations at Tel Bareqet (Israel)

Figure 20. L. 3063: EBII oval basin which reuses a Natufian narrow conical mortar as collection vat.

2.

Composite RCI including several concave cupmarks that form a circle, with one cupmark cut in the centre, on small flat bedrock (see table 1). 3. A dozen to fifty cupmarks, cut into levelled or unlevelled large bedrock. 4. Shallow round or oval small basins (with or without cups in their bottoms). 5. Round deep, small to medium, basins, with and without cups in their bottoms. 6. Large round concave basins. 7. Concave mortar. 8. Narrow conical mortar and wide conical mortar. 9. Reuse of narrow conical mortar as the collection cup for a large concave basin. 10. Wine press – a complex RCI.

the same complex (L. 3042), a hard stone (part of the rock formation, Eitam 2008, but see Nadel et al. 2009), that had been reduced by pecking and abrading, was left at the bottom of the small narrow conical mortar (Fig. 10). This is a common phenomenon in Epipaleolithic RCIs including the Natufian one (Eitam 2013). As described above, the identified Natufian narrow conical mortars are part of complexes of RCIs, and in some (for ex. L. 3042) the nearby cupmarks can also be regarded as Natufian although the possibility that the complex was a result of accumuled activity over several time periods cannot be rejected (see the following PPNA RCIs). The function of these Natufian installations as devices for processing wild cereals (barley) was recently tested in experimental study (Eitam et al. 2015). The results support the following reconstruction. In the wide conical mortar, the threshed spikelets were hummelled, (the hard base of the awns was removed to enable eating the grain in the form of groats or porridge). In the narrow conical mortar the hulled grains were de-husked, (the chaff peeled off, but see other suggestions, Hayden et al. 2013; Nadel and Lengyel 2009; Rosenberg 2008; Rosenberg and Nadel 2014). The grain is then milled into fine flour (wear marks of vertical and radial motions of a long pestle were observed at the narrow conical mortars, Eitam et al. 2015), eventually for making proto-bread – unleavened small thick loaves baked on ash and wooden coal.

Many of the RCIs can be dated based on their typological affinities to three periods: Late Epipaleolithic (and, in particular, Late Natufian, 13,500-11,700 calBP), Early Neolithic (PPNA) and EBII. Other RCIs, such as cupmarks (with no particular set or distinctive form) or small shallow basins, cannot be related to a particular function or period. The Late Epipaleolithic is represented by at least six RCIs. The narrow conical mortar (n=5; loci 3030, 3041, 3042, 3048) is exclusive to the Late Natufian (Eitam 2009, 2013; Rosenberg and Nadel 2014). At Tel Bareqet, the upper parts of four out of five Natufian narrow conical mortars were cut by later activity, possibly for secondary use during the EB II, in a way that allowed the new RCI to use the deep end of the narrow conical mortar as part of its function. A small wide conical mortar, familiar both from the Early and Late Natufian periods, was cut into bedrock close to a narrow conical mortar (Eitam et al. 2015; Rosenberg and Nadel 2011). At

The PPNA stone tools, unlike the varied Natufian ones, are homogenous in form, and primarily contain various combinations of cupmarks in different sizes. Clearly, cupmarks are in no way exclusive or typical only to the PPNA communities of the southern Levant. They served as preliminary food processors and small containers for different materials from a very early 37

Stone Tools in the Ancient Near East and Egypt

Number

Illustration

Typological date

Type of RCI

Suggested use

Numbers of RCI

1

Cupmarks

Varied usages e.g., grinding depression, container

Many: e.g. L. 3042

2

Circle of cupmarks

Unknown

4: Loci 3035, 3041, 4230, 3065

PPNA

3

Many cupmarks on bedrock

Cupmarks threshing floor

17

PPNA

4

Shallow round basins

Unknown

5

Round deep basins with collection cup

Unknown

10

PPNA

6

Large round concave basins

Unknown

5 or 8: Loci 3030, 3041, 3042, 3048, 3063; L. 3049, 3059, 3069

EB II

7

Concave mortar

Pounding-blending device

1

8

Narrow conical mortar

De-husking glumed cereals

5: loci 3030, 3041, 3042, 3048

Late Natufian

9

Small wide conical mortar

Hummeling wild barley

1: L. 3042

Late Natufian

10

Large concave basin with narrow deep cup

Simple oil installation

5: loci 3030, 3041, 3042, 3048

EB II

11

Wine press

Wine production

1: L. 2030

EB II

PPNA

Table 1. Types of rock-cut installations; average measurements in cm; abbreviations: NCM – narrow conical mortar; RCI – rock-cut installation.

time, and continued to be used in Byzantine and even later periods (Eitam 1996). Nonetheless, the PPNA stone tool assemblages – ground stones as well as RCIs – are characterized by the prevalence of cupmarks, which most likely served primarily as crushing-grinding devices (Eitam 2013).

(with 15 to 65 cupmarks), and 22 are smaller complexes with 3-14 cupmarks. The composite installation with numerous cupmarks was possibly invented in the end of the Natufian period (Huzuq Musa, Eitam n.d.; Rosenberg et al. 2010), and became common in PPNA sites like Zur Natan (Marder et al. 2008; Lechevallier and Ronen 1994) and Hatula (Eitam 2013). We suggest that this composite installation of dozens of cupmarks, cut into levelled and unlevelled large bedrock, was the PPNA threshing floor. The cupmarks in these composite RCIs at Tel Raqefet,

At Tel Bareqet, 39 complexes of cupmarks were cut into leveled and unlevelled bedrock exposures, and on the heads of elevated rocks. 17 of them are large complexes 38

D. Eitam: Survey of Rock-Cut Installations at Tel Bareqet (Israel) as well as in other sites, did not bear any use marks and were possibly used as small containers for the threshed spikelets and grains of wild cereals (but see, Grosman and Goren-Inbar 2016). A few small, round, shallow bases, with or without cups within the bottom, were cut among the many cupmarks (Figs. 7, 8, L. 3039) and possibly were part of the composite installations where grain was processed and prepared as food in the field. Other complexes of several cupmarks could also be typologically dated to the PPNA, in particular a composite installation containing about six concave cupmarks arranged in a circle, with a single cup positioned in its centre (L. 3035, 3041, 4230, 3065). This type of composite installation was cut into a small area of flat rock surface. A similar installation was found at Netiv Hagdud (Eitam 2013: Fig. 26:4).

with the collection vat was used in EB II as a simple olive oil press, based on its resemblance to similar, but smaller, Late Chalcolithic oil installations discovered in sites in the Samarian hills and the nearby Shephelah region (Eitam 1993, 2008). 5. Conclusions The study of Tel Bareqet RCIs provides new perspectives on the production of foodstuffs by the site inhabitants, and significantly enhances our knowledge of the course of occupation at the site in ancient times. Discovery of the narrow conical mortar – an RCI type that is exclusive to the Late Natufian period – sheds light on the Late Epipaleolithic occupation at Bareqet. Nevertheless, following the character of the finds, it seems that Tel Bareqet, during the Late Epipaleolithic, was a ‘satellite’ site connected to a larger settlement nearby. Such ‘satellite’ sites, with a few RCIs including narrow conical mortars, were discovered in Mount Carmel, the southern Jordan Valley and the Samaria fringes of the desert (Rosenberg and Nadel 2011; Eitam et al. 2015, respectively).

The Early Bronze Age II is best represented by the wine press (L. 2030). Wine presses similar in shape and size were found in EBII single-period fortified sites in the Samarian hills (Eitam 1987: Fig. 2). The wine press of Tel Bareqet is located on a large bedrock exposure at the northern part of the upper city, close to the fortification line. The winepress has a typical small, kidney-shaped pressing floor, ca. 3.20 m2, with a shallow side cut into the bedrock surface. The volume of the collection vat – a bell-shaped basin with a small sunken vat at the bottom – is ca. 5.6 cubic meters. An adjacent vat with a volume of approximately ca. 5.7 cubic meters was cut and surrounded by the continuation of the pressing floor’s curved side (L. 3022). It was possibly used as a second press for grape skins, to squeeze out the yeasts (concentrated on the skins) and the extra drops of juice. Several cupmarks were cut into a rock surface close to the winepress (L. 2023-3025). The one cut into the pressing floor was most likely part of the wine press itself.

More frequent RCIs, typical of the PPNA, are associated with the frequent lithic evidence and few other finds, suggesting the presence of settlement with a new organisation of agrarian production at Tel Bareqet. In light of the nature of the rock-cut installations and the excavation results, we may suggest that during the PPNA Tel Bareqet was a seasonal open site, occupied during the cereal harvest season, like the site of Zur Natan (Marder et al. 2008). These sites did not contain stone walls or mudbrick dwellings, but could have had ephemeral shelters. They are located where wild cereals grew extensively, and exposed rock surfaces enabled the carving of the agricultural installations needed for various stages of food processing. Thus, the cereals’ spikelets were collected, threshed, and possibly brought (after preliminary processing) to a nearby PPNA settlement.

Five large round basins were found on the western part of the upper EB II town (L. 3030, 3041, 3042, 3048, 3063). Loci 3049, 3059, 3069, all bearing similar features, may also be assigned to this period. The round basins (excluding 2 oval basins, Figs. 19, 21) are characterized by a bell-shaped section, in which their slightly concave bottoms, the style of carving, and their size suggest an EB attribution following the similarity in shape and size to bowl-shaped basins cut into the bedrock exposure of a MB II site at the eastern Samarian hills (Eitam, 1996: Fig. 5). The purpose of these later basins is unknown, but seems to be related to food preparation.

Last, the presence of only one wine press and the very few simple oil installations possibly dated to the EB II may point towards the minor place held by wine and olive oil in the Early Bronze II diet, as opposed to bread and other cereal dishes. The location of the few oil, and the single wine, devices in the upper city (a phenomenon repeated in other EBII fortified cities in the Central Highland of Israel, see Eitam 1987: 20, note 13; Fig. 2) may suggest that these luxury foodstuffs were served only to the elite living in the upper part of the city.

Significantly, four of the large basins of Tel Bareqet have deep narrow collection vats cut in their bottoms. The typical shape of the very narrow conical vat (unsuitable for collecting liquid) and the use marks on its walls clearly revealed that the narrow Natufian vat used as a mortar was reused as a collection vat (Figs. 9:21, 16:1 upper, 1 lower). We suggest that the basin

Here again the incorporation of landscape archaeology – the survey of installations cut into bedrock exposures on a site’s surface – and the results of archaeological excavations provide a significant insight into the life of 39

Stone Tools in the Ancient Near East and Egypt the inhabitants of the site throughout its history, adding a previously unknown occupation in Bareqet in the Late Natufian, and beyond in ancient times, and pointing toward new forms of settlement patterns during these different periods, and different trajectories in the development of agricultural techniques in the protohistorical era.

Eitam, D. 2008. ‘Plant Food in the Late Natufian: the Oblong Conical Mortar as a Case Study’. Journal of the Israel Prehistoric Society 38: 133-151. Eitam, D. 2009. ‘Cereal in the Ghassulian Culture in Central Israel: Grinding Installation as a Case Study’. Israel Exploration Journal 59/1: 63-79. Eitam, D. 2010. ‘Late Epipalaeolithic Rock-Cut Installations and Groundstone Tools in the Southern Levant–Methodology and Classification System’. Paléorient 35: 77-104. Eitam, D. 2013. Archaeo-Industry of the Natufian Culture: Late Epipalaeolithic Rock-Cut Installations and Groundstone Tools in the Southern Levant. PhD dissertation, Hebrew University of Jerusalem (Hebrew with English summary). Eitam, D. n.d. ‘New aspects of the Natufian Culture: Huzuq Musa, an Exceptional site in the Jordan Valley’. PPN8, Nicosia. Eitam, D., M., Kislev, A. Karty, O. Bar-Yosef 2015. ‘Experimental barley flour production in 12,500-year-old rock-cut mortars in Southwestern Asia’. PLoS ONE 10(7): e0133306. Eitam, D. and Squitieri, A. n.d. ‘Methodology and Classification system of Iron Age Ground Stone Tools and Rock-cut Installations in Israel’. Frankel, R. 1999. ‘Wine and Oil Production in Antiquity in Israel and other Mediterranean Countries’. JSOT/ ASOR Monograph Series 10, Sheffield Academic Press, Sheffield. Gopher, A. 1997. ‘Ground Stone Tools and Other Stone Objects from Netiv Hagdud’, in Bar-Yosef, O., and Gopher, A. (eds). An Early Neolithic Village in the Jordan Valley, Part I: The Archaeology of Netiv Hagdud. Harvard University Press, Cambridge, MA: 151176. Grosman, L., and N. A. Goren-Inbar 2016. ‘Landscape Alteration by Pre-Pottery Neolithic Communities in the Southern Levant – The Kaizer Hilltop Quarry, Israel’. PLoS ONE 11(3): e0150395. Hayden, B., N. Canuel, J. Shanse 2013. ‘What was Brewing in the Natufian? an Archaeological Assessment of Brewing Technology in the Epipaleolithic’. Journal of Archaeological Method Theory 20: 102-150. Lechevallier, M., and A. Ronen 1994. Le site de Hatoula en Judée Occidentale, Israel. Mémoires et Travaux du Centre de Recherche Français de Jérusalem 8. Association Paléorient, Paris. Marder, O., A. N. Goring-Morris, H. Khalaily, I. Milevski, R. Rabinovich, and V. Zbenovich 2008. ‘Tzur Natan, a Pre-Pottery Neolithic A Site in Central Israel and Observation on Regional Settlement Patterns. Paléorient 33(2): 79-100. Miller, V, S. Filin, D. Rosenberg, and D. Nadel 2014. ‘3D Characterization of Bedrock Features: A Natufian Case Study’, Near Eastern Archaeology 77, 3, Special Issue: Cyber-Archaeology: 214-218. Nadel, D., G. Lengyel 2009. ‘Human-Made Bedrock Holes (Mortars and Cupmarks) as a Late Natufian

Acknowledgments Excavations at Tel Bareqet were conducted under the Institute of Archaeology, University of Tel Aviv. R. Gophna served as an academic advisor on behalf of the Institute of Archaeology at Tel Aviv University in the 2004 season. Thanks go to Ram Gophna for his invitation to conduct the RCI survey in Tel Bareqet, and to Isaac Paz and Sapir Paz for supporting the RCI survey providing the volunteers. The latest also provided the data on the excavations, and the plans of the site and the RCI survey (figures 1 and 2). Thanks go to Ron Shimelmitz for providing the results of his investigation on the lithics and of his constractive criticism and comments, reading the first draft of this paper. Bibliography Amit, D., Torgë, H., Gendelman, P. 2008. ‘Horvat Burnat, a Jewish Village in the Lod Shephelah during the Hellenistic and Roman Periods’. Qadmoniot 136: 96108 (Hebrew). Buonasera, T. 2005. ‘Fatty acid analysis of prehistoric burned rocks: a case study from central CalifornIa’. Journal of Archaeological Science 32: 957-965. Dubreuil, L. 2004. ‘Long-Term Trends in Natufian Subsistence: A Use-Wear Analysis of Ground Stone Tools’. Journal of Archaeological Science 31(11): 16131629. Elliott, C. 1983. ‘The Ground Stone Industry, An Outline of Ground Stone Industry’. Levant 15: 11-37. Eitam, D. 1993. ‘“Between the (olives) rows, oil will be produced (wine?) presses will be trod…” (Job 24,11)’, in M.-C. Amouretti and Brun (eds), Oil and Wine Production in the Mediterranean Area. Bullettin de Correspondance Hellenique, Supplement 26 Athens, Paris: 65-90. Eitam, D. 1987. ‘Olive Oil Production during the Biblical Period’. in M. Heltzer and D. Eitam (eds), Olive Oil in Antiquity, Israel and Neighbouring countries, Haifa University, Haifa: 16-36. Eitam, D. 1996. ‘Survey of Agricultural Installations’, in A. Zertal (Ed.), The Menasseh Hill Country Survey, The Eastern Valleys and the Fringes of the Desert Vol. II. University of Haifa, Tel Aviv: 681-738 (Hebrew). Eitam, D. 2005. ‘The Installations and the Ground Stone Tools’, in A. Zertal (eds), The Manasseh Hill Country survey, from Nahal Bezek to the Sartaba Vol. IV. Univwrsity of Haifa, Haifa: 649-723 (Hebrew). 40

D. Eitam: Survey of Rock-Cut Installations at Tel Bareqet (Israel) Social Phenomenon’. Archaeology, Ethnology and Anthropology of Eurasia 37(2): 37-48. Nadel, D., D. Rosenberg, and R. Yeshurun 2009. ‘The Deep and the Shallow: the Role of Natufian Bedrock Features at Rosh Zin, Israel. Bulletin of the American Schools of Oriental Research 355: 1-29. Paz, Y., and S. Paz 2007. ‘Tel Bareket – Excavations in a Fortified City of the Early Bronze Age II in the Central Coastal Plain’. Qadmoniot 134: 82-89 (Hebrew). Peterson, J. D. 1999. ‘Early Epipaleolithic Settlement Patterns: Insights from the Study of Ground Stone Tools from the Southern Levant’. Levant 31: 1-17. Power, R. C., A. M. Rosen, and D. Nadel 2014. ‘The Economic and Ritual Utilization of Plants at the Raqefet Cave Natufian Site: the Evidence from Phytoliths’. Journal of Anthropological Archaeology 33: 49-65. Rosenberg, D. 2008. ‘The Possible Use of Acorns in Past Economies of the Southern Levant: a Staple Food or a Negligible Food Source?’ Levant 40(2): 167-175. Rosenberg, D. and A. Gopher 2010. ‘Food Processing Tools and other Groundstone Implements from Gilgal I and Gilgal III’, in O. Bar-Yosef, M. A. Goring-

Morris, and A. Gopher, (eds). Gilgal: Early Neolithic Occupations in the Lower Jordan Valley: The Excavations of Tamar Noy. American School of Prehistoric Research Monograph 4. Brill, Winona Lake: 139-176. Rosenberg, D., R. Yeshurun, I. Groman-Yaroslavski, H. Winter, A. Zertal, R. Brown-Goodman and D. Nadel 2010. Huzuq Musa: a preliminary report on the test excavation at a Final Epipaleolithic/PPNA site in the Jordan Valley. Paléorient 36 (2):189-204. Rosenberg D., and D. Nadel 2011. ‘Natufian Nahal Oren and its Satellite Sites: some Regional and Ceremonial Aspects’. Before Farming 3: 1-16. Rosenberg, D. and D. Nadel 2014. The Sounds of Pounding: Boulder Mortars and their Significance to Natufian Burial Customs. Current Anthropology 55(6): 784-812. Torgë, H. 2012. ‘Settlement Remains from the Late Hellenistic and Early Roman Periods at Khirbat Burnat (Southwest)’. ‘Atiqot 69, 1*-68* (Hebrew with English summary): 157-159. Wright, K. 1992. ‘A Classification System for Ground Stone Tools from Prehistoric Levant’. Paléorient 18(2): 53-81.

41

Stone Tools in the Ancient Near East and Egypt

Appendix A. Rock-cut installations at Tel Bareqet. Type, subtype

Upper/ lower diameter/ depth*

Shape; section

Face

Locus location Preservation

Note

1*

winepress

floor 3.185 sqm; vat vol.: ca. 5.6 cubic meters

kidney shape floor; bell shape vat; sec vat: irregular

carve on floor; flak & coarse pecked in vats

L. 3020; N-W good; sec vat corner near city eroded wall

EBII; see detailed description

2*

round, deep basin

19.5/9/15

shape; concave

carve on side& end; no peck or abrade

L. 3030; on a small rock surface; near no 1; by no. 3

upper part eroded

different shape – possibly late

3+

shallow cupmark

11/2

flat end

rough

L. 3030; 80 cm of no 2

eroded

4-12

round-conical cupmarks

10-9/3/6-5

round-conical shape

fine ground

L. 3032; N of tumulus on small bedrock

good

7 cups arranged in circle

11-17

shallow-concave cupmarks

6/4/2

fine ground

L. 3032; N of tumulus

good

2 cupmarks under tumulus;

18-19

shallow cupmark

13-10/11-9/2

rough

L. 3032; N of tumulus on small bedrock

good

16 cups on bedrock L. 3032

20-24

round-conical cupmarks

11-10/7-5/2.5

L. 3033; part of bedrock 3032; W of tumulus

25 26

shallow-concave cupmarks

9-8.5/7/2

L. 3033; part of bedrock 3032; W of tumulus

27

shallow-concave cupmark

7/6/1

28

concave cupmark

10/5/4

29

shallow cupmark

12/10/3

30

small, shallow basin & cup

16-15/1514/3 7/4/1

round; flat bottom

L. 3034; 5 m NE sides: eroded to L. 3033; 5 m bell shape of no 30

31*

small, shallow basin & cup

25-/ 22 (22)/5 12/7/2.5

round; flat bottom

L. 3034; 5 m NE sides: eroded to L. 3033; 5 m bell shape of no 29

32-39

shallow cupmarks

diameter: 1211-10; depth: 4-3-2.5

flat bottom

40

large deep basin & cup

25/(17)/ 17

cup: 5 /(9) on flat bottom

41-42

shallow cupmarks

diameter: 11, 8; depth: 3, 1

flat bottom

43

concave cupmark

11/8/5

No

flat end

bedrock goes under tumulus good; pretumulus 7 cups on bedrock L. 3033

on rocks E of tumulus L. 3032 – pre tumulus

6 cups arranged in circle

L. 3035; on flat rock

sides eroded

L. 3036; 6 m S of very eroded L3034

25 cupmarks around tumulus 2 shallow basins

7 cupmarks basin & 3 cupmarks

L. 3036; 6 m S of L3034 L. 3036; 6 m S of L3034

Measurements in cm; (4) estimated dimensions of complete tool; ;140 *28 – number and figure number; M – mortar; m. - meter; L –length; Dep- depth; Diam –diameter; N – north; S – south; W – west; E – East; GS – Ground stone; NCM – narrow conical mortar; LN – Late Natufian.

42

D. Eitam: Survey of Rock-Cut Installations at Tel Bareqet (Israel)

No

Type, subtype

Upper/ lower diameter/ depth*

diameter: 14, shallow 44-48 13, 11; cupmarks & cup depth: 3,4.

Shape; section

Face

Locus location Preservation

Note

flat end & cup; diameter: 4.5/6/3.5; depth: 1.5/2.5/0.5

L3037

4 cupmarks & cup basin & 3 cupmarks

49*

small basin & cup

20-26/(20-26) /108/4

oval; flat bottom

L. 3038; oblong and narrow rock

50-52

3 cupmarks: 1 concave & 2 with cup

diam: 12, 14, 13; depth: 4, 5, 4

flat & cup: 5.5/2;4/1

L 3038

53 *1

small basin & cup

30/(30/8 11/5

round; flat end round; cone

smooth smooth

L. 3039; beside no. 54, 55

good

54

shallow cupmark

6/1

slightly concave

smooth

L. 3039

good

55

conical cupmark

6/3

cone

smooth

L. 3039

good

56

shallow cupmark

7/1

slightly concave

smooth

L. 3039

good

57

shallow cupmark

6/1

slightly concave

smooth

L. 3039

good

58

conical cupmark

11/6/4

cone

rough

L. 3039

eroded

59

concave cupmark

12/5/5

concave

smooth

L. 3039

good

60 *2

medium basin & cup

32-28/32-28/ 4-5 10/10/6

flat end round; cone

smooth smooth

L. 3039; beside no 59, 61

eroded sides

61

shallow cupmark

11-10/7-8 /4-2

flat

smooth

L. 3039

good

62

shallow cupmark

9/7-8/0.5-2

flat

smooth

L. 3039

good

63

shallow cupmark

/8/1

flat

smooth

L. 3039

good

64

shallow cupmark

10/10/1

flat

smooth

L. 3039

good

65

concave cupmark & cup

10-11/4/1 dep: 5

smooth rough

L. 3039

66

concave cupmark & cup

10/3 3-3.5/0.5

concave

smooth rough

L. 3039

67

concave cupmark & cup

12/4 4/1.5

concave

smooth rough

L. 3039

68

concave cupmark

9/5/5

concave

smooth

L. 3039

good

69

shallow cupmark

9-8/2.5

flat bottom

smooth

L. 3039

good

70

shallow cupmark

7/1

flat bottom

smooth

L. 3039

good

71

shallow cupmark & cup

10-11/2 5/1

flat bottom

smooth rough

L. 3039

72

shallow cupmark & cup

11/2.5 4-3.5/1

flat bottom

smooth rough

L. 3039

73

shallow cupmark & cup

12/2 5-4.5/1

flat bottom

smooth rough

L. 3039

74

shallow cupmark & cup

10.5-11/2 5-4.5/0.5

flat bottom

smooth rough

L. 3039

43

on the north edge of rock

Stone Tools in the Ancient Near East and Egypt

No

Type, subtype

Upper/ lower diameter/ depth*

75

shallow cupmark & cup

10-11/2 5-4.5/1

flat bottom

smooth rough

L. 3039

76

shallow cupmark & cup

11/3 5/1

flat bottom

smooth rough

L. 3039

77

shallow cupmark & cup

10-11/3 5-4.5/0.5

flat bottom

smooth rough

L. 3039

78

shallow cupmark & cup

12/3 5-6/2

flat bottom

smooth rough

L. 3039

79

shallow cupmark & cup

11/1 6.5/3

flat bottom cone

smooth rough

L. 3039

80

shallow cupmark & cup

10/2 5-4/1

flat bottom

smooth rough

L. 3039

81

shallow cupmark & cup

9/1.5 4/0.5

flat bottom cone

smooth rough

L. 3039

82

shallow cupmark & cup

11/1.5 5.5/2.5

flat bottom cone

smooth rough

L. 3039

83

shallow cupmark & cup

11/3 4-3.5/1

flat bottom

smooth rough

L. 3039

84

shallow cupmark & cup

11-12/3-4 4/0.5

flat bottom shallow

smooth rough

L. 3039

85

shallow cupmark & cup

10/2 4/0.5

flat bottom shallow

smooth rough

L. 3039

86

shallow cupmark & cup

11/1 6/2

flat bottom cone

smooth rough

L. 3039

87

shallow cupmark

12/1

flat bottom

rough

L. 3039

88

shallow cupmark

9/2

flat bottom

89

shallow cupmark

8/7/1

flat bottom

rough

L. 3039

90

shallow cupmark & cup

16/2 8/6

flat bottom cone

smooth rough

L. 3039

91

shallow cupmark & cup

10/1 5/2

flat bottom

smooth rough

L. 3039

92

concave cupmark

8.5/4

concave

smooth

L. 3039

94

shallow cupmark & cup

12/2 7-6/5

flat bottom

rough

L. 3039

95

concave cupmark

7/3

concave

smooth

L. 3039

96

shallow cupmark & cup

9/0.5 6/1

flat bottom

smooth rough

L. 3039

97

shallow cupmark & cup

9-10/1 7.5/5

flat bottom

smooth rough

L. 3039

98

shallow cupmark & cup

10-9/2 7-6/3

cone

smooth rough

L. 3039

99

shallow cupmark

6-7/2

concave

smooth

L. 3039

100

concave cupmark

7/2

concave

smooth

L. 3039

101

concave cupmark

6-7/4

concave

smooth

L. 3039

Shape; section

Face

Locus location Preservation

on edge of rock

L. 3039

44

Note

good

D. Eitam: Survey of Rock-Cut Installations at Tel Bareqet (Israel)

No

Type, subtype

Upper/ lower diameter/ depth*

102

concave cupmark

7/2

cone

smooth

L. 3039

103

shallow cupmark

6-8/1

flat bottom

rough

L. 3039

104

concave cupmark

10/4/6

concave

smooth

L. 3039

105

concave cupmark

8/4/5

concave

smooth

L. 3039

106

concave cupmark

10/4/7

concave

smooth

L. 3039

107

shallow cupmark

5/0.5

flat bottom

rough

L. 3039

108

concave cupmark

7.6/2

concave

smooth rough

L. 3039

109

concave cupmark

10/3/6

smooth

L. 3039

110

concave cupmark

10/3/6

smooth

L. 3039

111

concave cupmark

10/3/7

smooth

L. 3039

112

shallow cupmark

8/1

flat bottom

rough

L. 3039

113

concave cupmark

9/4/4.5

concave

smooth

L. 3039

114

medium basin & cup

29-30/ (27-29)/16 12/5/5

round; flat end straight sides; slanted end

pecked end

L. 3040

115 *1

concave mortar

17/10/4

116 *2

conical cupmark

12/11/6 4-3.5/1

cone

smooth rough

L. 3041; beside no 118

117 *3

round conical cupmark

9.5/-11/8

round; cone

smooth

L. 3041

120 *6

conical cupmark

10/6

cone

smooth

L. 3041; beside no 121

5 inst in a row

121 *7

conical cupmark

14-12/8

cone

smooth

L. 3041

5 inst in a row

122 *8

conical cupmark

10-13/?

cone?

smooth

L. 3041

123 *9

concave cupmark

12/10/4

concave

smooth

L. 3041

124 *10

concave cupmark

13-11/5

concave

smooth

L. 3041; beside no 121

125 *11

shallow cupmark

9-7/2

slightly concave

smooth

L. 3041; beside no 123

126 *12

conical cupmark

9.5/7

concave

smooth

L. 3041

eroded on 1 side

127 *13

conical cupmark

11-9/7

concave

smooth

L. 3041; beside no 128

eroded rim

128 *14

conical cupmark

7/4

cone

smooth

L. 3041

Shape; section

Face

Locus location Preservation

oval; curved side& end: coarse L. 3041; on edge sides & concave peck of rock bottom

45

eroded sides

Note

on south part of rock of L. 3041

eroded bottom

5 RCI in a row

5 inst in a row

Stone Tools in the Ancient Near East and Egypt

No

Type, subtype

Upper/ lower diameter/ depth*

129 *15

conical cupmark

12/4

cone

smooth

L. 3041; in a row with no 120/1/4/6

130 *16

cupmrk

11-10/6

cone

smooth

L. 3041

131 *17

concave cupmark

11/7

round; cone

smooth

L. 3041

132 *18

conical cupmark

10-7/8

concave

L. 3041; beside no 132

133 *19

cupmark

14-10/5

slightly concave

L. 3041

eroded sides

134 *20

ocm

15-17/3/23

narrow cone

2/3 sides: coarse ground; lower part: fine ground

L. 3041

eroded near end & rim

narrow cone oval & deep concave

fine ground & coarse ground coarse crave

L. 3041; beside OCM 134

Shape; section

Face

Locus location Preservation

135 *21

ocm with 14(23/4/45(70) concave basin 41-29/17/23

136 *22

concave basin

48/23/22

round; shallow concave end

up: crave & coarse ground; bottom: ground

L. 3041; beside basin 135

137 *23

shallow cupmark

8/1

slightly concave

smooth

L. 3041; beside no 130

138 *24

shallow cupmark

14/11/3

roun

L. 3041; near no 136

139 *25

conical (?) cupmark

8-6/6?

cone

L. 3041; beside no 140

140 *29

conical (?) cupmark

7.5-8/6?

cone

L. 3041; beside no 141

142 *26

conical cupmark

8/6

cone

smooth

L. 3041; beside no 143

143 *27

conical cupmark

10/6

cone

smooth

L. 3041; beside no 144

144 *28

cupmark

10/6

cone

smooth

L. 3041

145 *1

large basin with ocm

50-60/48/22 14-16(18)/2/

striaght side & end narrow cone

crave & coarse ground fine ground

L. 3042; beside no 146, 147

146 *2

shallow cupmark

13-14/3

slightly cone

smooth

L. 3042; beside no 145

147

shallow cupmark

10/4

flat bottom with small cup

148153

9 conical cupmarks

diam: 14-9 dep: 7-6

cone

154155

2 shallow cupmarks

11-9/3

155169

13 conical cupmarks

170173

Note 5 inst in a row

N side eroded

some reused OCM erosion on m as cups sides up sides eroded

invasive erosion

debatable 3 cupmarks

eroded bell shape basin

reused OCM 2 cups near OCM

L. 3042; beside no 145

eroded sides & bottom?

smooth

L. 3042; on W edge of bedrock

good

1 cup cut another 1

sightly con

smooth

L. 3042; on W edge of bedrock

good

with no 148153

11-9/6-7

cone

smooth

L. 3042; on N side of bedrock

good

7 in circle, 1 in middle

3 shallow cupmarks

9-10/4

slightly concave

smooth

L. 3042; beside no 155-169

good; 1 cup eroded

with cup 155-169

174180

6 concave cupmarks

13-8/5-7

concave

4 cupmarks smooth

L. 3042; on centre of rock

4 eroded, 3 good

181

concave cupmark

11/8/2

concave

smooth

L. 3042; on centre of rock

46

with 174 -180

D. Eitam: Survey of Rock-Cut Installations at Tel Bareqet (Israel)

No

Type, subtype

Upper/ lower diameter/ depth*

182 *11

shallow cupmark

12/2

slightly concave

183

shallow cupmark

7/2

slightly concave

L. 3043; on narrow bedrock

group of 13 cups

184

shallow cupmark& cup

9/2.5 5/2

flat

L. 3043; on narrow oblong bedrock

group of 13 cups

185

shallow cupmark

10/6/3

flat

L. 3043; on narrow bedrock

good

group of 13 cups

186

concave cupmark

11-12/7-7/7

round

L. 3043; on narrow bedrock

eroded

group of 13 cups

187

shallow cupmark

7/6/1

flat

L. 3043; on narrow bedrock

group of 13 cups

188

shallow cupmark

5/4/1

flat

L. 3043; on narrow bedrock

group of 13 cups

189

shallow cupmark

7/5/2

flat

L. 3043; on narrow bedrock

group of 13 cups

190

shallow cupmark

7-8/1

slightly concave

L. 3043; on narrow bedrock

group of 13 cups

191

concave cupmark

10-11/5/6

round

L. 3043; on narrow bedrock

group of 13 cups

192

shallow cupmark

7/1

slightly con

L. 3043; on narrow bedrock

group of 13 cups

193

conical cupmark & cup

14/9

cone with cup in end

L. 3043; on narrow oblong bedrock

group of 13 cups

194

shallow cupmark

14/3

slightly concave

L. 3043; on W edge of bedrock

group of 13 cups

195

concave cupmark

12/5/7

round

L. 3043; on narrow bedrock

group of 13 cups

196198

3 cupmarks

L. 3043; on elevated rock

199201

3 cupmarks

L. 3043; on bedrock

202204

3 conical cupmarks

L. 3043; on bedrock

205206

2 small cupmarks

L. 3043; on elevated rock

207213

7 shallow cupmarks

214252

39 cupmarks

253

concave cupmark

254256

3 shallow cupmarks

257258

2 small shallow basins

259262

4 cupmarks

Shape; section

Face

Locus location Preservation

rough

L. 3042; on E side of bedrock

11-14/1.5-4

rough most cups concave, or cone, few flat

10/7/10

eroded

Note debatable; isolated

L. 3044; on bedrock

NE to L. 3034

L. 3045; on square bedrock

ES to L. 3034

smooth

L. 3045; on elevated rock

rough

L. 3045; on elevated rock

E to no 244

flat

L 3046; beside no 259-262

on low oblong rock step

flat

L. 3046; beside 2 basins

on low oblong rock step

round

47

Stone Tools in the Ancient Near East and Egypt

Upper/ lower diameter/ depth*

Shape; section

Face

Locus location Preservation

No

Type, subtype

272275

4 cupmarks

conical & concave

L 3047; on 1 elevated rock

near 2nd rock

276279

5 cupmarks

conical & concave

L. 3047; on 2nd elevated rock

near 3rd rock

276279

1 cupmark

concave

L. 3047; on 3rd elevated rock

beside 4th rock

280281

2 cupmarks

conical

L. 3047; on 4th elevated rock

282 *1

large basin & coll cup

45-50/ (55)/20 17/8/6

slanted botton; wide cone

carve & coarse ground coarse ground

L. 3048; attached no 283

eroded sites; 1 side broken

283 *10

small basin & cupmark

28-30/10 5/3

straight side & end; cone

coarse ground carve

L. 3048; attached no 282

eroded sites; contemporary 1 side to no 282 broken

284/5, *2, 3

2 shallow cupmarks

10-12/3 9/4

flat bottom; slightly concave

rough; find ground

L. 3048; 1 beside other 2

286 *4

small basin & cup

17/3 6/4

flat bottom; cone

coarse ground carve

L. 3048; beside no 285

eroded on 1 side

287 *7

small basin & cup

17-(15)/4 5/4

concave bottom/ cone

coarse ground; coarse ground& carve

L. 3048; beside no 285

eroded on 1 side

288 *8

small basin & cup

17/3 6/4

concave bottom/ cone

coarse ground carve

L. 3048; beside no 285

eroded on 1 side

289 *5

ocm widen into basin

(21)/3/48; 26-24/17/13

conical m; curve sides

carve & fine ground coarse crave

L. 3048; beside no 290

good/ eroded

290 *6

wide conical mortar

21-23/4/17

conical m; small cilindrical shaft in bottom

carve on sides; coarse ground on shaft

L. 3048; beside no 289

eroded sides

291*

large basin & shallow cup

50-51/ (50)/22 17/2

stright side & end; slightly fine ground

side: carve; end: coarse ground fine ground

292309

18 cupmarks

310*

small con basin

311312

2 cupmarks

313314

2 small shallow basins

315317

3 cupmarks

L. 3052; on elevated rock

318320

3 small cupmarks

L. 3052; on bedrock

321

small shallow basin

L. 3052; on elevated rock; beside no 322

E of bedrock 318-320

322

cupmark

L. 3052; beside no 321

E of bedrock 318-320

323

small shallow basin

L. 3052; on elevated bedrock

3 m N to no 323

L. 3049; on S eroded sites; edge of L 3048; 1 side attached no 283 broken L. 3050; 3 m W to L 3049

15-17/13/5

concave

fine ground

L. 3051; on low small rock L. 3051; on elevated rocks

Note

7 inst on oblong bedrock

hard stone in mortar

7 inst on oblong bedrock 2 circles of 5 & 6 cups

good

SW corner of the Tell area with few RCI

L. 3052; on small bedrock

flat bottom

48

D. Eitam: Survey of Rock-Cut Installations at Tel Bareqet (Israel)

No

Type, subtype

324341

18 shallow cupmarks

342347

4 concave cupmarks

348357

10 shallow cupmarks

376379

Upper/ lower diameter/ depth*

Shape; section

Face

Locus location Preservation

Note

L. 3053; on close elevated rocks

with 4 concave cupmarks

L. 3053; on close elevated rocks

with 19 cupmarks

slightly concave/ flat

L. 3053; on 3 elevated rocks

5 m N the 22 cupmarks

1 cupmark

concave

L. 3047; on 3rd elevated rock

beside 4th rock

380381

2 cupmarks

conical

L. 3047; on 4th elevated rock

382 *1

large basin & coll cup

45-50/ (55)/20 17/8/6

slanted bottom; wide cone

carved & coarse ground coarse ground

L. 3048; attached no 283

eroded sites; 1 side broken

383 *10

small basin & cupmark

28-30/10 5/3

stright side & end; cone

coarse ground; carved

L. 3048; attached no 282

eroded sites; contemporary 1 side to no 382 broken

384/5, *2, 3

2 shallow cupmarks

10-12/3 9/4

flat bottom; slightly con

rough; fine ground

L. 3048; 1 beside other 2

386 *4

small basin & cup

17/3 6/4

flat bottom; cone

coarse ground; carved

L. 3048; beside no 285

eroded on 1 side

387 *7

small basin & cup

17-(15)/4 5/4

concave bottom/ cone

coarse ground; coarse ground & carved

L. 3048; beside no 285

eroded on 1 side

388 *8

small basin & cup

17/3 6/4

concave bottom/ cone

coarse ground; carved

L. 3048; beside no 285

eroded on 1 side

389 *5

ocm widen into basin

(21)/3/48; 26-24/17/13

conical m; carving sides

carved & fine ground; coarse ground

L. 3048; beside no 290

good/ eroded

390 *6

wide conical mortar

21-23/4/17

conical m; small cilindrical shaft in bottom

carved on sides; coarse ground on shaft

L. 3048; beside no 289

eroded sides

391*

large basin & shallow cup

50-51/ (50)/22 17/2

stright side & end; slightly fine ground

392409

18 cupmarks

410*

small con basin

411412

2 cupmarks

413414

2 small shallow basins

415417

3 cupmarks

L. 3052; on elevated rock

418420

3 small cupmarks

L. 3052; on bedrock

421

small shallow basin

L. 3052; on elevated rock; beside no 322

E of bedrock 418-420

422

cupmark

L. 3052; beside no 321

E of bedrock 418-420

slightly concave

side: carved; end: L. 3049; on S eroded sites; coarse ground edge of L 3048; 1 side fine ground attached no 283 broken L. 3050; 3 m W to L. 3049

15-17/13/5

concave

fine ground

L. 3051; on low small rock L. 3051; on elevated rocks

7 inst on oblong bedrock

hard stone in mortar’s bottom

7 inst on oblong bedrock 2 circles of 5 & 6 cups

good

SW corner of the Tell area with few RCI

L. 3052; on small bedrock

flat bottom

49

Stone Tools in the Ancient Near East and Egypt

Upper/ lower diameter/ depth*

Shape; section

Face

Locus location Preservation

No

Type, subtype

423

small shallow basin

L. 3052; on elevated bedrock

3 m N to no 423

424428

5 single cupmarks

L. 3053; on few bedrocks

10 m EN to L. 3052

428442

group of 5 cupmarks

L. 3053; on bedrock

10 m EN to L 3052

443449

7 cupmarks

L. 3054; on square rock

to N & beside no 450-473

450468

18 shallow cupmarks

L. 3054; on close elevated rocks

with 4 Con cupmarks

469473

4 concave cupmarks

L. 3054; on close elevated rocks

with 19 shallow cupmarks

474484

9 shallow cupmarks

L. 3054; on elevated bedrock

5 m N the above rock

485

conical cupmark

486488

2 shallow cupmarks

489495

6 conical cupmarks

496501

5 deep cupmarks

502508

slightly concave

slightly con, flat

Note

L. 3054; on elevated bedrock L. 3054; on bedrock

5 m N the above rock

L. 3054; on small bedrock

5 m W to L. 3054

cone or concave (partly covered)

L. 3055; on brdrock

with no 486488

6 shallow cupmarks

flat bottom

L. 3055; on bedrock

509530

21 cupmark

concave & shallow

L. 3056; on small long slanted bedrock

7 m W of excavated courtyard

530534

4 cupmarks

L. 3057; on bedrock

accumulated

535

small shallow basin

L. 3057; beside no 530-534

536544

8 conical cupmarks

L. 3057; on square bedrock

beside no 530-535

545

small concave basin

L. 3058; on small bedrock beside 546-551

20 m N to L. 3057

546551

5 small cupmarks

2 cups connected

L. 3058; beside no 545

20 m N to L. 3057

552

large round basin

(partly covered)

L. 3059; on bedrock

553557

4 scattered cupmarks

558564

concentration of 6 cupmarks

567

barrel-shape basin ?

568571

4 concave cupmarks

slightly concave

8-6/4-7; 8-6/2-3

eroded

2.5 m to above rock

L. 3059; on small rocks

around no 552

3 concave/ 3 shallow

L. 3060; on small slanted rock

10 m S of excavated area

round; (partly covered)

L. 3061; beside Roman winepress

on N slope

L. 3062; on small bedrock

1.5 m W to excavated pebble floor

6-9/4-6

50

D. Eitam: Survey of Rock-Cut Installations at Tel Bareqet (Israel)

No

Type, subtype

Upper/ lower diameter/ depth*

572573

2 shallow cupmarks

5-7/1.5

slightly concave

574 *1

large basin with ocm

40-30/ (37)/13 (17)/2/(30)

oval; slanted end; narrow cone

575 *2

small basin with cup

15-17/9/8 6/3

576 *3

small basin with cup

577 *4

sallow cupmark

578579

2 small shallow basins

580*

small basin & cup

580588

9 shallow cupmarks

L. 3064; on elevated rock

589601

12 shallow cupmarks

L. 3065; on wide rock step

4 m E to L. 3041

602

small shallow basin

L. 3065; on the above rock step

4 m E to L. 3041

603613

11 cupmarks

L. 3066; on small long narrow rock

in a row, 4 m E to excavated area

614615

2 cupmarks

L. 3066; on round bedrock

in a, 4 m E to excavated area

616

cupmark

L. 3066; on small low bedrock

in a row, 4 m E to excavated area

617640

25 cupmarks

L. 3067; on bedrock

641642

2 small shallow basins

643645

3 cupmarks

646*

large concave basin

647642

6 cupmarks

643647

6 conical cupmarks

Shape; section

Face

Locus location Preservation

Note

L. 3062; on the above bedrock

1.5 m W to excavated pebbled floor

coarse ground fine ground

L. 3063; on wide bedrock, beside no 575

bell shape by 10 m S to erosion/ L. 3042, 7 m good W to quarry

round & deep; concave

carve& ground fine ground

L. 3063; beside no 574

side eroded

10 m S to L. 3042

18/15/7 5/4.5

stright sides & end concave

carve& ground fine ground

L. 3063; beside no 575

side eroded

10 m S to L. 3042

14-12/4

round and concave

L. 3063; N &, beside no 576

flat bottom

L. 3063; N to no 577

eroded

10 m S to L. 3042, 7 m W to quarry

straight side & bottom; concave

L. 3063; beside & W to no 574

good

10 m S to L. 3042, 7 m W to quarry

16-18/8 7/5

falt

55 x 45 x 25

concave-cone

51

10 m S to L. 3042, 7 m W to quarry

accumulated

circle of 6 cups & 1 in middle

L. 3068; on rocks near quarry

near Roman quarry

L. 3068; on 3 rocks

near Roman quarry

L. 3069; on rock step beside Roman quarry

6 m E to fortified wall

L. 3070; on small bedrock

pounder found beside L

L. 3071; on long & narrow bedrock

W to L. 3043

Ayn Asil and Elephantine (Egypt): remarks on classification and function of ground stone implements Clara Jeuthe Keywords: groundstone tools, classification, quarries, distribution, Egypt

1. Introduction

2002, Soukiassian 1997, Jeuthe et al. 2014, Jeuthe 2012, Jeuthe in print a). The second site in the Balat area included in this paper belongs to the Sheikh Muftah community, the local semi-nomadic culture of Dakhla Oasis and its surroundings (research here has been generously supported by the Deutscher Akademischer Austauschdienst, DAAD). Here, a large camp site, located just north of the large enclosures of Ayn Asil, has been recently investigated (Jeuthe 2017, Jeuthe 2014, Jeuthe in print b). That site dates to the early fourth dynasty (ca. 2600 BC) and is therefore older than the Pharaonic Egyptian structures at Balat. It is contemporary to the early Pharaonic Egyptian presence in the oasis.

In most Pharaonic Egyptian settlements, the ground stone implements form one of the major find groups in addition to the flint tools. Although publications mentioning ground stone implements are rare (e.g. Prell in Forstner-Müller et al. 2015, Stevens 2012, Prell 2011, Boyce 1995 for the later periods, but fewer for our periods of interest, such as Jeuthe 2012, Egger 1994, Kopp et al. in print), they clearly display a variety of crafts and possible centres of production, the definition of which, we hope, will be one of the outcomes of our project. This project, generously supported by a research fellowship of the Deutsches Archäologisches Institut (DAI) and carried out in cooperation with the Institut français d’archéologie orientale (IFAO), is based on two major settlement sites in Ancient Egypt, both the focus of long-term research: Elephantine Island/ Aswan in Upper Egypt (DAI) and Ayn Asil/Balat in Dahkla Oasis in the Western Desert (IFAO).

A total of 1764 ground stone implements have been recorded at Balat since 2008. In the Governor’s Palace, this number includes 381 already published items found in a workshop area of the First Intermediate Period (Jeuthe 2012) as well as further 1083 recorded tools from other contexts in the palace. In addition, 290 implements come from the Sheikh Muftah camps. Despite being often less easy to study, as they bear fewer and less-pronounced wear marks than those found in the Pharaonic Egyptian contexts, the Sheikh Muftah tools help us to outline cultural contacts, local traditions, technological and procurement selection strategies, and organization.

2. The ground stone tools in context The sites provide us with a large number of wellstratified finds. The documentation of the ground stone tools is already well advanced at Balat, but work on the finds from Elephantine has only just started. We believe that we have recorded the majority of finds and now only some fragments are missing, which should not significantly change the general picture presented in this paper. However, a careful review of the find contexts is still to be undertaken, although we have already excluded from this study large fills and levelling layers as well as (building) pits and trenches. Thus, this paper provides a general overview of the material and addresses the first key issues, though a detailed stratigraphic and spatial analysis is not yet possible.

The bulk of the tools from Balat originated in Ayn Asil (n = 815), where two areas of the recent excavations in the Governor’s Palace have been selected for this study (Jeuthe et al. 2014). The first one is the so-called South-Eastern enclosure, from which 572 stone tools originated, mainly dating to Phase II, during the First Intermediate Period. The South-Eastern enclosure consists of two rows of room units grouped around a long courtyard (61 m x 32 m), of which the northern and southern ends have been excavated. The room units show relevant features indicating they were used as storage, work zones, and also as domestic spaces, which makes it difficult for the moment to understand the main function of this part of the Governor’s Palace. 234 stone tools were collected from the second research area at Ayn Asil, the so-called Phase 3 structures, dating to the Early Middle Kingdom. Until now, five buildings

The area of Balat contains several sites, of which two are the most important for this paper. The main site is Ayn Asil, in particular the Southern Enclosure: the so-called Governor’s palace (Fig. 1). It was built during the sixth dynasty and remained in use until the Early Middle Kingdom (ca. 2350-1900 BC), and had three major building phases (see Soukiassian et al. 52

C. Jeuthe: Ayn Asil and Elephantine (Egypt)

Figure 1. Overview Balat and research areas (D. Laisney, G. Soukiassian, after Jeuthe in print c, figure 1, Laisney 2011, figure 12, copyright IFAO).

53

Stone Tools in the Ancient Near East and Egypt

Figure 2. Overview Elephantine Island and research areas (after Kopp et al. in print, figure 1, Ziermann 2003, figure 1, Ziermann 1993, figure 2, Laisney 2011, figure 12, copyright DAI).

54

C. Jeuthe: Ayn Asil and Elephantine (Egypt) from this phase have been excavated, which appear to be mainly self-contained supply and storage units, covering an area of at least 45 m x 100 m. Phase 3 ended with a large fire, which preserved a large number of finds in-situ.

Quarries and corresponding activities are well attested in the region of Aswan (Harrell 2016, Harrell 2012, Brown & Harrell 2009, Harrell & Storemyr 2009, Kelany et al. 2009, Klemm & Klemm 2008, Klemm & Klemm 2001. See also http://www.eeescience.utoledo.edu/faculty/ harrell/egypt/quarries/Quarries_Menu.html). At the current stage our study is not yet able to relate the finds to specific quarries. For the moment, we can only assume that the majority of stone materials used in the research areas were of regional, if not local origin.

While the finds from Balat constitute the majority of our research corpus, a smaller group of 450 implements originating from Elephantine Island is included in this study as comparison. Elephantine Island, which is also located in a border region, has a long history from Predynastic times onwards. The two research areas chosen here (Fig. 2) have already been published, but without a detailed analysis of the finds. These areas are located close to the Satet Temple, and the contexts included in our project date mainly to the mid-2nd to 4th dynasties (ca. 2800-2500 BC). Satet North (Ziermann 2003) consists of an enclosed area of roughly 15 m x 20 m, which includes small rooms, courtyards, silos, and other storage installations with various rebuilding phases and modifications. These have changed the general character of the area from one used mainly for supply and storage to one with a more pronounced domestic purpose, though still including workshops and storage facilities. 176 ground stone implements have been recorded from Satet North so far.

Below is a list of raw materials and their varieties used for tools from our research areas at Balat and Elephantine (see also below, Table 1 to Table 3). In the following, we use the term ‘pebble’ also for cobbles. 3.1 Sandstone • •

(1)1 Pale to yellow soft, fine to medium grained sandstone, attested more commonly in Balat and less so Elephantine. (1)2 Pale to occasionally yellowish hard, fine grained and dense sandstone, attested in Balat and seldom at Elephantine.

The sandstone found in both sites may derive from the Nubian Sandstone series, though there are different formations that go under the term of sandstone (Harrell 2016: 20, Kindermann 2010: 51f., Klemm & Klemm 2008: 169). Two samples from Elephantine (used in temples) can be related to the Gebel el-Silsila area, apparently from both Nile banks (Klemm & Klemm 2008: 206-209).

The second area at Elephantine Island, Satet West (Kopp 2011, Kopp et al. in print), of around 5 m x 7.5 m, was excavated only recently and it yielded a house with packed sequences of minor plan renewals, without significant alterations. In addition, there are only a few specific features which may help to understand the activities carried out here. A bulk of 274 ground stone tools coming from this area is included in the corpus.

3.2 Quartzitic sandstone / conglomerate •

The ground stone tools at Elephantine had been previously recorded and partly published (Kopp et al. in print) and in addition, P. Kopp kindly provided us with his full documentation of the discarded pieces, allowing the artefacts to be fully reviewed. Furthermore, the descriptions and classification have been modified to make them comparable with the database used for the finds from Balat.

•

•

3. Raw materials The raw material attested in Ayn Asil are almost exclusively local (that is coming from the area close to Balat), or regional (that is coming from the oasis depression and its immediate surroundings). Exotic materials are almost absent. However, it should be noted that petrographic analyses have yet to be carried out.

•

The material classification found at Elephantine Island now follows the same classification as those in Balat, although they originated from different regions.

(2)1 / (2)2 Variants of black to blackish-violet conglomerate, mainly coarse-grained and spongy. Apparently, similar variants are attested in both Ayn Asil and Elephantine. (2)3 Grey to black, irregular but fine to medium grained and relatively soft, a burnt variant of (2)1 / (2)2? Hardly attested in the Pharaonic Egyptian contexts at Balat but found on the Sheikh Muftah site; not attested at Elephantine Island. (2)4 Light to medium grey, hard, almost silicified sandstone. Variants attested both in Balat and Elephantine. (2)5 Similar to variant (2)1, this is a silicified variant with dark brown to blackish colour, also attested in a more reddish variant. Variant (2)5 is still coarse but with a better granular binding, which means that presumably a higher amount of quartz fills the free pore space. This variant is only attested at Elephantine.

Without a petrographic determination, this category (2) refers to those stone materials that appear to be 55

Stone Tools in the Ancient Near East and Egypt neither sandstone nor fully developed quartzite (i.e. the quartz grains are still visible but do not bind fully) (see also Kindermann 2010: 50). Both sites have direct access to sources for of this material, although not all the types of quartzitic sandstone are attested in the Balat area. However, the conglomerate (2)1 / (2)2 occurs approximately 30 km from Balat. In Aswan, possible corresponding quarries for variant (2)5 are known at Qubbet el-Hawa at the West bank (Harrell & Storemyr 2009: fig. 7, fig. 31).

naturally occurring granite pebbles. Quarries are also well attested there (Kelany et al. 2009, Harrell & Storemyr 2009: fig. 7, Brown & Harrell 2009, Klemm & Klemm 2008: 233-267). Granite does not occur in the Dakhla Oasis. However, it has been used at the prehistoric Djara site, located on the plateau (Kindermann 2010: 51). Varia Dolerite was used occasionally at Elephantine Island; diorite (or granit diorite) is even less common and marl is only exceptionally attested in Ayn Asil. River pebbles of different types of hard stone found at Elephantine come from locally available sources, while tools made of quartz are rare in both sites.

3.3 Quartzite This comprises variants of fine-grained, dense and hard fully developed quartzite (i.e. silicified sandstone) from light grey to almost blackish grey in colour, occurring naturally as pebbles but also as thin slabs. These tabular pieces are only attested in Balat, while the examples at Elephantine are mainly of a lighter grey colour, in some cases slightly reddish, and can display larger grains.

4. Classification The classification system comprising four main categories plus a cluster of non-classified items was established in 2008 (Jeuthe 2012), following the work of Prell 2011, and since then has been modified and enlarged as necessary. The action involved in the tool usage, that is long or short strokes, the application of pressure or sweeps, whether soft or strong etc., became one of the main criteria for classification (based on the ideas of de Baune 2000: 20-23, 33-42). These differences in action, force, angle and direction are expressed in different use wear (i.e. stress marks and polished areas), whose nature is in relation to both the contact material and the material of the tool (depending, for example, on the hardness, abrasion resistance, or brittleness, of both the tool and contact materials). Moreover, the position of these wear marks (work face or edge) may also contain important information as well as indicating the manner of handling of the tool (e.g., active, passive, mounted, ease of handling, etc). Other criteria help define the main function of the tool and rely on the combined observation of shape, manufacture, form of the working face(s), and body shape, as well as the raw material chosen. The tools often appear to be multifunctional and/or of unique shape. An individual tool may have had various functions within a single category but use marks indicate activities from multiple categories less often. As scientific analyses, which may in the future help for identification of tools’ uses, have not yet been carried out, the categories of this classification aim to narrow the activities to which a given tool was destined. This is a first step towards a more in-depth functional interpretation. Consequently, tools are always classified according to their last definable use and not by their primary function. In contrast, ad-hoc use is very rarely identifiable, and is not included as a main criterion.

The fully-developed quartzite variant belongs to the hardest type of stone used at both sites. Quarries are well attested in the region of Aswan (Harrell & Storemyr 2009: fig. 7, Klemm Klemm & Klemm 2008: 219-227), while at Balat, no contemporary quarries have been identified, although the raw material was observed there in the close surroundings (Jeuthe 2010: 213ff.). 3.4 Limestone • •

(4)1 Limestone of varying degrees of hardness in the shape of small and hard pebbles, which are not attested at Elephantine. (4)2 Fossil limestone conglomerate, coarse and spongy, either yellowish or grey, only known in large blocks. This variant does not occur at Elephantine Island and only rarely at the Sheikh Muftah site, but is more common at Ayn Asil.

While these limestone varieties belonging to the Egyptian Limestone Plateau are directly accessible in Dakhla Oasis (see Kindermann 2010: 51), they needed to be imported into Elephantine and here they appear to have been only exceptionally used for tools. 3.5 Flint Variants of flint nodules can be collected locally in Ayn Asil, but the nodules found at Elephantine were imported. However, flint was only rarely used for tools on both sites (the knapped flint/chert industries are not included in the present study).

Moreover, a category can be subdivided into different clusters, mainly based on presumed functions and movements. Clusters include different tool types or tool groups, the latter comprising different types according

3.6 Granite Variants of granite, including pebbles, were only used at Elephantine where a number of tools are made of 56

C. Jeuthe: Ayn Asil and Elephantine (Egypt) to their use marks and also shape. For example, the group ‘hammer’ consists of individual hammer types such as ‘simple’, ‘cubic’ and ‘flat-convex’ amongst others.

The second cluster of ‘battering tools’ includes hammer stones which are likely to have been used for battering/ shattering/cutting (i.e. a long and/or sweeping movement but forceful pressure towards a point), and comprises types such as ‘pick-like’ (Fig. 3: 3567-32), ‘axe/hack/chopper-like’, ‘adze-like’ (Fig. 3: 3318-1), ‘simple hammer’, ‘ball/pebble hammer’ (Fig. 3: 3688-4, 3579-2, 3652-1), ‘gravel hammer’, and also anvils.

The first category includes tools showing use marks from pressure and forceful (usually) targeted movements, while the second category designates grinding tools exclusively related to grain processing (excluding, tools for grinding pigments, or tools for grinding non-cereal plants, as far as can be distinguished by the direction of movement and/or shape of the tool). Tools of the third category usually show stress marks resulting from long, wide, sweeping, but less forceful movements. Netherstones (this term comes from the classification proposed by Adams 2014 where it is used as the most neutral term among those proposed for this type of object) of indistinct and often various uses form the fourth category. All tools which are either too poorly preserved, or whose corresponding activities are not fully defined, are referred to as unclassified. Moreover, specific industries, such as the stone vessel production in Elephantine, are excluded from the four main categories.

‘Wheel-shaped’ hammers (Fig. 3: 2666-7) differ from the above because they often show wear marks closely related to the third category (see Prell 2011: 35 for similar observations) but they also fit into the second cluster of hammer. Without trace analyses, we cannot define their main function. Hence, they remain in the first category for the time being. Again, tools may bear traces of multiple activities, but these are often within the same general cluster. All these sub-classifications are based on the form of the tool, in particular on the working edge. Nonetheless, during the object recording, it was obvious that certain hammer types are strongly related to specific marks and position of work faces, thus truly defining a type. 4.2 Tools of the second category – ‘grinding stones’

4.1 Tools of the first category – ‘pressing and battering tools’

Grinding stones (i.e. handstones, querns, or slabs) or are well attested in the Pharaonic Egyptian contexts and are also represented in daily life models found, for example, in the tomb of Meket-Re/Theben (Winlock 1955) or in the tomb of Henu/ Dayr el-Barsha (De Meyer 2007). Long and massive lower stones, usually saddleshaped querns, are typically depicted as non- or hardly portable items, while most of the upper handstones are used with two hands and are boat-shaped. However, large querns are only known in Ayn Asil during the Phase I / 6th dynasty, and were replaced thereafter by small, portable, often flat slabs, usually not more than 30 cm in length (Fig. 4: 3576-1). Apart from showing corresponding stress marks in the longitudinal direction, some of these small portable slabs were found in situ in shallow depressions in the floor or on working platforms, hence their interpretation as lower stones is certain (see also Prell 2011: 72-80 for portable slabs in Piramesse during the New Kingdom). They were carefully shaped and show pick marks all over the body. The grinding stones at Ayn Asil are made almost exclusively from a sandstone conglomerate of the variant (2)1 and only seldom from fossil limestone, (variant (4)2, see below, Table 1), which was more popular for the large querns. Related handstones with corresponding stress marks in the transverse direction (Fig. 4: 3623-22) occur less often than slabs but may be hidden among some indeterminate grinding stone fragments which only show clear evidence of use on the working surface and/or facial abrasion without any clear direction of movement. We may also anticipate a

Tools of the first category are involved in various activities related to the use of forceful pressure (i.e. impacting, smashing, crushing, battering), and also related to a pressure applied through short movements for pulverizing, mashing, and polishing (Fig. 3). Unlike in the previous study (Jeuthe 2012), polishing tools are now placed in this category. The most common marks visible on these tools are battering scars and damage from scaling, which can form shallow depressions and polished areas, often in combination with short, sometimes circular, scratch marks. The preferred choice of raw material suitable for such activities is a hard stone, often in a pebble form (see below, Table 1 to Table 3). Tool groups included in this category are hammer stones, pounders, pestles, polishing tools or – in the case of no detailed tool definition possible – generic ‘pressure tools’ only. This category can be divided into two main clusters of which the first consists of different ‘pressing tool’ types, involved in activities entailing only short, abrupt movements (i.e. percussion within a small area). Corresponding ‘passive’ tool types, such as mortars, occur only rarely. More common, though still rare, are the palettes with traces of pigment. ‘Active’ tool types are pounders, pestles, polishing tools, ‘cubic hammers’ (Figure 3: 3655-11) and ‘flat-convex hammers’. Both show similar marks and are therefore grouped together. 57

Stone Tools in the Ancient Near East and Egypt

Figure 3. Tools of the first category from Ayn Asil: ‘wheel hammer’ 2666-7, ‘cubit hammer’ 3655-11, ‘pebble hammer’ 3688-14, 3652-1, 3579-2, ‘pick-like’ 3567-32, ‘adze-like’ 3318-1 (scale 1:2, drawn by A. Hussein, C. Jeuthe, copyright IFAO).

58

C. Jeuthe: Ayn Asil and Elephantine (Egypt)

Figure 4. Tools of the second category from Ayn Asil: slab 3576-1, handstone 3623-22 (scale 1:2, drawn by A. Hussein, copyright IFAO).

higher number of grinding balls than the few classified items (see below, Table 1, Table 2).

Thus, tools either classified within other categories, such as ‘pressure tools’ in category 1 or ‘abrasive/ 59

Stone Tools in the Ancient Near East and Egypt

Figure 5. Tools of the third category from Ayn Asil: abrasive stones 3635-25, 3906-13, 3535-2, 3906-12 and whetstones 3567-11, 3927-37, 2667-1 (scale 1:2, drawn by E. Gossens, A. Hussein, C. Jeuthe, copyright IFAO).

60

C. Jeuthe: Ayn Asil and Elephantine (Egypt) smoothing tools’ in category 3, may have been also used for grinding activities (possible as multifunctional tools), but the use wear is not clear enough to able to classify them.

different stress marks, and the edges are (in some cases) rounded or appear to have a scraper-like shape. Tools within this category are often used in multiple ways, with whetstones designated as the tools with the most specific purpose. Whetstones bear clear whetting channels, either with a V-shape (possibly indicating edges such as knives etc.; Fig. 5: 3567-11) or more often a U-shape to rounded section (which may imply work on bone or wooden items; Fig. 5: 3927-37). A whetstone can also bear both kinds of channels (Fig. 5: 2667-1).

The grinding stones from Elephantine do not differ significantly from those from Balat, although larger querns may occur in a greater variety of contexts on Elephantine Island according to P. Kopp. Apart from one fragment, they are also absent in the selected research areas. Instead, most classified grinding tools are handstones because of their half-round section and relative thickness. Some handstones bear clear transverse stress marks, but the majority have only indistinct concentrations of marks or abrasions. Therefore, the number of slabs may be higher than we currently see due to misclassification as handstones. Small handstones that can be used in one hand have also been found, but the same problem of misclassification mentioned above may account for the low number of grinding balls listed at Elephantine Island. Unlike Ayn Asil, the group of non-classified tools contains a higher number of pebble tools without (or with only a very few) distinct use marks. These may have also been used for grinding.

4.4 Tools of the fourth category – ‘multifunctional netherstones’ This category comprises all remaining netherstones (i.e. work boards, lapstones, slabs, …) without distinct traces of a specific use (Fig. 6: 3621-16). They are predominately made from harder stone materials and often from (broken) grinding stones. Netherstones are in most cases carefully made and relatively small; massive, non-portable items occur only exceptionally. They can bear multi-directional traces (such as multidirectional scratch marks; Fig. 6: 3567-24), as well as glossy areas and/or shallow depressions (Fig. 6: 3824-5). One may presume that they were used in various ways and therefore were multi-functional. A careful analysis of the find contexts is needed to try to reconstruct their main functions. While this has not yet been undertaken in detail, preliminary observations at Ayn Asil relate these objects not only to craftwork but also to food processing.

4.3 Tools of the third category – ‘abrasive/smoothening tools’ Tools of this category found in Ayn Asil rarely show traces of intentional forming, and quite a few of them may have been used only briefly or even ad hoc. Intentional shaping has been observed more often at Elephantine, where there is also evidence of more intensive use of these tools. This category includes sharpening tools, mainly of soft stone with a high abrasion factor, such as whetstones and abraders. It also includes smoothing tools (Fig. 5). Smoothing may also be considered polishing, but we use the term ‘smoothing’ to define activities in which less use of pressure than polishing is assumed, because of the soft raw material that was selected. Consequently, the work surfaces on tools in this category are often abraded and/or have a highly compacted work face but have not developed the shiny gloss characteristic of polishing tools. Abrasive stones can show clear stress marks, such as scratches and deeper striations, often indicating long movements in various directions, including curved motions (Fig. 5: 3635-25, 3906-13, 3906-12). They are also attested as netherstones used for abrasion or sharpening (Fig. 5: 3535-2). These netherstones are often carefully shaped and their concave working surface commonly shows longitudinal or slightly curved striations, although never in a transverse direction. Use wear also depends on the contact material, thus the distinction between ‘abrading’ and ‘smoothing’ refers to the types of stress marks rather than to the precise function. Moreover, a tool can have more than one working surface showing

The same is true for the Sheikh Muftah group at Balat; here netherstones are often related to food processing as well. But since, in general, stress marks on ground stone tools from the Sheikh Muftah camps are less pronounced than on the ‘pharaonic’ tools from Ayn Asil, and the tools seem to have been used more often in a multi-functional way, most of the netherstones from Sheikh Muftah cannot be included in any of the first three categories. Hence, they have been assigned to the fourth one. 5. General observations and remarks on the stone tool assemblages The assemblages from each site show specific patterns in the relationship between raw materials and tool categories, for which stone characteristics such as hardness, or degree of abrasion, certainly play a key role (see Table 1 to Table 3). Not only were local stone variants used almost exclusively, but those with the most direct access predominate. Additionally, the raw material selection in the Sheikh Muftah community demonstrate the same selection strategy; the preference for the nearest available stone is even more obvious here than at the other sites. 61

Stone Tools in the Ancient Near East and Egypt

Figure 6. Tools of the fourth category from Ayn Asil: netherstones 3621-16, 3527-24, 3824-5 (scale 1:2, drawn by E. Gossens, A. Hussein, copyright IFAO).

Despite the prominent selection of local materials, larger workshops for ground stone implements are absent from all the sites under study. Only the remains from a limited production of simple quartzite flake tools have been found in Satet West. Concentrations of remains from other tool-shaping activities have been never recorded, although these may have been discarded as fragment or splint by excavators. Still, we can assume that tool making may have taken place in

the settlements (perhaps on an ‘on demand’ basis) as workshops are also unknown at the quarries. The only exception to this are from Pharaonic Egyptian culture where we find grinding stones that were made in specific workshops at the quarries (see below). Local sandstone varieties are the most common materials in Balat, both in Ayn Asil (44.4%) and at the Sheikh Muftah site (50.3%). Locally available harder 62

C. Jeuthe: Ayn Asil and Elephantine (Egypt) stone types such as quartzite and granite predominate at Elephantine Island (35.5% respectively 17.1%), where limestone is almost absent (1.3%). Pebbles from all sites show a wide range of all kinds of stones. Nonetheless, at least in the case of Ayn Asil, we see a degree of selection, as flint nodules are seldom used as pebble tools. This may be a result of the poor compositional quality of the flint available.

for which the preferred choice is a mainly a dark conglomerate of hard quartzitic sandstone (variants (2)1/(2)5). Despite its porousness and large cavities, the hardness and low abrasion factor seem to have made it the preferred choice at both sites. While this pattern of selection and provision might be less surprising for Nile Valley sites, the comparison with the semi-nomadic Sheikh Muftah community at Balat demonstrates differences from Pharaonic Egyptian culture in the oasis (Table 1, Table 3). Only a few grinding stones have been found in the Sheikh Muftah camp sites and all but one of them were made from hard sandstone (variant (1)2), and one from a quartzitic sandstone (variant 2(4)). Neither of these, by contrast, was used for grinding stones at Ayn Asil. Moreover, both preferred choices at Ayn Asil, the conglomerate (variant (2)1/2, 85.5%) and the fossil limestone (variant (4)2, 11.2%), are almost completely absent from the Sheikh Muftah site. Only a small fragment of a grinding slab made from the ‘Pharaonic’ conglomerate has been found, in an eroded area. However, it is possible that the grinding stones found on the Sheikh Muftah site may have been merely collected from the surface of other sites, either prehistoric or Pharaonic Egyptian. In the latter case, we would observe a change in preferred

Collected pebbles are most common in the first category (see Table 1 to Table 3), which for this reason shows the widest range of materials. Furthermore, the fourth category of (multi-purpose) netherstones also shows a wide range of, mainly hard, materials. The data appears to be slightly biased by the number of reused grinding stones in this category, especially in the case of Elephantine. Excluding them, the preferred material for such items would be (hard) sandstone and quartzite at Ayn Asil and granite at Elephantine, thus in both cases the hardest material available. In contrast, tools of the third category are made of soft sandstone as this stone is more suitable for the activities and movements related to this category. The strongest relationship between material selection and tool type can be found in the grinding stones, total n = 815

total %

Cat. 1 (n = 207 Cat. 2 (n = 152 / 25.4%) / 18.7%)

Cat. 3 (n = 337 Cat. 4 (n = 77 / non-classified / 41.3%) 9.4%) (n = 42/ 5.2%)

sandstone

362

44.4%

13 / 6.3%

308 / 91.4%

(1) pebble

5

0.6%

1 / 0.5%

3 / 0.9%

(1)1

294

36.1%

2 / 1%

271 / 80.4%

10 / 13%

11 / 26.2%

34 / 10.1%

11 / 14.3%

8 / 19%

21 / 27.3%

20 / 47.6% 1 / 2.4%

(1)2

63

7.7%

10 / 4.8%

quartzitic sandstone

244

30%

50 / 24.2%

130 / 85.5%

27 / 8%

32 / 41.6%

5 / 11.9%

(2)1 / (2)2

216

26.5%

31 / 15%

130 / 85.5%

27 / 8%

24 / 31.2%

4 / 9.5%

8 / 10.4%

1 / 2.4%

14 / 18.2%

8 / 19%

1 / 1.3%

3 / 7.1%

13 / 16.9%

5 / 11.9%

10 / 13%

8 / 19%

1 / 1.3%

1 / 2.4%

1 / 1.3%

6 / 14.3%

8 / 10.4%

1 / 2.4%

(2)4

28

3.4%

19 / 9.2%

sil. sandstone / quartzite

96

11.8%

71 / 34.3%

2 / 1.3%

(3) pebble

58

7.1%

52 / 25.2%

2 / 1.3%

(3) quartzite

38

4.7%

19 / 9.2%

limestone

73

9%

38 / 18.4%

(4) pebble

23

2.8%

21 / 10.1%

(4)1

19

2.3%

12 / 5.8%

(4)2

31

3.8%

5 / 2.4%

(5) flint pebble

16

2%

16 / 7.7%

varia

4

0.5%

3 / 1.4%

1 / 0.3% 1 / 0.3%

marl

2

0.2%

1 / 0.5%

quartz

2

0.2%

2 / 1%

non-classified

20

2.5%

16 / 7.7%

pebbles

15

1.8%

15 / 7.2%

other

5

0.6%

1 / 0.5%

1 / 0.3% 1 / 0.3%

17 / 11.2%

17 / 11.2%

1 / 2.4% 3 / 2%

Table 1. Ayn Asil: raw materials of stone implements attested and their frequency rates within the individual categories.

63

1 / 2.4%

Stone Tools in the Ancient Near East and Egypt

total n = 450

total %

53

11.8%

(1) sandstone

3

0.7%

3 / 5.7%

(1)1

41

9.1%

sandstone

Cat. 1 (n = 234 Cat. 2 (n = 69 / Cat. 3 (n = 53 / Cat. 4 (n = 18 / 52%) 15.3%) 11.8%) / 4%) 3 / 1.3%

1 / 1.4%

44 / 83%

non-classified (n = 76 / 16.9%)

3 / 16.7%

2 / 2.6%

38 / 71.7%

2 / 11.1%

1 / 1,3%

(1)2

9

2%

3 / 1.3%

1 /1.4%

3 / 5.7%

1 / 5.6%

1 / 1,3%

quartzitic sandstone

106

23.6%

11 / 4.7%

65 / 94.2%

6 / 11.3%

7 / 38.9%

17 / 22.4%

(2)1 / (2)2

30

6.7%

1/ 0.4%

21 / 30.4%

2 / 3.8%

3 / 16.7%

3 / 3.9%

(2)4

19

4.2%

7 / 3%

1 / 1.4%

1 / 5.6%

11 / 14.5%

(2)5

57

12.7%

sil. sandstone / quartzite

43 / 62.3%

99

22%

83 / 35.5%

(3) pebble

49

10.9%

44 / 18.8%

(3) quartzite

50

11.1%

39 / 16.7%

limestone

6

1.3%

3 / 1.3%

3 / 3.9%

(4) pebble

3

0.1%

2 / 0.9%

1 / 1.3% 2 / 2.6%

(4) limestone

3

0.1%

1 / 0.4%

(5) flint pebble

4

0.9%

4 / 1.7%

granite

53

11.8%

40 / 17.1%

(6) pebble

14

3.1%

13 / 5.6%

4 / 7.5% 1 / 1.9%

3 / 3.9% 1 / 5.6%

14 /18.4% 5 / 6.6%

1 / 1.9%

3 / 4.3%

1 / 1.9%

1 / 5.6%

5 / 27.8%

9 / 11.8%

4 / 5.2% 1 / 1.3%

(6) granite

39

8.6%

27 / 11.5%

varia

30

6.7%

22 / 9.4%

8 / 10.5% 1 / 1,3%

diorite

4

9.9%

3 / 1.3%

dolerite

10

2.2%

10 / 4.3%

quartz

16

3.6%

9 / 3.8%

river pebble

43

9.6%

31 / 13.2%

non-classified

56

12.4%

37 / 15.8%

pebble

45

10%

30 / 12.8%

other

11

2.4%

7 / 3%

3 / 4.3%

1 / 1.9%

5 / 27.8%

3 / 3.9%

7 / 9.2% 1 / 1.9%

1 / 5.6%

10 / 13.2%

1 / 5.6%

18 / 25% 15 / 19.7%

1 / 5.6%

3 / 3.9%

Table 2. Elephantine: raw materials of stone implements attested and their frequency rates within the individual categories.

sources within Egyptian Pharaonic culture in Dakhla Oasis between the 4th dynasty and Ayn Asil from the 6th dynasty onwards. Nonetheless, the difference in strategies remains striking despite being governed by the same environmental conditions and available stone materials.

in Dahkla Oasis (yet), the grinding stones found at Elephantine Island may come from the quarries and workshops at Qubbet el-Hawa on the west bank (Harrell & Storemyr 2009: fig. 7, fig. 31). At first sight, the stone there strongly resembles our variant (2)5. This remains to be confirmed by petrographic analysis. Moreover, these workshops seem have been used specifically for the production of grinding stones, because the raw materials at these quarries were never the primary choice for other tools attested in our research areas. Instead, these materials occur in other tool categories only as reused former grinding stones. Therefore, the production and distribution of grinding tools appears to be state-organised. The same strategy has been also observed in the production of sickle blades made of chert (Jeuthe in Forstner-Müller et al. 2015, Jeuthe 2012, Tillmann 2007), which may relate the production of different tools related to harvesting and bread production to a state-organised economic system.

In this way Egyptian Pharaonic culture, both in the Nile Valley and the oasis, demonstrate a specific pattern regarding grinding stones. Moreover, it should be noted that on-site production has never been identified. Evidence available indicates specific quarries with adjunct workshops for the production of grinding stones known elsewhere in prehistoric and dynastic times (Harrell 2012, Harrell & Storemyr 2009, Storemyr 2014, see also Bloxam et al. 2009, see also Prell 2011: 72f. for the restricted material selection on grinding stones in Piramesse during the New Kingdom). While specific quarries and workshops have not been located 64

C. Jeuthe: Ayn Asil and Elephantine (Egypt)

sandstone

total n = 290

total %

146

50.3%

Cat. 1 (n = 27 / Cat. 2 (n = 8 / Cat. 3 (n = 46 / Cat. 4 (n = 93 / non-classified 9.3%) 2.8%) 15.9%) 32.1%) (n = 116 / 40%) 4 / 14.8%

6 / 75%

44 / 95.7%

37 / 40%

55 / 47.4%

42 / 91.3%

8 / 8.6%

31 / 26.7%

(1) pebble

1

0.3%

(1)1

83

28.6%

2 / 2.4%

1 / 0.9%

(1)2

62

21.7%

2 / 2.4%

6 / 75%

2 / 4.3%

29 / 31.2%

23 / 19.8%

quartzitic sandstone

56

19.3%

1 / 3.7%

2 / 25%

1 / 2.2%

26 / 28%

26 / 22.4%

(2)1 / (2)2

9

3.1%

1 / 12.5%

1 / 2.2%

(2)3

26

9%

(2)4

21

7.2%

1 / 3.7%

sil. sandstone / quartzite

48

16.6%

4 / 14.8%

(3) pebble

8

3.1%

2 / 2.4%

(3) quartzite

40

13.8%

2 / 2.4%

limestone

27

9.3%

5 / 18.5%

(4) pebble

21

7.2%

4 / 14.8%

(4)1

4

1.4%

1 / 3.7%

(4)2

2

0.7%

(5) flint pebble

4

1.4%

4 / 14.8%

varia

1

0.3%

1 / 3.7%

non-classified

8

2.8%

7 / 25.9%

pebble

7

2.4%

7 / 25.9%

other

1

0.3%

1 / 12.5% 1 / 2.2% 1 / 2.2%

5 / 5.4%

9 / 7.8%

8 / 8.6%

18 / 15.5%

13 / 14%

6 / 5.2%

26 / 28%

17 / 14.7%

2 / 2.2%

4 / 3.4%

24 / 25.8%

13 / 11.2%

4 / 4.3%

17 / 14.7% 16 / 13.8%

3 / 3.2% 1 / 1.1%

1 / 0.9%

1 / 0.9% 1 / 0.9%

Table 3. Balat/Sheikh Muftah: raw materials of stone implements attested and their frequency rates within the individual categories.

Regarding the occurrence rate of the different types of grinding stones (slabs/handstones) in the individual research areas, their ratio is always problematic (Table 4). In Ayn Asil the lack of active grinders (with 22.1% versus slabs with 55.3%) may be related to problems with the classification of grinding balls. Their frequency at Elephantine presents the opposite situation; there handstones/grinding balls predominate (with 79.2% versus slabs with 8.7%). A quick glance at the difference in percentages between slabs (33.3%) and handstones (41.6%) in Satet North suggests actual grinding, but their find contexts do not support this, as most of them are isolated finds from different rooms. It seems more likely that grinding took place in specific zones (such as close to at bakeries) and the items found distributed within the settlements may have been reused for a different purpose, or simply discarded. We need to review the features (such as silos) to be certain about this assumption.

Satet West (Kopp et al. print). Here, as in Satet North, the find contexts and associated finds do not imply any production. Regarding the first category, the research areas Phase 3 at Ayn Asil and Satet North at Elephantine Island show a preference for the cluster of ‘pressing tools’ (61% respectively 59.4%), while the ratio between the clusters is more or less balanced with the other research areas (Table 4). This might be a first indication of differences in main activities, but this must still be analyzed alongside the find contexts and features. However, there is no specific preference for the ‘battering’ or ‘pressing’ tool clusters obvious in the Sheikh Muftah site; this deserves an analysis related to the individual camps as done elsewhere (Jeuthe in print b). Moreover, first category tools are the most frequent in both areas at Elephantine Island (49.8%). In contrast, at Ayn Asil, the number of tools classified in the third category (41.3%) is almost twice as the first one (25.4%). We have noted the low frequency of battering and pressure tools in Ayn Asil elsewhere, hence we presume that they are more likely to have been in use for a longer time and were rarely thrown away during the occupation of the buildings. A large bulk of these

In other categories as well there are certain similarities between the sites when looked at overall, but the picture becomes more ambiguous when comparing individual research areas (Table 4). For example, it is interesting to note that stone vessel production tools are absent at Balat whereas they have been almost exclusively found 65

Stone Tools in the Ancient Near East and Egypt

Ayn Asil SE enclosure total n = 815 n = 572

Phase 3 n = 243

Elephantine Satet West Satet North Sheikh Muftah total n = 450 n = 274 n = 176 n = 290

Category 1

n = 207 / 25.4%

n = 130 / 22.7%

n = 77 / 31.7%

n = 234 / 49.8%

n = 138 / 46.7%

n = 96 / 54.5%

n = 27 / 9.3%

cluster 1 (pressing tools)

110 / 53.1%

63 / 48.4%

47 / 61%

113 / 48.2%

56 / 40.6%

57 / 59. 4%

13 / 48.1%

cluster 2 (battering tools)

95 / 45.9%

65 / 50%

30 / 39%

105 / 44. 9%

66 / 47. 8%

39 / 40.6%

13 / 48.1%

non-classified

2 / 1%

2 / 1.5%

16 / 6.8%

16 / 11. 6%

Category 2

n = 152 / 18.7%

n = 121 / 20.2%

n = 31 / 12.8%

n = 69 / 15.3%

n = 57 / 20.8%

grinding balls

6 / 3.9%

6 / 19.4%

1 / 1.4%

1 / 1.7%

handstones

28 / 18.4%

25 / 20.7%

3 / 9.7%

53 / 76.8%

48 / 84.2%

5 /41. 6%

slabs

84 / 55.3%

65/ 53.7%

19 / 61.3%

6 / 8.7%

2 / 3.5%

4 / 33.3%

non-classified

34 / 22.4%

31 / 25. 6%

3 / 9.7%

9 / 13.1%

6 / 10. 5%

3 / 25%

Category 3

n = 337 / 41.3%

n = 259 / 45.3%

n = 78 / 32.1%

n = 53 / 11.8%

n = 33 / 12%

n = 20 / 11.7%

n = 46 / 15.9%

abrasive stones

181 / 53. 7%

150 / 57.9%

31 / 40%

29 / 54.7%

20 / 60. 6%

9 / 45%

19 / 41.3%

smoothing tools

69 / 20.5%

49 / 18.9%

20 / 25.6%

13 / 24.5%

7 / 21.2%

6 / 30%

13 / 28.2%

whetstones

43 / 12.8%

33 / 12.7%

10 / 12.8%

6 / 11.3%

3 / 9.1%

3 / 15%

9 / 19.6%

16 / 20.5%

5 / 9.4%

3 / 9.1%

2 / 6.7%

scraper-like

2 / 0.6%

2 / 0.8%

netherstones

33 / 9.8%

17 / 6.6%

1 / 3.7% n = 12 / 17.4%

n=8/ 2.8%

8 / 100%

1 / 2.1%

non-classified

9 / 2.7%

8 / 3.1%

1 / 1.3%

Category 4

n = 77 / 9.4%

n = 44 / 7.7%

n = 33 / 13.6%

n = 18 / 4%

n=7/ 2.6%

n = 11 / 6.3%

n = 93 / 32.1%

non-classified

n = 42 / 5.2%

n =18 / 3.1%

n = 24 / 9.9%

n = 76 / 16.9%

n = 39 / 14.2%

n = 37 / 21%

n = 116 / 40%

1 / 1.3%

1 / 2.6%

24 / 100%

57 / 75%

21 / 53.8%

36 / 97.3%

116 / 100%

18 / 23.7%

17 / 43.6%

1 / 2.7%

stone disk

4 / 0.5%

4 / 22.2%

other tools related to stone vessel production

38 / 4.7%

14 / 77.8%

4 / 8.7%

Table 4. Frequency rates of functional categories within the sites and individual research areas.

tools have been found within the abandonment layers, supporting this suggestion. Thus, we have to assume that activities related to the first category of tools are more common than our data from the actual use contexts suggest. By contrast, third category tools are rarely carefully shaped – if shaped at all – and not intensively used. They are also easy to replace, because the outcrops are located directly next to the enclosures of Ayn Asil (Jeuthe 2012: 240, 271f.).

Elephantine Island. A detailed analysis into functional classes needs to be undertaken in the near future, but even in this early stage of research it has become obvious that the number of reused pottery sherds is not only significantly higher than in Ayn Asil, the pieces also have a variety of use traces on their edges and faces. In most cases, these seem to correspond with the use traces observed on stone tools of the third category. With the exceptions of marl clay sherds, sherds are too weak or small to be related to activities in the first or second categories and show none of the corresponding use marks (such as battering) scars) that we would associate with those tool categories. However, the rare use of ceramic sherds as palettes is attested. We can conclude that the number of abrasive and smoothing implements at Elephantine Island is much larger than that shown by the stone tools alone. Thus we propose that the reuse of ceramic sherds on the island correlates, with few exception only, with the use of the soft sandstone in the oasis. Following that hypothesis both sites would have relied on the same strategies

Another peculiarity of tool distribution within the individual research areas is demonstrated by the frequency of third category tools (Table 4). In comparison with Ayn Asil (41.3%) these tools are notably underrepresented at Elephantine Island (11.8%). Moreover, the tools appear to be more carefully shaped and intensively used than in both Ayn Asil and the Sheikh Muftah camps. Consideration of all finds, in particular the reused ceramic sherds which are very frequent at Elephantine Island, suggests a possible explanation for the small number of these tools at 66

C. Jeuthe: Ayn Asil and Elephantine (Egypt) for ad-hoc or on-demand tools by choosing a directly available material.

tool and material selections. Particularly concerning grinding stones, we find much stronger similarities between Ayn Asil and Elephantine Island. This emphasises the importance of tools related to bread production, also apparent from the production and supply of sickle implements and in the large bakeries. Hence, a sufficient supply is needed for groups of people and work teams (see as first introduction into the Ancient Egyptian system of provision Kemp 1989: 163, 171-184).

While the reused ceramic sherds might well explain the low occurrence rate of third category tools at Elephantine Island, abrasive and smoothing tools are also not very frequent at the Sheikh Muftah camps. Their numbers may be slightly biased as further fragments without a preserved shape or clear traces of use made from soft sandstone variants remain unclassified. But given the easy access to the stone material, the number remains low, especially compared with their frequency at Ayn Asil. In this case, we may presume that the corresponding activities were carried out on a small scale only, as reused ceramic sherds with use traces corresponding to the third category occur only occasionally.

In terms of tool frequency, in both Egyptian Pharaonic sites and their individual research areas, we observed similarities (for example, a higher number of press tools and netherstones, highlighting the similarities between Satet North at Elephantine Island and Phase 3 in Ayn Asil). We understand this as a first indication of the functional differences within a settlement, whereby corresponding activities are less well visible within the ground stone assemblages. In future research, we hope to elucidate in more detail the activities occurring at these sites through in-depth functional and spatial analyses of the tools, and their frequencies.

Finally, the frequencies within the third category show small differences between the individual areas, although the general order of occurrence of the individual tool types is the same (Table 4). Abrasive stones are less dominant in Satet North (45%) than at Satet West (60%) and Elephantine Island. In Balat they are also more frequently attested in the SouthEastern enclosure (57.9%) than in Phase 3 (40%). The same pattern not only occurs in the cluster of ‘pressing tools’ within the first category (see above) but also by the netherstones of the fourth category. They were also found more often in the area of Satet North (6.3%) and the structures in Phase 3 (13.6%) than in the other areas. First observations at Ayn Asil hint at a predominance of contexts related to food preparation rather than craftwork, which also applies to the Sheikh Muftah site. There, netherstones are the largest group of classified tools (32.1%) and although often being multifunctional, they were more likely to have been found with other items and features related to cooking areas than in other activity zones. However, netherstones are far more common at Ayn Asil (9.4%) than at Elephantine Island (4%), for which we lack a satisfying explanation. Smaller-scale food processing cannot be excluded as yet in the selected research areas there, but we need a detailed review of their find contexts.

Bibliography Adams, J. 2014. Ground Stone Analysis: A Technological Approach, 2nd ed. Salt Lake City: The University of Utah Press and American Southwest. de Baune, S. 2000. Pour une archéologie du geste: Broyer, moudre, piler. Des premiers chasseurs aux premiers agriculteurs. Paris: CRNS Éditions. Bloxam, E., P. Storemyr, and T. Heldal 2009. Hard stone quarrying in the Egyptian Old Kingdom (3rd millennium BC): Rethinking the social organization, in Y. Maniatis (ed.) ASMOSIA VII, Proceedings of the 7th International Conference of the Association for the Study of Marble and Other Stones in Antiquity, Thassos 15-20 September 2003 (Bulletin de Correspondance Hellénique supplement 51): 187-201. Athènes: École Française d’Athènes. Boyce, A. 1995. The finds, in B. J. Kemp Amarna Report 6 (Egypt Exploration Society Occasionally Publications 10): 44-136. London: Egypt Exploration Society. Brown, V. M. and J. A. Harrell 1998. Aswan granite and granodiorite. Göttinger Miszellen, Beiträge zur Ägyptologischen Diskussion 164: 33-39. Eger, C. 1994. Steingeräte aus dem Umfeld der Roten Pyramide in Dahschur. Mitteilungen des Deutschen Archäologischen Instituts. Abteilung Kairo 50: 35-42. Forstner-Müller, I., C. Jeuthe, V. Michel and S. Prell 2015. Grabungen des Österreichischen Archäologischen Instituts Kairo in Tell el-Dab’a/Avaris: A. Das Areal R/III, Zweiter Vorbericht. Ägypten und Levante 25: 17-71. Harrell, J. A. 2016. Varieties and sources of sandstone used in ancient Egyptian temples. Journal of Ancient

6. Conclusions This first overview of ground stone tools from Balat and Elephantine Island shows strong similarities in strategies within the Pharaonic Egyptian culture, though adapted to the different environments. This is at least partly underscored by comparing with the semi-nomadic Sheikh Muftah community at Balat, which is contemporary to the early Pharaonic Egyptian presence in the oasis, though earlier in date to Ayn Asil. Nonetheless, they both shared the same environment and access to raw material, but show differences in 67

Stone Tools in the Ancient Near East and Egypt Egyptian Architecture 1: 11-37. http://www.egyptianarchitecture.com/JAEA1/JAEA1_Harrell. Harrell, J. A. 2012. Utilitarian Stones, in W. Wendrich (ed.) UCLA Encyclopedia of Egyptology. Los Angeles: University of California at Los Angeles. http:// digital2.library.ucla.edu/viewItem.do?ark=21198/ zz002bqsfg. Harrell, J. A. and P. Storemyr 2009. Ancient Egyptian quarries – an illustrated overview, in N. Abu Jaber, E. G. Bloxam, P. Degryse and T. Heldal (eds) QuarryScapes: Ancient Stone Quarry Landscapes in the Eastern Mediterranean (Norwegian Geological Survey of Norway, Special Publication 12): 7-50. Trondheim: Geological Survey of Norway. Jeuthe, C. 2014. Initial results: The Sheikh Muftah occupation at Balat North/1 (Dakhla Oasis). ArchaéoNil 24: 103-114. Jeuthe, C. 2012. Balat X. Ein Werkstattkomplex im Palast der 1. Zwischenzeit in Ayn Asil (Fouilles de l’Ifao 71). Cairo: Institut français d’archéologie orientale. Jeuthe, C. 2017. Balat/Dakhla Oasis: The Sheikh Muftah camps during the Old Kingdom, in B. MidantReynes and Y. Tristan (eds) Egypt at its origins 5. Proceedings of the Fifth International Conference ‘Origin of the State. Predynastic and Early Dynastic Egypt’, Cairo, 13th – 18th April 2014 (Orientalia Lovaniensia Analecta): 165-174. Leuven: Peeters. Jeuthe, C. 2018. The Governor’s palaces at Ayn Asil/ Balat (Dakhla Oasis/Western Desert), in M. Bietak and S. Prell (eds), Ancient Egyptian and Near Eastern Palaces. Vol. I. Proceedings of the Conference on Palaces in Ancient Egypt, held in London 12th – 14th June 2013 (Contributions to the archeology of Egypt, Nubia and the Levant): 125-141. Wien: Verlag der Österreichischen Akademie der Wissenschaften. Jeuthe, C. in print b. Balat XII. The Sheikh Muftah Site (Fouilles de l’Ifao). Cairo: Institut français d’archéologie orientale. Jeuthe, C., V. Le Provost, and G. Soukiassian 2014. Ayn Asil, palais des gouverneurs du règne de Pépy II. Etat des recherches sur la partie sud. Bulletin de l’Institut français d’archéologie orientale 113: 203-238. Kelany, A., K. M. Negem, A. Tohami, and T. Heldal 2009. Granite quarry survey in the Aswan region, Egypt: shedding new light on ancient quarrying, in N. Abu Jaber, E. G. Bloxam, P. Degryse and T. Heldal (eds) QuarryScapes: Ancient Stone Quarry Landscapes in the Eastern Mediterranean (Norwegian Geological Survey of Norway, Special Publication 12): 87-98. Trondheim: Geological Survey of Norway. Kemp, B. J. Ancient Egypt: Anatomy of a civilization. 2nd Edition. London: 2006. Kindermann, K. 2010. Djara – Zur mittelholozänen Besiedlungsgeschichte zwischen Niltal und Oasen (AbuMuharik-Plateau, Ägypten) (Africa Praehistorica 23). Köln: Heinrich-Barth-Institut. Klemm, R. and D. Klemm 2008. Stones & Quarries in Ancient Egypt. London: British Museum Press.

Klemm, R. and D. Klemm 2001. The building stones of ancient Egypt – a gift of its geology. African Earth Sciences 33: 631-642. Kopp, P. 2011. Siedlung und Stadterweiterung westlich des Satettempels, in D. Raue et al., Stadt und Tempel von Elephantine. 37./38./39. Grabungsbericht. Mitteilungen des Deutschen Archäologischen Instituts. Abteilung Kairo 67: 186-198. Kopp, P., R. Colman, A. van den Driesch, E.-M. Engel, H.-W. Fischer-Elfert, I. Forstner-Müller, A. Hood, J. Peters and D. Raue 2018. Elephantine XXIV: Funde und Befunde aus der Umgebung des Satettempels. Grabungen von 2006-2009 (Archäologische Veröffentlichungen 104). Wiesbaden: Harrassowitz. Laisney, D. 2011. Balat IX. Cartographie de Balat (Fouilles de l’Ifao 61). Cairo: Institut français d’archéologie orientale. de Meyer, M. 2007. The Tomb of Henu at Deir el-Barsha. Egyptian Archaeology 31: 20-24. Prell, S. 2011. Einblicke in die Werkstätten der Residenz. Die Stein- und Metallwerkzeuge des Grabungsplatzes Q I (Forschungen in der Ramses-Stadt. Grabungen des Pelizaeus-Museums Hildesheim in Qantir--PiRamesse 8). Hildesheim: Gerstenberg. Stevens, A. 2012. Akhenaten’s Workers: The Amarna Stone Village Survey, 2005-2009, Vol. 1 & 2 (Egypt Exploration Society Excavation memoir 100-101) London: Egypt Exploration Society. Soukiassian, G. 1997. A Governor’s Palace at Ayn Asil, Dahkla Oasis. Egyptian Archaeology 11: 15-17. Soukiassian, G., M. Wuttmann and L. Pantalacci 2002. Balat VI. Le palais des gouverneurs de l’époque de Pépy II. Les sanctuaires de ka et leurs dépendances (Fouilles de l’Ifao 46). Cairo: Institut français d’archéologie orientale. Storemyr, P. 2014. A prehistoric grinding stone quarry in the Egyptian Sahara, in: L. Selsing, L. (ed.) Seen through a millstone (AmS-Skrifter 24): 67-82. Stavanger: Museum of Archaeology, University of Stavanger. Tillmann, A. 2007. Neolithikum in der späten Bronzezeit. Steingeräte aus Auaris-Piramesse (Forschungen in der Ramses-Stadt. Grabungen des Pelizaeus-Museums Hildesheim in Qantir--Pi-Ramesse 4). Hildesheim: Gerstenberg. 4. Winlock, H. E. 1955. Models of Daily Life in Ancient Eygpt (Publication of the Metropolitan Museum of Art, Egyptian Expedition 18). Cambridge-Massachusetts: Harvard University Press. Ziermann, M. 1993. Elephantine XVI: Befestigungsanlagen und Stadtentwicklung in der Frühzeit und im frühen Alten Reich. (Archäologische Veröffentlichungen 87). Mainz: P. von Zabern. Ziermann, M. 2003. Elephantine XXVIII: Die Baustrukturen der älteren Stadt (Frühzeit und Altes Reich). Grabungen in der Nordoststadt (11.-16. Kampagne) 1982-1986 (Archäologische Veröffentlichungen 108). Mainz: P. von Zabern. 68

Documentation: Non-Archaeological and Archaeological Sources in Comparison

69

70

Mill-songs. The soundscape of collective grinding in the Bronze and Iron Age Near East and eastern Mediterranean Luca Bombardieri Keywords: soundscape, collective grinding, figurative documents, coroplastic art, Bronze and Iron Age Eastern Mediterranean

1. Introduction

2. A socializing activity. Grinding genre scenes in the figurative art of the ancient Near East and eastern Mediterranean

Ethnographic as well as experimental studies emphasise that grain grinding with a non-rotating tool similar to the grinding slab is a very intense physical action which involves repetitive motions backwards and forwards, combined with striking, or circular grinding motions in basin-shaped grinding stones (Bennet and Elton 1898; Curwen 1946; Cane 1989). Such an intense action can be rightly considered one of the most time-consuming processing techniques in pre-modern technological contexts.

A set of Near Eastern prehistoric grinding scenes was initially discussed by R. S. Ellis (1995) and then updated by M. Trokay (2000). This evidence can be completed by an interesting set of coeval Cypriote examples (Karageorghis 1991; Bombardieri 2007; 2010) and finally compared with the rich data-set analyzed by L. A. Moritz (1958), who presented the major documents from Greek and Roman figurative art.

Current archaeological debate largely converges in suggesting that grinding, along with spinning, was not only one of the major economic activities within the ancient household but also – more significantly – a primary shared lived experience for the members of the household (or community) responsible for it (Goody 1982; Steel 2016a; 2016b: 87). It is thus not surprising that a large array of grinding scenes also entered the early figurative representation from the Near East and central-eastern Mediterranean.

Among the earliest secure representations of a grinding scene we may mention a clay sealing from Susa, stylistically dated to the Uruk period (Ellis 1995: 402; Bombardieri 2010: 53) (Fig. 1). Two figures are represented on this seal kneeling on the opposite sides of a sloping work table, which can be most likely interpreted as a quern. In the scene the person on the top side of the quern is clearly grinding using a hand-stone, while the second kneeling on the bottom side is collecting the flour in a container (a basin or a large bag). Interestingly, this image suggests that grinding was a cooperative activity since its earliest representation.

In this article, after briefly discussing a set of figurative documents connected with grinding in the ancient Near East and eastern Mediterranean, I will analyze peculiar cases of forced grinding as punishment and then focus on collective grinding as evidence of socialised behaviour and cooperative activities.

A communal grinding scene also appears on one of the decorated bronze bands that adorned the main doors of

Our sources of evidence for collective grinding are three-fold: multiple grinding installations from excavated domestic buildings (houses or households) and public buildings; figurative documents; and literary evidence. Each of these will be discussed before turning to look at what they and other avenues of enquiry, including ethnographic counterparts, might together tell us about the soundscape of collective grinding and the possible association of grinding with rhythm and musical performance.

Figure 1. Susa. Uruk period. Clay sealing with grinding scene (after Ellis 1995: Fig. 1).

71

Stone Tools in the Ancient Near East and Egypt An interesting set of contemporary terracottas with grinding representations comes from Iron Age Cyprus. Among these, a Cypro-Archaic terracotta, said to come from a funerary context at Episkopi or Kourion, is kept in the Cesnola collection at the Metropolitan Museum of Art in New York (Palma di Cesnola 1894: Pl. X.75; Karageorghis 2000: 161) (Fig. 3). As in the Near Eastern examples mentioned above, two figures are involved in the grinding scene represented by this terracotta; on the right a female figure holds with both hands an elliptical implement, pushing it on a large quern that rests inside a trough with upright sides. Dots of black paint on the upper surface of the quern most probably represent corn that is about to be ground (Karageorghis 1989: 88). A second female figure on the left holds a sieve, while a shovel is represented at the far left of the scene.

Figure 2. Balawat/ Imgur-Enlil. NeoAssyrian period. Scene incised on decorated bronze bands, with two soldiers in the Assyrian camp of king Shalmaneser III (after Trokay 2000: Fig. 5).

Two further terracottas come from 6th century BC contexts at Tamassos (central Cyprus) and are now part of the collections of the Cyprus Museum in Nicosia (Karageorghis 2006: 119, no. 88). The better preserved one consists of two figures grinding and sieving corn. The grinding figure stretches both arms forward to hold a large rubber resting on a fixed quern. The second figure sits on the opposite side, engaged in using a round punctured sieve. Unlike the example held in New York, there is no trace of painted decoration on the terracottas from Tamassos.

public and sacred buildings at Balawat (ancient ImgurEnlil) in Assyria (Micale, Nadali 2004). This scene was included in a large narrative in bas-relief dedicated to the military expedition to Syria carried out by King Shalmaneser III in the year 858 BC. Two soldiers of the Assyrian army are here again represented kneeling, one next to the other, while grinding on two sloping querns (King 1915: Pl. 30; Trokay 2000: Fig. 5) (Fig. 2). This scene is part of a larger depiction of daily life within the Assyrian camp, where a series of food processing activities were performed in order to feed the troops (Bombardieri 2010: 54).

The most evident exception to the standard grinding activity is represented by a very peculiar scene carved on one of the Assyrian wall reliefs in the South-West

Figure 3. Episkopi (?). Cypro-Archaic period. Terracotta depicting a grinding scene. Credits: The Metropolitan Museum of Art, New York. The Cesnola Collection, Purchased by subscription, 187476 (accession number 74.51.1643).

72

L. Bombardieri: Mill-songs

Figure 4. Ninive. SW Palace. Room XXXIII. Particular of the wall relief with two Babylonian prisoners forced to grind their father’s bones (adapted from Layard 1853: Pl. XLV-XLVI; drawn by G. Albertazzi).

Palace at Nineveh. The whole figurative program of Room XXXIII is dedicated to the expedition of King Ashurbanipal to Elam (Barnett 1970: Pl. X); in particular, the celebrated battle conducted on the river Ulai is narrated over three reliefs, where an isolated scene of prisoners’ punishment is also depicted. The cuneiform inscription describes the captives as they are forced to grind their father’s bones (Fig. 4).

this hypothesis. Such installations appear in diverse contexts and different periods in the archaeological record of the Near East and eastern Mediterranean (Bombardieri 2010: 46-48). A comprehensive discussion of these cases is beyond the scope of this article, but it is worth mentioning that multiple-quern installations are documented in Early Bronze Age through Iron Age domestic contexts (e.g., EBA Tell Bderi, Pfälzner 1996: 120, EBA-MBA Hama, Fugmann 1958: 30, Fig. 29b, IA II-III Tell Barri, Bombardieri 2010: 47-48), as well as in public buildings in Mesopotamia, Syria and the Levant. The presence of such multiple grinding installations is strictly connected with special food preparation in dedicated spaces of palatial buildings, as attested in the West Unit of the Early Syrian Royal Palace G at Ebla (Matthiae 2013: 53, Fig. 2.14) and much more clearly in the Old Syrian Western Palace, where 16 querns where located over low benches on three sides of the same large room L.3135 (Matthiae 1989: Pl. 88).

According to Reade (Reade 1979: 99) it could be a sort of ‘evocative’ scene and the prisoners may be the sons of the Babylonian King Nabushum-eresh, who were compelled to grind up their father’s bones in Nineveh. Grinding up paternal bones is also indicated as a ritual punishment in Hittite culture, where it is documented in the so-called Soldier’s Oath: ‘[…] so let the oath deities seize the man who transgresses this oath […], mill his bones, apply the heat to him, and grind him up’ (Hoffner 1974: 134).

An apparently different situation is documented in Bronze Age Cyprus. At Early and Middle Bronze Age Marki Alonia and Ambelikou Aletri, as well as in the Late Bronze Age Apliki Karamallos and Arediou Vouppes, grinding stones were mostly found individually and always set at floor level or on low benches (Webb 2016: 381; Webb and Frankel 2013; Steel and McCartney 2008; Kassianidou 2007).

3. Multiple grinding systems between ritual and sociality With the sole exception of the grinding captives at Nineveh, who can be ritually linked to a specific event, all the scenes described above indicate that grinding is likely to produce social interactions in cooperative contexts. Indeed, grinding is a social task in ancient food preparation; it is carried out on a daily basis by necessity and it never appears to be performed by a single person for his or her own purposes as shown above. There is little doubt that these are elements of a social activity. Archaeologically, the presence of benches supporting fixed sets of querns strengthens

Multiple grinding on a raised bench was clearly not a routine domestic practice (Knapp 2013: 288). At Sotira Kaminoudhia one of the main features of a unit identified by the excavator as a ‘cult place’ was a waisthigh grinding platform with two sets of grinding stones 73

Stone Tools in the Ancient Near East and Egypt in situ and room for two more (Swiny 2008). A similar bench with a single very large quern is documented at Middle Bronze Age Erimi Laonin tou Porakou (Bombardieri 2017: 51). The body of evidence provided by figurative art confirms that grinding could involve several people at the same time in a collective work session. In particular, the Cypriot coroplastic art of the Bronze Age appears to be highly informative; terracotta figurines were particularly rich in genre scenes, especially those concerning daily food preparation. Among these are some representations of collective grinding on raised benches. This allows us to suggest that the modelled scenes show preparations for a singular communal event, for which cereal was processed on a special bench using a multiple grinding system (Webb 2016: 381). 4. Grinding and performance in the Coroplastic Art of Bronze Age Cyprus The coroplastic art of Bronze Age Cyprus reaches what Karageorghis (1991: 123) has defined as its finest expression in the so-called scenic compositions typical of the Early and Middle Bronze Age periods. A number of Red Polished ware vases and terracotta models with scenic compositions provide us with a data-set of complex collective human representations and genre scenes. The scenes are composed of figures attached to vases or forming free-standing terracotta groups with analogous subjects, normally taken from everyday life.

Figure 5. Provenace unknown. RP amphora with modelled complex scenic composition. Musée National de Céramique at Sèvres (after Morris 1985: fig. 493).

or trough, leaning forward with their arms stretched forward to hold a grinder in the trough. To the left of this group a fragmentary figure stands in front of a basin/mortar which contains a lumpy substance. A further female figure stands on the extreme left of the scene.

An unprovenanced EC IIIA large Red Polished ware amphora now in the Musée National de Céramique at Sèvres (France) is relevant to our discussion (Stewart 1962: 278; Karageorghis 1991: 125, SC3) (Fig. 5). Numerous figures modelled in the round were attached to this amphora, depicting various scenes associated with domestic tasks, such as pottery-making and breadmaking. On side A, a female figure is represented seated in the middle of the neckline, leaning against the neck. She holds a swaddled child in her right arm. In front of this woman, on a lower level on the shoulder of the amphora, there is a group of three female figures. A fourth figure on the extreme right is now detached. These women are represented kneeling and bending forward with both arms on the inner border of what appears to be an elliptical trough. To the left of the group another female figure stands in front of a conical mortar fixed on the ground, while she holds with both the hands a long pounder or pestle (partially missing).

A comparable composition also appears on an unprovenanced large Red Polished spouted bowl, formerly part of the Desmond Morris collection at Oxford (Morris 1985: Pls. 292-302; Karageorghis 1991: 117, SC1). On side A of this RP bowl a seated (female?) figure is attached below the rim and leans against one side of the spout. This figure is holding an unidentifiable object in her arms. Near the leaning figure, two groups respectively of two and three human figures stand in front of two similar ledges with conical lumps. Another standing figure completes the scene, depicted holding a long pestle and crushing something in a basin-shaped mortar. A further parallel for this scenic composition can be observed in a fragmentary Red Polished ware bowl probably from the area of Marki and now in the Cyprus Museum (des Gagniers and Karageorghis 1976: Pl. XI: 6; Karageorghis 1991: 120-121, SC9). On side A and side B of this bowl figures stand on a horizontal ledge with their arms stretched to touch the bowl surface; a female

An analogous scene can be found on a Red Polished ware deep bowl, coming from an MC II funerary context at Kalavassos (Todd 1986: 183; Karageorghis 1991: 125, SC5). On side A of this bowl a group of six figures modelled in the round is attached below the rim. They all stand behind a semicircular ledge forming a basin 74

L. Bombardieri: Mill-songs figure with an infant against her chest is depicted below the rim near to a group of three figures turned towards the bowl that seem to manipulate a substance on the ledge itself. The central figure is here holding a cylindrical object in the left hand, punctured at the top, that has been interpreted as possibly representing grain (Karageorghis 1991: 121).

The recurrent scenic composition described above is also perfectly represented by a Middle Cypriote terracotta with no provenance, that is part of the collections of the Musée du Louvre (Caubet et al. 1992: 34-35, cat. 15; Karageorghis 1991: 144-145, VII.6) (Figs. 6 and 7). This terracotta depicts seven human figures around a large elliptical trough. The latter appears horizontally subdivided into two sections, one deeper and one raised, serving as a ledge, along which five out of the seven figures are represented from the buttocks upwards, attached to the outer wall of the trough itself. Among these, only two figures are explicitly female, while all of them wear turbans across the forehead and multiple necklaces. They all stretch both arms forward with their hands working on the ledge surface as in the scenic compositions on vases discussed above.

A similar pattern can be detected in these four scenic compositions, where three basic elements appear associated: 1. 2. 3.

a single standing figure crushing something in a basin-shaped mortar; a raised bench (or trough) with several grinders working together; a single woman holding her child, seated in the vicinity.

It is worth observing that at the two extremities of the trough three additional human figures participate in the scene. On the left, there is a standing woman facing the working figures. Unlike the other human figures, she has large flat ears perforated once, and wears three decorated bands around the neck and two further decorated bands on the forehead. The standing woman holds a child on her left arm, who wears bands around the neck and forehead, like its mother.

Discrete elements related to this composite scene recur in a few other Early and Middle Cypriote modelled compositions. On a fragment of a Red Polished ware deep bowl probably from Marki two groups of three figures stand on a ledge, with their arms stretched forward to manipulate a lumpy substance. While they may be bread-makers rather than grinders, they appear to be engaged in a collective work-session (Karageorghis 1991: 121, SC12).

At the opposite right extremity of the trough a second standing figure holds a deep conical basin in both hands. Along the top of the trough between the two standing figures a series of small bowls are located.

We may also observe elements of this recurrent composition on an unprovenanced Red Polished ware deep conical bowl in the Pierides collection at Larnaka (Karageorghis 1991: 120, SC8; Knapp 2013: 336). On side B of this bowl a man and a woman are represented seated. Near the woman’s right hand there is a ledge with a pounder on it, while a cradled infant is visible near the woman and the ledge.

A comparable association of elements in what appears to be a collective grinding scene, also appears in different contexts and periods. The parallel between the Middle Cypriote terracotta in the Louvre and a later

Figure 6. Provenance unknown. Red Slip terracotta. Musée du Louvre, Paris (adapted from Caubet et al. 1992: 34-35; drawn by G. Albertazzi).

75

Stone Tools in the Ancient Near East and Egypt is providing a rhythmic accompaniment for the work performed. 4.1 Grinding, singing and playing: a soundscape? What all these representations seem to imply is that grinding and pounding make characteristic sounds and thus create a recognizable (and evocative) soundscape. These sounds must have been familiar to all members of an ancient household or a larger community. More specifically, grinding was one of those activities in which two spheres of cultural activity could combine: food preparation and singing (Wilkins 2000: 124). Weaving was probably another. Thus, M.-L. Nosch (2014: 95) describes an imagined domestic acoustical landscape created by repeated familiar weaving sounds. These are the memorable sounds that Iphigenia recalls when she regrets that she will never again hear the histos kalliphthonges (‘the beautifully sounding loom’ Eur. IT 208-225). The grinders could also orchestrate recognizable repeated percussive rhythms providing the base for a single monodic line, either played or performed vocally.

Figure 7. Provenance unknown. Red Slip terracotta. Musée du Louvre, Paris. Detail of the standing woman holding her child (adapted from Caubet et al. 1992: 34).

Textual evidence about the combination between grinding, singing and playing goes back to the Sumerian period. A Sumerian text is known to be a possible fragment of a song sung by the grinders as they worked (Civil 2006: 121-138). This work-song was named elalu in Sumerian, and has been compared with the so-called alala, sung while ploughing and harvesting.

archaic terracotta from Boeotia in the same museum appears particularly suggestive. In fact, a terracotta from Thebes in the Louvre Museum, dated to the end of 6th century BC, shows four women grinding, while a bearded standing man plays the flute on one side (Blümner 1912: 63-64) (Fig. 8). The grinding process is similarly carried out in a kneading trough divided into four sections by deep grooves, presumably intended for catching the flour. Even if one can question whether the four women represented are grinding (Lindet 1899: 419; Moritz 1958: 31) rather than shaping oblong loaves (Pottier 1899: Fig. 8), there is little doubt that the flautist depicted in this Boeotian clay model

An Assyrian proverb also mentions the connection between grinding and singing: ‘As they say, I do not know the work song, If the millstone gets lost, I will not suffer’ (Stol 2016: 352). The evidence from the biblical narration of Samson’s punishment by the Philistines may suggest an interesting connection between blindness, grinding and music. In Judges, it said that the Philistines:

Figure 8. Thebes, Boeotia. Terracotta. Musée du Louvre, Paris (after Pottier 1899: Fig. 8).

76

L. Bombardieri: Mill-songs ‘seized him, gouged out his eyes […] and he was set to grinding grain in the prison’ (Judges 16:21).

mill-song is also noted by Plutarch at the end of the 1st century AD (Mortiz 1958: 31, n. 2; Beaton 2004: 147), in which the woman who grinds and sings is called ‘foreign’, again possibly to suggest a slave status:

While it has been argued that the mill used by Samson is likely an Olinthus mill, not in use before the end of the 6th century BC (Hasegawa 2012; Bombardieri 2010; Frankel 2003), this passage appears generally consistent with wider evidence from the pre-classical Near East, where female slaves and prisoners of war were commonly used as corn grinders (in Akkadian geme2-kikken), as they constituted service personnel and performed the heaviest menial works (Charvat 2007: 259).

‘I heard a foreign woman singing at the mill when I was staying in Ephesus: “Grind, mill, grind. Of course Pittakos used to grind when he was king of great Mytilene”’ (Plut, Mor. 157e; see Watkins 2000: 124). Interestingly, the Greek historian Alcimus from Sicily, quoted by Athenaeus at the end of the 2nd century AD, refers more generally to bread-making when he mentions ‘kneading to the music of the flute’ (Ath. 518b; see Blümner 1912: 64; Moritz 1958: 31, n. 3).

It is also interesting to note that persons could be deliberately blinded and then put to work as a flourgrinder or as a weaver or a singer in ancient Mesopotamia (Stol 2016: 351). In several early Akkadian and Ur III lists of names blind weavers are associated with blind musicians (Sommerfeld1999: 66, no. 14; Heimpel 2009: 46, no. 6); boys and girls were blinded to be trained as musicians in Mari (Ziegler 2011: 306); and, according to the myth of Enki and Ninmakh, being a singer was the destiny of those born blind (Shehata 2009: 36-39). In this Sumerian myth the two gods happened to be drunk and they created seven people out of clay, each with a defect; after having realised them Enki and Ninmakh tried together to find a purpose for these freaks and decided that the person who had the fate to born blind was given the destiny of becoming a musician.

Finally, in the Pseudo-Vergil’s Moretum, presumably derived from a Greek original, a man grinds at a circular mill and sings to accompany his constant and repeated motion: ‘Then he called his two hands to the task, dividing the work between them; the left was dedicated to supplying the grain, the right to doing the work, turning in constant motion and moving the mill (the bruised grain fell through under the rapid blows of the stones). At times the left hand came to the aid of her weary sister and took her turn. Soon he began to sing a country song and ease his task with a rustic tune’ (Pseudo-Vergil, Moretum, 24-28; see Watkins 2000: 129).

All these examples seem to point to an association between blindness and grinding, as well as a correlation between grinding and music. Again, the percussive sounds created by repeated work actions, such as grinding and weaving, appear indirectly associated with singing and music.

J. Burckhardt (2002: 198-199) reports the names of thirteen popular traditional flute melodies (nomoi) that were probably variations of a basic type and were distinguished by special rhythms. These flute melodies were played as dance tunes and come down to us from Late Antiquity through the same source that gives us a list of songs arranged by trades or work activities. They include songs sung by women while weaving or spinning wool, or nursing babies; by people trampling out grain, toiling in the fields; and finally, by people grinding grain.

It is also fairly common to find references to blind singers and musicians in the Homeric poems (and – of course – Homer was blind himself), like Demodocus who performed when he was blind in the Odyssey (VIII 44). But, no direct correlation between grinding and singing or playing music can be found in the Homeric epics.

In modern Greek folk poetry at least two traditional songs sung by women to accompany the action of grinding flour are documented by R. Beaton (2004:147). Both bear an evident rhythmical resemblance in their opening to the ancient mill-song reported by Plutarch.

There is some significant evidence for work songs in classical antiquity and, specifically, the singing of women has been interpreted by J. Wilkins (2000: 124) as an aspect of their life which was communal and shared. We also know from literary sources that in Classical and Hellenistic Greece many daily activities and work tasks commonly had musical accompaniment. Men sung herdsmen’s songs, while winnowing songs and mill-songs were sung by women (Ath. 618d, 619a-d; Aristoph. Thesm. B, fr. 352).

ἄλει, μύλα, ἄλει (Plut, Mor. 157e. Tr: ‘Grind, mill, grind’). Ἄλεθε, μύλο μου, ἄλεθε (Politis 1914: 241, no. 234. Tr: ‘Grind, my mill, grind’). Ἄλεσε, μύλε μ’ ἄλεσε σιτάριν καὶ κριθάριν (Petropoulos 1959: 173. Tr: ‘Grind, my mill, grind wheat and barley’).

More specifically, Aristophanes mentions a mill-song sung by a woman while milling barley (Nub. 1358). A 77

Stone Tools in the Ancient Near East and Egypt Some evidence of mill-songs is also documented by E. Lelli (2016: 119) in the modern folk poetry of southern Italy, especially from Alcantara in Sicily and from specific areas of inner Calabria, where an allusive reference to grinding and sexual performance appears more evident than in the Greek counterparts. It is worth mentioning that the ancient original mentioned by Plutarch appears as a direct inspiration for a traditional mill-song from the Aspromonte in Calabria.

a possible specific working soundscape where the complex rhythmic action of multiple grinding was associated with music. Thus, the work group also became an ensemble, the workplace a stage and the work session a performance. While a mill-song performance can be only imagined in the scenic compositions of multiple grinding from Early and Middle Bronze Cyprus (maybe led by the woman holding her child?), a musical session appears explicitly represented on the Archaic terracotta from Thebes in the Louvre, where a flute-player also appears in the multiple grinding scene.

This evidence of a continuous song tradition may be explained as a consequence of the same sounds and rhythms produced by the action of grinding from antiquity to modern traditional societies. As observed in the literary sources, the emphasis is always put on the rhythmic aspect of grinding motions. Fendin (2006: 161) argues that the rhythm could also have been held steady through singing, as in a modern example recorded among the Zuni Indians in the early twentieth century:

Sounds themselves are missing from the archaeological record. Nevertheless, figurative documents and comparable evidence from literary and ethnographic sources, suggest that mill-songs existed in the ancient Mediterranean and Near East. Mill-songs, like modern work songs, possibly featured call and response structures, as they were probably designed to enable the labourers who sang them to coordinate their efforts in accordance with the rhythms of the songs.

‘As there are eight mills in this house, there are 16 grinders. An aged woman, said to be the only one living who knows the two original grinding songs by heart, sits before the mills and leads the grinders in the song’ (Stevenson 1904; Mobley-Tanaka 1997).

As D. Hendy (2013) argues, rhythm and even noise have been comfortingly social since the most distant human past, with sacred places clearly chosen for their acoustic potential, and what can appear to us as noise (like stone striking or pounding) may create a characteristic and evocative soundscape.

5. Conclusions The archaeological and ethnographic evidence converges to indicate that grinding was one of the hardest and most time-consuming tasks in pre-modern technological systems. It is thus not surprising to find grinding associated with punishment in the specific cases mentioned above.

With this in mind, we may conclude this brief article with the words of the celebrated avant-garde composer John Cage: ‘if we try to ignore noise, it’ll drive us nuts; but if we concentrate on it and listen deeply, it’s as fascinating as Bach’.

Does this mean that grinding was considered of low value in such contexts? This may not be the case if the culture recognises its dependence on processed flour, as was true for the vast majority of pre-literate societies and modern primitive communities.

Acknowledgements I wish to thank the editors’ initiative and particularly Andrea Squitieri for his invitation to contribute to this volume. I also express my gratitude to Francesca Iurlaro for her suggestions and to Jenny Webb for her insightful comments on an earlier draft of this paper. Finally, many thanks to Giulia Albertazzi for producing the beautiful original drawings.

The cultural relevance of grinding in the ancient Mediterranean and Near East may also be connected to the social aspect of this activity. While in domestic (or household) contexts each work session normally involved at least two grinders for a long span of time daily, specific spaces dedicated to multiple grinders are documented in public buildings or on more formal (ritual) occasions.

Bibliography Barnett, R. D. 1970. Assyrian Palace Reliefs in the British Museum. London: British Museum. Beaton, R. 2004. Folk Poetry of Modern Greece. New York and London: Cambridge University Press. Bennet, R and J. Elton 1898. History of corn milling 1. Handstone, slave and cattle ‘mills’. London: Simpkin and Marshall Publishers Ltd. Blümner, H. 1912. Tecnologie und Terminologie der Gewerbe und Kunst bei Griechen und Römen. Berlin: De Gruyter.

Our sources of evidence are three-fold: in situ grinding installations, literary evidence and figurative documents. Each of these have been discussed, before turning to look at what they might together tell us about the ‘real’ context of a grinding session. The analyzed evidence suggests a meaningful relationship between grinding and rhythm, indicating 78

L. Bombardieri: Mill-songs Bombardieri, L. 2007. La macinazione nella raffigurazione iconografica a Cipro tra II e I Millennio a.C.: sviluppo della rappresentazione e confronti con il Vicino Oriente e l’area egea. Mesopotamia LXII: 23-36. Bombardieri. L. 2010. Pietre da macina, macine per mulini. Definizione e sviluppo delle tecniche per la macinazione nell’area del Vicino Oriente e del Mediterraneo orientale antico (British Archaeological Reports International Series 2055). Oxford: Archaeopress. Bombardieri, L. 2017. Erimi Laonin tou Porakou. A Middle Bronze Age Community in Cyprus (Studies in Mediterranean Archaeology 145). Uppsala: Åstrom Editions. Burckhardt, J. 2002. History of Greek Culture. New York: Dover Publications. Cane, S. 1989. Australian Aboriginal seed grinding and its archaeological record. A case study from the Western Desert, in D. H. Harns and G. C. Hillman (eds) Foraging and farming. The evolution of plant exploitation (One World Archaeology 13): 171-194. London: Unwin Hyman. Caubet. A., A. Hermary and V. Karageorghis 1992. Art Antique de Chypre au Musée du Louvre: du Chalcolithique à l’époque romaine. Paris: A.G. Leventis Foundation. Charvat, P. 2007. Social in Early Dynastic Babylonia, in G. Leick (ed.) The Babylonian World: 251-264. London and New York: Routledge. Civil, M. 2006. The Song of the Millstone, in G. del Olmo Lete, L. Feliu and A. Millet Albà (eds), šapal tibnim mû illakū. Studies Presented to Joaquín Sanmartín on the Occasion of his 65th birthday (Aula Orientalis Supplementa 22): 121-138. Barcelona: Sabadell. Curwen, C. E. 1946. Plough and pasture. London: Cobbett Press. des Gagniers, J. and V. Karageorghis 1976. Vases et figurines de l’Age du bronze à Chypre: Ceramique rouge et noire polie. Quebec: Presses de l’Université Laval. Ellis, R. L. 1995. Mühle, B. Archäologisch. Reallexikon der Assyriologie und Vorderasiatischen Archäologie 8: 401402. Fendin, T. 2006. Grinding processes and Reproductive metaphors, in A. Andrén, K. Jennbert and C. Raudvere (eds), Old Norse religion in long-term perspectives. Origins, changes and interactions. An international conference in Lund, June 3-7, 2004: 159-163. Lund: Nordic Academic Press. Frankel, R. 2003. The Olinthus Mill, Its Origin, and Diffusion: Typology and Distribution. American Journal of Archaeology 107/1: 1-21. Goody, J. 1982. Cooking, Cuisine and Class. Cambridge: Cambridge University Press. Hasegawa, C. S. 2012. Samson and Mill (Judg 16, 21). Biblische Notizen 152: 29-34. Heimpel, W. 2009. Blind Workers in Ur III Texts. Kaskal. Rivista di storia, ambienti e culture del Vicino Oriente antico 6: 43-48. Hendy, D. 2013. Noise. A Human History of Sound and Listening. London: Profile books.

Hoffner, H. A. 1974. Alimenta Hethaeorum. Food Production in Hittite Asia Minor (American Oriental Series 55). New Haven: American Oriental Society. Karageorghis, V. 1991. The Coroplastic Art of Ancient Cyprus I. Chalcolithic – Late Cypriot I. Nicosia: A. G. Leventis Foundation. Karageorghis, V. 2000. Ancient Art from Cyprus: The Cesnola Collection in the Metropolitan Museum of Art. New York: The Metropolitan Museum of Art. Karageorghis, V. 2006. Aspects of Everyday Life in Ancient Cyprus. Nicosia: A. G. Leventis Foundation. Kassianidou, V. 2007. Ground Stone Tools from Apliki Karamallos, in B. Kling and J. D. Muhly Joan du Plat Taylor’s excavations at the Late Bronze Age mining settlement at Apliki Karamallos: 277-306. Sävedalen: Åström Editions. King, L. W. 1915. Bronze reliefs from the gates of Shalmaneser. London: The British Museum. Knapp, A. B. 2013. The Archaeology of Cyprus: From Earliest Prehistory through the Bronze Age. Cambridge: Cambridge University Press. Lelli, E. 2016. Sud antico: Diario di una ricerca tra filologia ed etnologia. Milano: Bompiani. Lindet, L. 1899. Les Origines du Moulin à Grains. Revue Archéologique 35: 413-444. Matthiae, P. 1989. Ebla. Un impero ritrovato. Torino: Einaudi. Matthiae, P. 2013. The Royal Palace. Functions of the Quarters and the Government of the Chora, in P. Matthiae and N. Marchetti (eds), Ebla and its Landscape. Early State Formation in the Ancient Near East: 49-65. Walnut Creek: Left Coast Press. Micale, M. G. and D. Nadali 2004. The shape of Sennacherib’s camps: strategical functions and ideological space. Iraq LXVI: 163-176. Mobley-Tanaka, J. L. 1997. Gender and ritual space during the pithouse to Pueblo transition. Subterranean mealing rooms in the North American Southwest. American Antiquity 62: 437-448. Moritz, L. A. 1958. Grain Mills and Flour in Classical Antiquity. Oxford: Oxford University Press. Morris, D. 1985. The Art of Ancient Cyprus. Oxford: Phaidon. Nosch, M.-L. 2014. Voicing the Loom: Women, Weaving, and Plotting, in D. Nakassis, J. Gulizio and S. A. James (eds), KE-RA-ME-JA. Studies Presented to Cynthia W. Shelmerdine: 91-102. Philadelphia: Instap Academic Press. Palma di Cesnola, L. 1894. A Descriptive Atlas of the Cesnola Collection of Cypriote Antiquities in the Metropolitan Museum of Art, New York, vol. 2. Boston. Petropoulos, D. 1959. (ed.) Greek Folk Songs, vol. II. Athens (in greek). Pfälzner, P. 1996. Activity Areas and the Social Organization of Third Millennium B.C. Households’, in K. R. Veenhof (ed.), Houses and Households in Ancient Mesopotamia. Papers read at the 40th Rencontre Assyriologique Internationale. Leiden, July 5-8, 1993: 79

Stone Tools in the Ancient Near East and Egypt 117-127. Leiden, Instanbul: Nederlands Instituut voor het Nabije Oosten. Politis, N. 1914. Selection of the songs from the Greek people. Athens (in greek). Pottier, E. 1899. Nouvelles Acquisitions du Louvre (1897). Revue Archéologique 34: 1-14. Reade, J. E. 1979. Narrative Composition in Assyrian Sculpture. Baghdader Mitteilungen 10: 52-110. Shehata, D. 2009. Musiker und ihr vokales Repertoire. Untersuchungen zu Inhalt und Organisation von Musikerberufen und Liedgattungen in altbabylonischer Zeit. (Göttinger Beiträge zum Alten Orient, Band 3). Göttingen: Universitäts Verlag. Sommerfeld, W. 1999. Die Texte der Akkade-Zeit. 1. Das Dijala-Gebiet: Tutub (IMGULA 3/1). Münster: Rhema. Steel, L. 2016a. The social and economic roles played by the women of Alashiya, in S. Lynn Budin, J. M. Turfa (eds), Women in Antiquity. Real Women across the Ancient World: 386-398. London: Routledge. Steel, L. 2016b. Kitchenalia in Bronze Age Cyprus. Gastronomica. The Journal of Critical Food Studies 16/3: 79-90. Steel, L. and C. McCartney 2008. Survey at Arediou Vouppes (Lithosouros), a Late Bronze Age Agricultural Settlement on Cyprus: A Preliminary analysis of the Material Culture Assemblages. Bullettin of the American School of Oriental Research 351: 9-37. Stevenson, M. C. 1904. The Zuni Indians. Their Mythology, Esoteric Fraternities, and Ceremonies. 23rd Annual Report of the Bureau of American Ethnology. Washington, D.C: Smithsonian Institution. Stewart, J. R. 1962. The Early Cypriote Bronze Age, in P. Dikaios and J. R. Stewart, Swedish Cyprus Expedition Volume IV, Part IA. The Stone Age and the Early Bronze

Age in Cyprus: 205-401. Lund: Swedish Cyprus Expedition. Stol, M. 2016. Women in the Ancient Near East. BostonBerlin: De Gruyter. Swiny, S. 2008. Of cows, copper, corners and cult: The emergence of the Cypriot Bronze Age. Near Eastern Archaeology 71: 41-51. Todd, I. A. 1986. Vasilikos Valley Project 1: The Bronze Age Cemetery in Kalavasos Village (Studies in Mediterranean Archaeology 71.1). Göteborg: Åstrom Editions. Trokay, M. 2000. Le materiel de broyage en basalte du Tell Ahmar (Area C, fouilles de 1989-1996)’, in P. Matthiae, A. Enea, L. Peyronel, F. Pinnock (eds), Proceedings of the first International Congress on the Archaeology of the Ancient Near East, Rome, May 1823 1998: 1665-1672. Rome: La Sapienza Università di Roma. Watkins, J. 2000. Food preparation in Ancient Greece: Representations of Gender Roles in the Literary Evidence, in M. Donald, L. Hurcombe (eds), Representations of Gender from Prehistory to the Present: 118-136. London and New York: Macmillan Press. Webb, J. M. 2016. Women at home and in the community in prehistoric Bronze Age Cyprus, in S. Lynn Budin, J. M. Turfa (eds), Women in Antiquity. Real Women across the Ancient World: 375-385. London: Routledge. Webb, J. M. and D. Frankel 2013. Ambelikou Aletri. Metallurgy and Pottery Production in Middle Bronze Age Cyprus. Uppsala: Åstrom Editions. Ziegler, N. 2011. Music, the Work of Professionals, in K. Radner and E. Robson (eds), The Oxford Handbook of Cuneiform Culture: 288-312. Oxford: Oxford University Press.

80

Rotary Querns and the Presentation of the Past Jennie Ebeling Keywords: rotary quern, southern Levant, museums, heritage, identity

1. Introduction

2. Definition

The rotary quern became the primary domestic graingrinding apparatus in the southern Levant within several centuries of being introduced from the western Mediterranean world during the Roman occupation of Palestine. Rotary querns were used to grind grain and other foodstuffs in domestic contexts into the twentieth century CE and, although no longer a routine part of a woman’s kitchen equipment in the region, they have been documented recently in limited use in Jordan. Described in ethnographic studies and depicted in photographs from the nineteenth and early twentieth century, rotary querns are now appreciated as important artefacts of traditional daily life. Many unprovenanced examples can be seen in restaurants, private homes, and antiques shops in Israel, Palestine, and Jordan, and the image of a woman with a rotary quern is used in bakery and restaurant logos and representations of traditional Palestinian life on postcards, posters, and other artwork. They are also featured in daily life exhibits in Israeli, Palestinian, and Jordanian museums and heritage centres along with other ethnographic household objects. Since most of them cannot be assigned a date due to lack of provenance, their inclusion in such exhibits is fraught with problems.

Rotary querns, also called rotary grinding stones, hand mills, etc., consist of two circular stones of varying diameters that fit together: an upper, mobile stone and a lower, stationary stone. The change from a lateral motion required to use a handstone and grinding slab/saddle quern to a rotary motion signalled a great step forward in milling technology and this principle was further adapted to use animal, water, and wind power (Watts 2014: 21). Grain, pulses, and other foods and materials are fed through a hole in the centre of the upper stone and ground by turning it in a rotary motion with the aid of a vertical stick or rod placed into a socket in the upper stone. The crushed grains spill out from between the two stones and this material is collected and the process repeated depending on the desired final product (for example, coarse-ground grain or fine flour). Examples from the southern Levant are usually made of vesicular basalt, although examples are also known in granite, sandstone, beachrock, and other raw materials. 3. Archaeological and Ethnographic Sources Although examples from Europe and the Mediterranean world have been the focus of study for some time (e.g. Childe 1943), Levantine rotary querns have received inadequate attention by archaeologists. The earliest published example of a rotary quern in the region is from first century CE Masada, Israel, and, according to Frankel (2003: 18), rotary querns replaced the Olynthus mill as the primary hand mill in the region in the Byzantine period. Published examples from archaeological contexts are found in site reports dating from the Roman through the Ottoman period (Ayalon, Tal and Yehuda 2013: 273, figs. 12-13; Ebeling 2009: Ground Stone Artifacts Plate A: 4; Fischer and Tal 1999: fig. 10.1:15, 17; Netzer 1991: 290-291, ills. 464-465; Nevo 1991: 26, pl. 2:5; Tzaferis and Peleg 1989: fig. 72:30 and others). Unfortunately, ground stone artefact specialists have not studied them as a single assemblage.

This paper explores the place of the rotary quern in contemporary Middle Eastern culture and some of the issues involved in using unprovenanced, and thus undated, ground stone artefacts to illustrate women’s daily life activities in the recent past. Grinding stones have specific cultural meanings in the present and are valued, regardless of their date and provenance, as symbols of traditional women’s work that resonate with local audiences and tourists. Since archaeologists working in the southern Levant have generally overlooked these artefacts as a topic of serious study, there exists no typology; as a result, their value as artefacts that can inform about daily life from the Roman period through the nineteenth century in the region is currently limited. However, rotary querns persist in illustrations of traditional daily life activities and represent a way of life that is rapidly disappearing.

Therefore, there is much we do not know about changes in rotary quern form and technology over 81

Stone Tools in the Ancient Near East and Egypt their long period of use in the region. Examples displayed in museums and owned by private collectors vary in form and decoration, suggesting chronological and/or regional differences. I attempted to create a chronological typology by examining rotary querns as part of a larger study of traditional bread production in Jordan in fall 2012 (Ebeling 2014a, 2014b; Ebeling and Rogel 2015). Unfortunately, this was impossible due to the lack of archaeological contexts for the approximately fifty examples I examined on display in Jordanian museums, in the collection of the Department of Antiquities of Jordan, and in the houses and gardens of private collectors in Amman. As will be demonstrated below, rotary querns dating to the Roman and Byzantine periods might be included in traditional daily life displays in museum exhibits in the region.

illustrated by Musil (1908: 145-6), Dalman (1902; 1933), Avitzur (1976) and many others. A common trope in early photographs taken in the region is the authentic or staged image of Palestinian women, usually in pairs, operating a rotary quern (Fig. 1). This Orientalist image of the woman at the mill was particularly popular among pilgrims and tourists because of biblical allusions to this activity (Graham-Brown 1988: 151, Photo 9). Since Western viewers saw them as illustrations of New Testament passages, the observable present was conflated with the mysterious and unobservable past. Early archaeologists working in the region staged similar scenes of Palestinian women operating ancient grinding stones and included these photographs in archaeological publications as illustrations of ancient daily life. In doing so, they confused ancient and contemporary traditions and technologies (Ebeling and Rogel 2015: 346).

Information about the recent use of rotary querns in the southern Levant can be found in nineteenth and twentieth century descriptions in travellers’ accounts, ethnographic studies, and photographs and these sources universally describe and depict women operating them. The method of use is described and/or

Although the use of rotary querns is strongly associated with women, who have traditionally been responsible for most food preparation in household contexts (Ebeling 2010: 53; Meyers 2008), the manufacture of rotary querns by men has been documented among nomadic populations in the region. Musil (1908: 145)

Figure 1. ‘Two women at the mill.’ Matson Photographic Collection, Library of Congress, Prints and Photographs Division [LC-DIG-matpc-07552].

82

J. Ebeling: Rotary Querns and the Presentation of the Past

Figure 2. A rotary quern still in use in northern Jordan. Photograph by Dia’a Mazari Gharaibeh.

4. Rotary Querns in Museums and Heritage Centres

reported that the ‘Ammarin in southern Jordan made rotary querns from local stone while others used stone imported from Syria. Glueck (1965: 15-16) found evidence for the manufacture of granite rotary querns at a site north of Eilat, Israel; the presence of medieval and modern sherds led him to suggest that ‘… these mills are more or less modern in origin’ (16). In 1971, Bedouin in central Sinai were documented making rotary querns out of local sandstone (Goren 1999: Abb. 32). These studies bear witness to the continued use of rotary querns in the region through the second half of the twentieth century.

Most rotary querns are featured in ethnographic rather than archaeological exhibits in local museums, and it is likely that tourists might encounter these objects for the first time in museum displays. In summer 2010, I visited twelve university, regional, and private museums and heritage centres in Israel, Jordan, and Palestine in order to document and report on portrayals of gender and nationalism in ethnographic exhibits (Ebeling 2011). Although my initial goal was to document how ancient daily life, especially during the period of the Hebrew Bible, was interpreted in museums, I soon found that daily life exhibits universally represent traditional life in the recent past in Palestine and Transjordan. However, I also found that ancient artefacts, specifically, ground stone tools, were included in some of these ethnographic exhibits. I believe that this gives visitors an inaccurate picture of recent ways of life and conflates the past with the present in the same way that Orientalist photographs did.

More recent ethnographic studies have contributed to our understanding of the persistence of this grinding technology, albeit on a limited scale. Palmer (2002: 177) observed the use of rotary querns to split lentils and break bitter vetch fruits to feed livestock in northern Jordan. During my study of traditional clay ovens in northern Jordan, one rotary quern was observed still in use to grind wheat for semolina (samid) (Fig. 2). Although the date of its manufacture and acquisition is unknown, interviews with family members revealed that this set had been in the family for generations and might be one-hundred years old. Interestingly, the informants related that neighbourhood women borrow it every few days. This is a far cry from the situation only a generation or two ago, before imported flour was widely available, when every rural family likely owned its own rotary quern and relied on it to grind grain to make bread – the staple food in the traditional diet – and other grain-based dishes (see further Dalman 1987: 219-225).

The ethnographic exhibits sometimes include a setting – a reconstructed room in a house, a courtyard, or a Bedouin tent – in which female and male mannequins are situated among artefacts of daily life. In most cases, female mannequins are posed with rotary querns, traditional clay ovens, ceramic vessels, and other implements and installations related to food preparation or with looms. In the National Heritage Museum at the University of Jordan in Amman, for example, a female mannequin is posed behind a rotary quern and surrounded by vessels containing grain, a sieve, and a large ceramic vessel; a threshing sledge 83

Stone Tools in the Ancient Near East and Egypt

Figure 3. National Heritage Museum at the University of Jordan, Amman, Jordan. Photograph by Jennie Ebeling.

hangs on the wall behind her (Fig. 3). Nearby, another female mannequin is depicted kneading dough next to a hearth. Another section of the museum features a Bedouin tent scene; a female mannequin sits in front of it with a churn while a female companion works a horizontal loom. Labels in Arabic and English accompany many of the scenes and items in this museum, although a rotary quern sitting on its own is labelled in Arabic only. Throughout the museum, the objects’ provenance is usually unspecified other than ‘countryside.’ Malt (2005: 66) reports that the collections include purchases

and gifts from the community and donations from students; thus, the antiquity of the artefacts in these displays is most likely unknown. Lamentably now closed, the Joe Alon Museum of Bedouin Culture in Lahav, Israel, featured a similar Bedouin tent scene with a female mannequin sitting behind a rotary quern while bread baked on a saj nearby; another female mannequin stood behind her (Fig. 4). This exhibit, and others in this well-organised museum, was accompanied by detailed information

Figure 4. Joe Alon Museum of Bedouin Culture, Lahav, Israel. Photograph by Jennie Ebeling.

84

J. Ebeling: Rotary Querns and the Presentation of the Past

Figure 5. Museum of Jordanian Heritage at Yarmouk University, Irbid, Jordan. Photograph by Jennie Ebeling.

about the items and their functions in Hebrew, Arabic, and English. As in the National Heritage Museum in Amman, however, the exact provenance is not given, although most of material, which is from the collections of Kibbutz Lahav and a Bedouin museum in the Sinai Mountains, originated in the Sinai and Negev Desert (http://www.joealon.org.il). Again, specific dates were not provided for most of the artefacts and installations that were on display.

Museum in Jordan once contained a nice exhibit with mannequins, including one posed with a rotary quern, but the museum had been completely redone and these exhibits were gone in 2010. The Palestinian Heritage Center in Bethlehem offers something a bit different, as the centre, which is not a museum per se, includes ethnographic exhibits similar to those in the museums in Jordan and Israel described above. In addition to a large and well-appointed Bedouin tent in which visitors are encouraged to dress up in traditional costumes and pose for photos, there is a reconstructed Palestinian living room that includes many objects of traditional daily life, among them a rotary quern (Fig. 7). According to the description on the website, ‘[v]isitors to the Center are welcome to enjoy sitting, drinking coffee, and watching videos on the archaeology, history, and religions of Palestine’ in this room (http://www.palestinianheritagecenter. com/index.php/the-center). Although there are no mannequins in these exhibits, the centre produces postcards and posters featuring Palestinian women posed in this and other settings (Moors 2000; Semmerling 2004). In this image, labelled ‘Palestinian women practicing traditional Palestinian activities,’ a group of young women in traditional embroidered dresses pose with numerous elements of Palestinian heritage, including what appears to be the same rotary quern that is on display in the centre (Fig. 8). The rotary quern, like the colourful embroidery and other traditional objects, symbolises traditional Palestinian lifeways. Since the provenance of the objects is unclear – even the embroidered dresses on display in the centre and printed on postcards

Although similar sorts of ethnographic reconstructions are on display in the Museum of Jordanian Heritage at Yarmouk University in Irbid, Jordan, there are neither mannequins nor much in the way of wall text. Rotary querns and other artefacts and installations of daily life are arranged somewhat randomly in various rooms (Fig. 5), but most visitors would gain little more than an impression of daily life activities given the lack of explanatory information in any language. A faculty member in the Department of Archaeology informed me that most of the material was donated by the students’ families and their provenance and dates were unknown. An exhibit in the Madaba Archaeological and Folklore Museum, Jordan, includes a section with female and male mannequins wearing traditional clothing and posed with artefacts of daily life in several glass cases with minimal explanatory text. Unfortunately, when I visited the museum in 2010 and again in 2012, the female mannequin in the far left display case was gone and the rotary quern alone remained (Fig. 6)! Museum staff could not tell me what had happened to this display. The Salt Archaeological and Folklore 85

Stone Tools in the Ancient Near East and Egypt

Figure 6. Madaba Archaeological and Folklore Museum, Madaba, Jordan. Photograph by Jennie Ebeling.

Figure 7. Palestinian Heritage Center, Bethlehem, Palestine. Photograph by Jennie Ebeling.

and posters include a mix of pieces in director Maha Saca’s vast ethnographic collection and items recently commissioned by Palestinian women who are supported through her efforts – this image mixes clothing, jewellery, and objects from different periods and conveys a timeless sense of traditional Palestinian life rather than a living room in a specific time and place.

unknown and it is possible that artefacts dating to as early as the Roman period are used to represent traditional life in recent times. In two of the museums I visited, ancient artefacts are deliberately displayed among recent ethnographic material, albeit in very different ways. In the Bedouin Heritage Center in Shibli, Mt. Tabor, Israel, several mannequins are posed in a generic daily life scene with bedding, clothing, and tools of daily life. Near this scene sits a small collection of ancient ground stone artefacts, including vessels, mortars, and weights, among recent metal and wood

In all of the examples described above, the date of the rotary querns on exhibit is either imprecise or 86

J. Ebeling: Rotary Querns and the Presentation of the Past

Figure 8. Image from the collection of the Palestinian Heritage Center, Bethlehem, Palestine. Photograph copyright Maha Saca, Palestinian Heritage Center.

Figure 9. Bedouin Heritage Center, Shibli, Mt. Tabor, Israel. Photograph by Jennie Ebeling.

objects in a corner labelled ‘utensils’ in English (but ‘cooking utensils’ in Hebrew and ‘kitchen utensils’ in Arabic) (Fig. 9). The director of the museum confirmed that some of these objects were ancient, although there was no provenance information provided and the casual visitor would probably assume that these artefacts, some of which could very well date to the Bronze or Iron Age, were modern ethnographic items.

A mix of ethnographic and archaeological material is also on display in the Man and His Work Center in the Eretz Israel Museum in Tel Aviv, Israel. This recently-updated pavilion includes detailed exhibits on traditional everyday life in Palestine and Israel and features ethnographic photographs, accounts, and objects in twenty well-organised sections. There are no mannequins here, but photographs are mounted on the walls that show women participating in activities 87

Stone Tools in the Ancient Near East and Egypt like grinding grain and baking bread. The signage in Hebrew, Arabic, and English is generally very good and much of the collection originates in the 1930s, when Shmuel Avitsur began collecting ethnographic artefacts and utensils (http://www.eretzmuseum.org.il/e/121/). A sign at the entrance to the pavilion informs the viewer that most of the objects on display belong to recent generations although a few ancient artefacts are included with an explanation of their dates and origins. The website provides explicit justification for the inclusion of ancient artefacts: ‘The comparison between the old and the new attests to the great resemblance between the way of life in ancient times and that of recent generations, and highlights how minor the changes in work tools and daily artefacts has been over the years’ (http://www.eretzmuseum.org. il/e/121/). Unfortunately, the rotary quern on display – an unusual basalt example that is decorated with raised knobs – lacks a date and is only said to come from the Golan Heights (Fig. 10); it is thus unclear if this piece is considered to date to ‘recent generations’ or if it is much older.

and anthropology students in particular, in order to provide them with full and broad knowledge about Jordanian heritage in order to enrich their national feelings’ (The Heritage Museum brochure). While this is an admirable goal, it is difficult to know how successful this and other museums are in achieving it given the impressionistic character of its ethnographic exhibits. As mentioned above, it is impossible to know exactly which past is being represented in these displays because the specific period these scenes are intended to represent are rarely given. As Malt (2005: 69) notes, the presentation in the National Heritage Museum at the University of Jordan ‘…reflects correlations between natural, demographic, socioeconomic, and cultural facts in a blending of archaeological and ethnographic exhibits.’ As a result, they have a timeless quality that conflates the recent ethnographic past with the ancient past and ultimately provides a generalising picture of traditional ways of life in the region. The deliberate inclusion of ancient ground stone artefacts in primarily ethnographic exhibits in two of the museums further complicates this picture. Contrary to the stated justification for including archaeological materials with ethnographic ones in the Man and His Work Pavilion in the Eretz Israel Museum, there are important differences between the food-related technologies common two hundred years ago and those used two thousand years ago in the region. The lack of serious study of rotary querns, which apparently remained in near-continuous use for nearly two millennia, proves this point: we currently have no idea how this technology changed over this extraordinarily long period of use despite clear morphological differences in the artefacts in this corpus because they have not been studied adequately. The belief that the ethnographic present is a reflection of the ancient past common in travelogues and photographs a century ago seems to persist. Future study of rotary querns with known archaeological provenance will most certainly prove this to be false.

All of the museums and centres I visited display objects from private collections and it is therefore impossible to expect provenance information for every item. However, the signage in these museums and the texts associated with the exhibits should make visitors aware of the problems involved in displaying and interpreting these sorts of collections. This is important simply because the museums and centres explicity include ‘education’ in their mission statements and objectives. For example, the National Heritage Museum at the University of Jordan offers as its first museum objective ‘[t]o educate University students in general, and archaeology

In addition, the inclusion of rotary querns in these exhibits perpetuates stereotypes about traditional male and female activities. Malt, in her study of women working in the museum profession in Jordan and Morocco (2006), cites as an example the ethnographic exhibit of ‘Typical Life’ that used to be on display in the Salt Archaeological and Folklore Museum. In this exhibit, the female mannequins wear work clothes and weave, cook, and grind while well-dressed male mannequins are shown lounging, smoking, and drinking coffee; from her perspective, ‘[t]hese dioramas reinforce subservience and the inequality of women and perpetuate traditional stereotypes (121).’ Since the scenes are undated, these gender stereotypes also seem timeless and support

Figure 10. Rotary quern from the Golan Heights. Man and His Work Center in the Eretz Israel Museum, Tel Aviv, Israel. Photograph by Jennie Ebeling.

88

J. Ebeling: Rotary Querns and the Presentation of the Past popular ideas about male and female activities not only in the recent, ethnographic past, but also in the distant, archaeological, past. Visitors, both locals and foreigners, may come away from these exhibits with their preconceived notions about male and female roles in ‘the past’ reinforced. Although Malt (2006: 121) identified an example of an effort to challenge such notions in the Ethnographic Museum in Tetouan, Morocco when a female curator redesigned a family diorama to depict a fashionably-dressed female mannequin drinking tea in a social setting rather than grinding grain with a rotary quern or performing other tasks of daily life, such a shift was nowhere to be seen in the museums described above. 5. Rotary Querns Outside of Museums Figure 11. Haret Jdoudna Restaurant, Madaba, Jordan. Photograph by Jennie Ebeling.

Many rotary querns without provenance are on display in restaurants and other public places in Jordan as well as in private collections in Jordan and Israel, and the image of the woman grinding grain is still used in logos and signs in restaurants and bakeries. Visitors to Jordan can expect to find rotary querns in the décor in Middle Eastern/ Oriental restaurants frequented by tourists in places like Jerash, Amman, and Madaba. Sometimes these are displayed with other artefacts associated with traditional life, like threshing sledges, sieves, and ceramic vessels. In all cases, these objects are likely intended to lend an air of authenticity to the experience of dining in a restaurant that serves local cuisine. Rotary querns, including one located at the entrance to Haret Jdoudna Restaurant in Madaba, are usually unlabelled and not always treated with respect by customers (Fig. 11).

Rotary querns are also part of the garden décor in Israeli and Jordanian homes and they are available for sale in antique shops from the Old City in Jerusalem to the Eighth Circle in Amman (Fig. 12). The meaning and value of these objects must vary among those who collect them; unfortunately for archaeologists, the provenance of a rotary quern is usually lost by the time it ends up in the private garden of an affluent Ammani (Fig. 13) or Israeli who has no direct connection to those who made and used this essential tool of daily life. For those who are personally connected to the object, however, its meaning is clear. Palmer (2002: 177) noted in her study of food and food identity among fellahin and Bedouin in Jordan that ‘[r]otary querns remain

Figure 12. Beit Al-Turath Handicraft Company, Amman, Jordan. Photograph by M. Rogel.

89

Stone Tools in the Ancient Near East and Egypt

Figure 13. A private garden in Amman, Jordan. Photograph by Jennie Ebeling.

valued family possessions as a symbol of a way of life that has passed.’ The image of the woman at the mill popular in authentic and staged ethnographic photographs from the late nineteenth and early twentieth century lives on in bakery logos and folk art. The Mill Bakery near the Fifth Circle in Amman, for example, includes this familiar image on its sign and packaging (http://jo.jeeran.com/ en/p/the-mill-bakeries-amman/). Even though commercial bakeries are generally owned and operated by men, the image of the woman grinding grain is familiar and conveys authenticity to patrons, some of whom may remember a time when their daily bread was baked by women at home. This image can also be seen in a framed artwork displayed in the Palestinian Heritage Center in Bethlehem and a very similar piece that I purchased from the Arab Orthodox Society in the Christian Quarter in Jerusalem (Fig. 14). These examples of Palestinian art render the familiar image of the woman at the mill in embroidery, a popular expression of Palestinian identity in art. As Malhi-Sherwell writes in her study of the representation of women in Palestinian art (2001), ‘…

Figure 14. Embroidered artwork of a woman with a rotary quern. Photograph by Jennie Ebeling.

90

J. Ebeling: Rotary Querns and the Presentation of the Past female peasants in traditional costume have become the foremost signifiers of Palestinian national identity’ and ‘…artefacts from the peasant home in general, and the woman’s sphere in particular – such as cooking utensils, baskets, and clay jugs – have been fetishised and marked as objects of Palestinian heritage …’ (163). In these works, which were displayed among embroidered objects for sale in both the Palestinian Heritage Center and the Arab Orthodox Society, the artist associates a woman in a traditional Palestinian embroidered dress with a traditional domestic cooking utensil and, in doing so, reinforces both symbols as markers of Palestinian identity.

at the University of Evansville in summer 2010 and a National Endowment for the Humanities Postdoctoral Research Fellowship at the American Center of Oriental Research in Amman, Jordan, in autumn 2012. Thanks to the University of Evansville, the National Endowment for the Humanities, ACOR Director Barbara Porter and the staff and fellows at ACOR for their support, and Dia’a Mazari Gharaibeh for her important contributions to this project. Bibliography Avitzur, S. 1976. Man and His Work. Jerusalem: Carta/ Israel Exploration Society. Ayalon, E., O. Tal and E. Yehuda 2013. A Twelfth-Century Oil Press Complex at the Crusader Town of Arsur (Apollonia-Arsuf) and the Olive Oil Industry in the Latin Kingdom of Jerusalem. Journal of Eastern Mediterranean Archaeology and Heritage Studies 1/4: 259-291. Childe, V. G. 1943. Rotary Querns on the Continent and in the Mediterranean Basin. Antiquity 17: 19-26. Dalman, G. 1987. Arbeit und Sitte in Palästina, Band III: Von der Ernte zum Mehl. New York: Georg Olms Verlag. Dalman, G. 1902. Grinding in Ancient and Modern Palestine. The Biblical World 19/1: 9-18. Ebeling, J. 2014a. Traditional Bread Baking in Northern Jordan Part 1. Video online. Available at < https:// www.youtube.com/watch?v=up7nC_bVD3M> Ebeling, J. 2014b. Traditional Bread Baking in Northern Jordan Part 2. Video online. Available at < https:// www.youtube.com/watch?v=TaVca6KkMZQ> Ebeling, J. 2011. The Presentation of Women’s Lives in Antiquity in Museums in Israel and Jordan. Paper presented at the Annual Meeting of the American Schools of Oriental Research, San Francisco, CA. Ebeling, J. 2010. Women’s Lives in Biblical Times. London: T&T Clark, Int’l. Ebeling, J. 2009. Ground Stone Artifacts, in E. M Meyers and C. L. Meyers (eds) Excavations at Ancient Nabratein: Synagogue and Environs: 284-295. Winona Lake (IN): Eisenbrauns. Ebeling, J. and M. Rogel 2015. The Tabun and its Misidentification in the Archaeological Record. Levant 47/3: 328-349. Frankel, R. 2003. The Olynthus Mill, Its Origin, and Diffusion: Typology and Distribution. American Journal of Archaeology 107: 1-21. Glueck, N. 1965. Further Explorations in the Negev. Bulletin of the American Schools of Oriental Research 179: 6-29. Goren, O. 1999. Traditionelle Handwerksformen der Beduinen in Israel und im Sinai. Jerusalem: Israel Museum. Graham-Brown, S. 1988. Images of Women: The Portrayal of Women in Photography of the Middle East 1860-1950. New York: Columbia University Press.

6. Conclusion Rotary querns are unique in the Levantine ground stone assemblage because they have meaning and value for the contemporary local population. Since recent ethnographic examples are similar in form to archaeological examples from as early as the Roman period, they represent traditional domestic food processing activities generally in ‘the past.’ Valued as a symbol of a way of life that is extinct (or nearly so), rotary querns are kept, collected, and displayed in a variety of contexts in Jordan, Palestine, and Israel. This, I believe, has hindered the serious study of these essential tools of daily life, resulted in inaccurate representations of traditional activities in museum exhibits, and perpetuated popular stereotypes about traditional gender roles. The current state of affairs precludes the creation of a chronological typology of rotary querns using examples in museums and private collections. It may be possible to glean some useful information from those who still use rotary querns or have family heirlooms in their possession; based on my experience documenting traditional clay ovens in Jordan, however, I believe that an ethnographic study would be difficult to undertake. The best hope for ground stone artefact specialists is for archaeologists to continue collecting and publishing rotary querns found in excavations so that chronological and regional variability can be assessed and a typology developed. When this long-lived technology is better understood, it may be possible for archaeologists and others to address some of the problems that I have identified in museum exhibits and other presentations of the past in the region. In the meantime, ground stone specialists must be sensitive to the fact that some of ‘our’ humble ground stone artefacts are powerful symbols of identity for many in the present. Dare we imagine that they were imbued with such meaning in the past? Acknowledgements This research was supported by the Global Scholar program in the Institute of Global Enterprise in Indiana 91

Stone Tools in the Ancient Near East and Egypt Heritage Museum Brochure. Faculty of Humanities and Social Sciences, the University of Jordan. Malhi-Sherwell, T. 2001. Imaging Palestine and the Motherland, in T. Ben Zvi and Y. Lerer (eds) SelfPortrait: Palestinian Women’s Art, 160-166. Israel: Andalus Publishing. Malt, C. 2006. Women, Museums, and the Public Sphere. Journal of Middle East Women’s Studies 2/2: 115-136. Malt, C. 2005. Women’s Voices in Middle East Museums: Case Studies in Jordan. Syracuse: Syracuse University Press. Meyers, C. 2008. Grinding to a Halt: Gender and the Changing Technology of Flour Production in Roman Galilee, in S. M. Subías and M. S. Romero (eds), Engendering Social Dynamics: the Archaeology of Maintenance Activities, 65-74. Oxford: Archaeopress. Musil, A. 1908. Arabia Petraea. Wien: Buchhändler der Kaiserlichen Akademie der Wissenshaften.

Netzer, E. 1991. Masada III. The Yigael Yadin Excavations 1963-1965, Final Reports: The Buildings, Stratigraphy, and Architecture. Jerusalem: Israel Exploration Society and Hebrew University. Nevo, Y. D. 1991. Pagans and Herders: A Re-Examination of the Negev Runoff Cultivation Systems in the Byzantine and Early Arab Periods. Negev (Israel): IPS Ltd. Palmer, C. 2002. Milk and Cereals: Identifying Food and Food Identity among Fellahin and Bedouin in Jordan. Levant 34: 173-195. Semmerling, T. J. 2004. Israeli and Palestinian Postcards: Presentation of the National Self. Austin: University of Texas Press. Wyatt, S. R. 2014. The Life and Death of Querns: The Deposition and Use-Contexts of Querns in South-Western England from the Neolithic to the Iron Age. Southampton (UK): The Highfield Press.

92

Bourgul in Talmudic and Classical Literature, and Today 1

Rafael Frankel Keywords: Bourgul, Talmudic literature, Classical literature, glume wheat, food preparation

One of the many uses of stone mortars and similar objects was for the preparation of bourgul and this is an opportunity to describe how it is produced today and how it was produced in the past.1

4. Cracking the grain usually using a pestle or mallet and a large mortar, but sometimes using a loosely set rotary hand quern. 5. Sieving the grain into four sizes using four sieves. First the complete grains are removed and then the grits are separated into four sizes: bas bulgur (the largest), orta bulgur (the medium size), cok ince taneli (the smallest), and flour.

Among the most interesting and significant aspects of the history of human culture is the regional diversity of culinary techniques and customs, and their continuity over long periods of time (see for example Frankel 1999). This is particularly true for groats, crushed cereals (as opposed to flour, ground cereals). Bourgul is one type of groats. Gordon Hillman (1984: 135-140) described in great detail the pre-industrial methods of producing crushed wheat in Turkey. The product is called Bulgur in Turkish and also in Persian. The type of wheat used is Emmer (triticum diococcum) which is a glume wheat. Spelt (triticum spelta) is another glume wheat but the term ‘spelt’ often refers to glume wheats in general (see note 2). Glume wheats are similar to barley in that the grains remain enclosed in the glume/ husk after threshing. This husk must be removed before crushing or grinding into flour, unlike free- threshing or ‘naked’ wheat.

In the Arabic of the Southern Levant the same product is usually called bourgul and sometimes klili. In the past Nursi wheat was used but today Lachish wheat is used. The wheat is boiled and then dried on the roof for two weeks. Gustaf Dalman (1964: 249-251, 272275, pl. 56) reported that the grain was crushed in an edge-runner similar to that used to crush olives, but today the crushing is done mechanically. The cracked grain is then sieved into four sizes, the largest Bourgul Khishen or Kalila Kabira (large), the second Bourgul Vast (medium) or Kalila Mudavra (round), the third Bourgul Tsajir or Kalila Tsajira (small) and the smallest Tsuritsira (thin). It is then taken to the corn mill to be crushed with a special mill used only for bourgul (Fig. 1). After milling the grain is sieved once again at home. (Fig. 2). Below are the dimensions of the sieves of the Salman abu Yusef family. Note that the first sieve is not like the others, which are made of metal netting, but is made of string.

For the preparation of bulgur, Hillman describes five stages: 1. Cooking the grain and drying it in the sun. 2. Loosening the husk by one of two methods: a) with a wooden pestle or a mallet in a stone mortar; b) with an edge-runner. This device is a large stone wheel turned within a huge shallow stone bowl. It has a long wooden bar long wooden bar which is the horizontal axle for the wheel. One end is pivoted on top of a post set vertically in the centre of the bowl and the other serves as a handle. Similar edge-runners are used for crushing olives (Frankel 1999: 68-75). 3. Winnowing to separate the husks from the grain: when a slight wind is blowing the mixture of grain and husks is thrown into the air using wooden shovels or bowls and the grain is collected on a cloth or mat. 1 

Sieve no.

Diameter

Height

Size of holes in mm

1

64 cm

10 cm

4.5 x 3

2

42 cm

10 cm

2.5 x 2.5

3

36.5 cm

10 cm

1.6 x 1.6

4

30 cm

5 cm

1.4 x 1.4

Pliny the Elder (first century AD. He died in 79 AD in the eruption of Vesuvius that destroyed Pompeii) in his monumental Naturalis Historiae (Vol. V, book XVIII, Ch. XXIX #109-116: 256-263) described the preparation of four different types of crushed wheat. The first is Alica, about which, in another place, Pliny states ‘Alica is peculiarly Roman’ (Alica res Romana est – Vol. VI. book XXII, Ch. LXI #128: 384-385). It was produced from Zea which, as is clear from the text, is a glume

For a Hebrew version of this article see Frankel 2012.

93

Stone Tools in the Ancient Near East and Egypt this description, Pliny then discusses what he calls ‘spurious alica’ (Alica Adulterina) produced from Zea grown in Africa with larger and blacker grains. First, husks were removed and white lime added. Pliny does not mention any crushing but describes the sieving in great detail. ‘They bolt it through a flour sieve. The grain that stays behind in the sieve is called residuary (excepticia) and is the largest in size. That which goes through is sifted again in a finer sieve and is called seconds (secondaria) and likewise the name of ‘sieve-flour’ (cribrarius) is given to that which in a similar manner stays behind in a third sieve that only lets grains like sand through’. Pliny then refers to two more types of crushed wheat, but without going into details and referring only to what is exceptional. The third type (Pliny does not mention it by name) is first cooked, then dried in the sun, then wetted again and finally crushed in a mill (mola). The fourth type was called Tragum and in this case, instead of chalk, milk was added to give a white colour. In the Geoponica (III 7, 8, 9), the late Greek agricultural manual from the 10th century AD, three products are described: Χονδρου [Chondroy], Τραγου [Tragoy], and Πτισανης [Ptisanis]. Owen (1809: 109-110) in his English translation gave the Latin terms: Alica Tragum and Ptisane (Ptisane means barley groats). For the second and third the Greek and Latin are identical but Alica is Owen’s translation. The production of Alica is described in detail. It was made of glume wheat and after the husks were removed the grain was put in boiling water. Then gypsum and sand were added and the grain was pounded and sieved to three sizes. The Geoponica was based on various older sources and this section was probably based primarily on Pliny.

Figure 1. Crushing the grain, Yarka, Western Galilee.

wheat.2 The first stage involved removing the husks/ glumes using a wooden mortar and a pestle which had an iron cup attached to its end. After the husks were removed the grain was crushed using the same mortar and pestle. The grits were then divided into three sizes – minimum, secundarium, grandissimum. Chalk was added to give the Alica a white colour. This chalk was brought from a hill between Pozzuoli and Naples. Following

What all these examples have in common is that the crushed grain was sieved into several sizes, each with a different name, and most were made of glume wheats. The other characteristics varied. Some were cooked and dried in the sun, to some chalk was added, to some sand, and to some milk.

In the Loeb edition of Pliny’s Naturals Historiae Far is always translated as Emmer (e.g., Vol V Book XVIII ch XXIX # 111: 259) but Zea sometimes is not translated and appears as Zea in the English translation (ibid #112: 261). However, occasionally it is translated as Emmer (e.g., ibid vol VI Book XXII ch. LVIII #124: 383). The text does make it clear that Zea and Far are not identical, for example: ‘the places that use zea have not got far’ (qui zea utuntur non habent far ibid ch XIX # 82: 242-243). In the past, glume wheat, which retains its husk after threshing, was regarded as one variety. For example, in a LatinEnglish dictionary both Far and Zea were translated as Triticum spelta Linn. (see Lewis and Short 1879). Today, however, varieties of wheat are defined genetically and it has been shown that nearly all the main varieties of wheat have sub-varieties that are glume wheats. Thus it is clear that Far and Zea were different varieties of glume wheat. For discussion of varieties of wheat, see Zohary and Hopf 1994: 18-54. 2 

We can now turn to the Hebrew literature. The Hebrew term for glume wheat – kusemet appears in the Hebrew Bible three times) Exodus 9:32; Isaiah 28: 35; Ezekiel 4: 9) but in no case in connection to crushing. However, two terms in the Bible apparently refer to crushed grain. The first is – geresh3) Leviticus 94

R. Frankel: Bourgul in Talmudic and Classical Literature, and Today Zippori [Sephoris] (Jerusalem Talmud Pesahim ch 4: 1: 30 column 4). However, the most interesting passages are those referring to the three terms that usually come together: hilqa tragis tisane clearly corresponding to the Latin terms Alica, Tragum, Ptisane. They are mentioned several times (e.g., Mishnah Makshirin 6: 2, Babylonian Talmud Nedarim 55b). In one passage, they are included in ‘boiled material’ (Jerusalem Talmud Nedarim ch 6: 2: 39, column 3) showing that they were cooked before crushing similar to modern practice in Turkey and Galilee. However, the most detailed passage is in the Babylonian Talmud (Babylonian Talmud Mo’ed Katan 13b: ‘The Groats pounders make Hilka Tragus and Tisana privily for the requirements of the festival… Abaye explained Hilka means one [grain broken] in two, Tragis one into three Tisana one into four. When Rabbi Dimi came [from Israel] he said: all these are Kuntha (Aramaic for glume wheat)’. In the same passage it is explained ‘why is it called Hilka? because it has had its tunic (Haluk – the husk) taken off ’.

Figure 2. The four sieves of the Salman abu Yusef family from Yarka, Western Galilee.

Rabbi Dimi’s statement that they are all glume wheats is important but the question that arises is how to regard Rabbi Abaya’s explanation. Did the Latin terms change their meaning when they were adopted into Hebrew?4 Although this is very surprising exactly the same explanation appears in the Jerusalem Talmud: ‘Hilka to two, Tragis to three, Tisani to four’ (Jerusalem Talmud Sotah ch 2: 1: 17, column 4). It is also interesting to note that if Owen’s translation is correct in the Geoponica the three products also appear and in the same order. It must be stressed that that we have discussed so far refers to the eastern Mediterranean. In the West, groats are mainly used in porridge and as a result, for example in a book devoted to food in antiquity, the chapter heading of the section discussing many of the texts discussed here is ‘Braying of Porridge Grains’ (Thurmond 2006: 32-37). However, as a result of recent migration to the west, it is possible today to buy Bourgul – Bulghur/Bulgur in western countries. These terms did not appear in the Shorter Oxford English Dictionary of 1936, but do appear in the New Shorter Oxford Dictionary of 1993. Western products have also moved eastwards and in Israel today porridge takes its name from Quaker Oats and is usually called qvaker.

2:14, 16) which in the Vulgate, the Latin translation appears as farris fracta. The second – ripot) Samuel II 17: 19; Proverbs 27: 22) is translated in the Vulgate as ptisane and in Proverbs is mentioned with a mortar and pestle. In Talmudic literature, there are many passages that hint at a connection to the processes described above. For instance, in the description of the preparation of the Omer as a meal offering in the temple (Mishnah Menahoth 10: 4 – it was, however, barley not wheat), the husks were apparently removed in two ways: ‘they used to beat it with reeds and the stems of plants that the grain should not be crushed’ and ‘and then they put it into a hollow tube, wherein were holes so that the fire might prevail over all of it’ then ‘they put it in a grist-mill and took therefrom a tenth (of an Ephah) which was sifted through thirteen sieves’.3 The thirteen sieves are probably an exaggeration to distinguish the temple offering from the usual three sieves. As regards the Grist–Mill, David Eitam (1996) suggested that it was an edge-runner similar to that used to crush olives. ‘Groats Makers’ probably crushed the grain using mortars and pestles while the ‘Grist Millers’ used mills. They are mentioned together in the Mishnah (Moed Katan 2: 5), and in the Jerusalem Talmud it is stated that the former were from Akko and the latter from

H. M. Lazarus (1938: 76 ns. 3, 4, 5) suggests that Rabbi Abaya’s explanation is popular etymology, Hika’ is derived from divide [into halves]; Tragus from the Greek τριχως in three parts’, tisan Greek τισχαρα four. Lazarus however ignores the fact that products like these were divided into three sizes. He also does not refer to the Latin origin of the terms. 4 

For references to the Mishnah see Danby’s translation – Danby 1938. 3 

95

Stone Tools in the Ancient Near East and Egypt Conclusions

Frankel, R. 1999. Wine and Oil Production in Antiquity in Israel and Other Mediterranean Countries. Sheffield. Frankel, R. 2012. Bourgul in Talmudic and Classical Literature, and Today. Cathedra 14: 5.7-14 (Hebrew – English Absract: 213). Geoponica. 1895. H. Beckh (ed.) Leipzig. Hillman, G. 1984. Traditional Husbandry and Processing of Archaic Cereals in Recent Times. Part 1: The Glume Wheats. Bulletin on Sumerian Agriculture 1: 114-152. Lazarus, H. M. 1938. Mo’ed Katan translated to English with notes glossary and indices in The Babylonian Talmud Seder Mo’ed Volume IV [ed. I. Epstein] London: Soncino Press (separate pagination for each tractate). Lewis, Ch. T. and Ch. Short 1879. A Latin Dictionary Founded on Andrew’s Edition of Freund’s Latin Dictionary, revised enlarged and in great part rewritten by Ch. T. Lewis and Ch. Short. Owen, T. 1805. (trans.) ΓΕΩΠΟΝΙΚΑ Agricultural Pursuits. London. Plinius H. N. Naturalis Historiae Loeb Classical Library, Harvard University Press, 1938-1962, 37 books in 10 volumes, each book divided into both chapters and paragraphs. Thurmond, D. J. 2006. A Handbook of Food Processing in Classical Rome. Leiden and Boston. Zohary, D. and M. Hopf 1994. Domestication of Plants in the Old World. Oxford.

To sum up, the cereal food discussed here, Bourgul/ Bulgur/Alica/Tragum etc., is a fine example of cultural continuity spanning from the Roman period (and probably earlier) till today combined with marked regional diversity. The preparations varied slightly but in all our examples crushed grain was sieved into three or four sizes, each with a different name. Glume wheat was used in almost every case. The main variations in method involved preparing the grain by cooking and drying in the sun, or adding other substances like chalk or sand. A degree of regional diversity could be observed in the sub-types as cooking the grain was particular to the eastern Mediterranean and adding substances was preferred in the central Mediterranean. Additional research, especially of traditional food preparation in various countries, could clarify the picture further. Bibliography Dalman, G. 1964. Arbeit und Sitte in Palästina III. Hildesheim (Gûtersloh 1928-1943). Danby, H. 1938. The Mishnah. Oxford. Eitam, D. 1996. Groat-Makers Mill – A Device for making Groats in Roman Israel. Judea and Samaria Research Studies – the sixth conference Kedumim: 191-202 (Hebrew with English abstract).

96

Wine and oil presses in the Roman to Late Antique Near East and Mediterranean: Balancing textual and archaeological evidence Tamara Lewit and Paul Burton Keywords: Oil production, wine production, Cato, Pliny the Elder, Hero of Alexandria, Roman technology, late antique agriculture

1. Introduction

finally, by a more advanced ‘direct screw’ press (Fig. 3). Thus White posited an evolution, beginning with ‘The simple Catonian press with pulleys, ropes and manually operated capstan’, which is described as ‘not the most efficient way of doing the job’; after which ‘The first definite advance … dated a century later than Cato … was a spar fitted with a screw-thread’; and finally ‘The next advance was to do away with the lever altogether, using the screw to apply direct pressure …’ (White 1984: 68-70).1 More than three decades later, this sequence has become entrenched in historiography, for example in the recent Engineering and Technology in the Classical World: ‘The first development … was the addition of a winch … wrapped around a drum … typified by the lever-and-drum press described by Cato. … The second major innovation was the application of the screw. … It appeared in practical use … by the first century B.C.’; ‘[B]y the mid-first century AD, the beam was being discarded altogether in favour of the screw alone’ (Curtis 2008: 383). Schneider (2007: 157, 167; 2009) acknowledges that ‘technological change remained… spasmodic and dependent on specific contexts’ but does not substantially challenge the traditional view of a progressive evolution from lever and drum, to lever and screw, to direct screw, and makes no mention at all of publications referring to the Near East or the important archaeological work from the 1990s onwards.

From the Bronze Age to the 20th century, one of the most important and widespread uses of stone installations in the Near East was for the production of olive oil and wine, food staples of this region and indeed the entire Mediterranean by the Roman period. Over the past three decades, substantial archaeological finds pertaining to oil and wine installations of the Roman period and Late Antiquity have been uncovered in the Near East and analysed in detail by archaeologists, constituting a significant body of evidence for this important activity of the ancient rural world (Callot 1984; Hadjisavvas 1992; Amouretti and Brun 1993; Eitam and Helzer 1996; Frankel 1999; Kingsley 2002; Brun 2003, 2004; Ayalon, Frankel and Kloner 2009; Waliszewski 2014). These archaeological finds have revealed a wide variety of technological practices and solutions to the problems of physical pressure, capacity and ease of use presented by processes of oil and wine production. While relatively simple stone installations such as treading floors for grapes are very widely attested throughout antiquity, a common practice of the Roman era was the use of a more complex stone press. This mechanism was used for both wine and oil production, in order to extract further liquid from the fruit, following the initial treading of grapes or crushing of olives. The implications of the most recent archaeological evidence and analysis, however, have yet to be incorporated into the historiography of Roman technology. This rich new body of evidence, moreover, requires a fresh review of the literary evidence, which has been for so long the basis of historical accounts. Following the seminal account by White in 1975, which relied heavily on that of Drachmann (1932) and preceded the archaeological investigations outlined below, Roman press technology has been conceived as following an Empire-wide pattern. This pattern has been presented as a three-stage linear progression of ever-improving presses, occurring in the late 1st century BC to mid1st century AD. According to this view, a simple ‘lever and drum’ press (Fig. 1) was replaced by more complex and superior ‘lever and screw’ mechanisms (Fig. 2) and,

2. Archaeological evidence This picture of technological evolution, however, is completely at variance with the archaeological evidence for the Roman to late antique period, especially from the Near East. Many Near Eastern presses were of a ‘lever and weights’ type (Fig. 4) which has been virtually ignored in the historiography of Roman period technology.2 These Near Eastern presses employed several rectangular, cylindrical or bell-shaped weights, usually of around 600-800 kg Emphasis added. Curtis 2001: 386 mentions the Near Eastern lever and weights press in a few lines, only to say that it ‘can be dismissed quickly. … Although frequently found in the Near East, they are quite rare in the West’. 1  2 

97

Stone Tools in the Ancient Near East and Egypt

Figure 1. Lever and drum press, reconstructed according to Cato’s description in 1996 under the supervision of J.-P. Brun at Beaucaire (Gard), copyright Mas des Tourelles www. tourelles.com. Reproduced with permission.

Figure 2. Traditional lever and screw press, with a mobile weight, similar to that used in ancient Northern Syria and elsewhere in the Levant. Like the direct screw press described by Pliny, this press has a circular plate or ‘tympanum’ on top of the fruit. Reconstructed out of elements from different sites: 6 m oak beam (at least 150 years old) and limestone screw-weight from the village of Iksal in the Galilee; c. 4m oak screw from the Galilee; almond wood nut plate from the village of Sanur in northern Samaria. Collection of Eretz Israel Museum, Tel Aviv, Israel. Reproduced with permission.

Figure 3. Traditional direct screw press, similar to that used in ancient southern Levant and elsewhere. Probably from Western Galilee. Upper nut beam made from Atlantic Terebinth, with oak screw. Reconstructed stone piers. Collection of Eretz Israel Museum, Tel Aviv, Israel. Reproduced with permission.

98

T. Lewit and P. Burton: Wine and oil presses in the Roman to Late Antique Near East and Mediterranean

Figure 4. Reconstruction of the lever and weights press at Oilery FVIII, Chhîm, Lebanon, by Marek Puszkarski. Reproduced from Waliszewski, T. (2014). Elaion. Olive Oil Production in Roman and Byzantine Syria – Palestine, PAM Monographs 6. Warsaw, Fig. 43, with permission.

and approximately 1 m in height (Fig. 5), which were raised by winches attached to the lever. The fruit placed under the lever was compressed by these weights. The popularity and effectiveness of these presses are reflected in their regular appearance in Near Eastern contexts throughout the Roman period and up until Late Antiquity (Fig. 6). This traditional press type, little changed from ancient times, continued to be used in the region until the 20th century (Avitsur 1994: 113117; Gulick 1955: 72-73).

Freed 1991; Aydınoğlu and Çakmak 2011; Aydınoğlu et al. 2015). For example, the well-studied presses of the late antique villages of northern Syria were operated by a wooden screw attached to a single (usually cylindrical) stone weight weighing around 1000-1400 kg and measuring 0.8-1 m in diameter (Callot 1984). There were, however, many variants in the shape and design of Near Eastern weight-stones, as well as the methods of supporting and anchoring the lever, and liquid collection systems, which have been catalogued and examined in detail (Frankel 1999; Waliszewski 2014). These two types of press – lever and weights and lever and screw – were not always mutually exclusive, and both sometimes appear at the same site, as at Chhîm, Lebanon (Waliszewski 2014: 426442). Like lever and weights presses, lever and screw types also continued to be used in the Levant up to the 20th century (Avitsur 1994: 117-121).

The lever and screw mechanisms on which so much technological history has focussed are not absent from the archaeological record of the Near East, but they do not follow the straightforward late 1st century BC chronology which has been assumed in historical accounts. They appear to have become widespread in the Levant in particular from the 4th century AD onwards, the period to which most dated examples can be attributed (although precise dating is often difficult: Kingsley 2002: 61-66; Decker 2009: 167-169; Waliszewski 2014: 170-176), as well as in late antique Cyprus (Hadjisavvas 1992) and Cilicia (Rossiter and

The distribution of direct screw presses in the Roman world is hard to assess from the archaeological record because their reliance on wood components makes them difficult to trace (Frankel 1999: 137; Brun 2003: 99

Stone Tools in the Ancient Near East and Egypt 62). Judging by later historical and ethnographic evidence (e.g. Hadjisavvas 1992: 77; Avitsur 1994; Fig. 3), they may have been more widely used than the archaeological evidence suggests. However, a number have been identified in the southern Levant, and in Egypt (Brun 2004: 160, 166, 177, 181, 183, with further references). Like the lever and screw presses, securely dated examples are almost all of a 4th century AD or later date (e.g. Ben-David 2009; Roll and Ayalon 2009; Zerbini 2012: 10-11; Waliszewski 2014: 175, 467-469), as are their depictions in late antique mosaics, such as those at the 6th century churches of Saint Christopher at Kabr Hiram, Lebanon, and of Lot and Procopius at Mount Nebo, Jordan (Brun 2003: 216-217). Where the most intensive archaeological and technical analysis has been carried out, micro-regional patterns in the use of these Near Eastern press types have been discerned. Direct screw presses are more frequently attested in the southern regions of Palaestina than elsewhere in the Levant, and neither such direct screw presses nor the typical Levantine lever and weights presses appears in the villages of Northern Syria (Frankel 1996 and 1999; Waliszewski 2014: 140-152, 166177). There is some evidence that unusual double direct screw presses may have been used in a region of northeastern Palaestina (Ayalon 2015). Shapes of weightstones and their methods of attachment varied from

Figure 5. Typical limestone weight from a Levantine style lever and weights press, probably Byzantine, probably from the Galilee (Collection number MHW83.2011). Collection of Eretz Israel Museum, Tel Aviv, Israel. Reproduced with permission.

Figure 6. View of late 5th-6th century lever and weights press at Oilery FVIII, Chhîm, Lebanon (photo by Kazimierz Kotlewski). Reproduced from Waliszewski, T. (2014). Elaion. Olive Oil Production in Roman and Byzantine Syria – Palestine, PAM Monographs 6. Warsaw, Fig. 43, with permission.

100

T. Lewit and P. Burton: Wine and oil presses in the Roman to Late Antique Near East and Mediterranean one area to another (Frankel 2012), as did the designs of stone liquid collection systems and supporting piers (Frankel 1999: 82-83, 126-130; Waliszewski 2014: 146147, 153, 168).

Hispania and Italy (Peña Cervantes 2010: 214-215 and Table 8; Brun 2004: 27, 29).4 Some screw press weights have been found in Northern Italy but the only dating evidence is a terminus ante quem of the mid-3rd century for one reused in the wall of Verona, and others may even be medieval (Liverani 1987). There is very limited evidence for the use of direct screw presses in Hispania, southern Gaul and Italy (Peña Cervantes 2010: 213214; Brun 2005: 59-60, 116-118; Brun 2004: 49). In the North African provinces of Tripolitania and Tunisia, by contrast, lever and screw and direct screw presses are virtually entirely absent. In this region, exceptionally massive stone-built lever and weights presses were developed around the 2nd century AD, supported by huge stone piers, and with massive stone lintels to counteract the pressure, sometimes reinforced by further courses of masonry (Mattingly and Bruce Hitchener 1993; Mattingly 1996). A large (c. 7-9 m long) lever was weighted by a sometimes immense parallelepipedic stone counterweight (in Tripolitania, often more than 2-3 tonnes in weight), which was raised by a winch that was attached to the weight (as in Gaul), a method still used in this region up to the 20th century (Mattingly 1993: 494-496).

These well-investigated Near Eastern presses and their stone components differ in many – and often significant – ways from those developed in other regions of the Roman Empire. It is in Italy (Frankel 2001; Brun 2004: 24-25, 29, 43), Istria (Matijašić 1993; Brun, 2004: 55-57), Dalmatia (Brun, 2004: 61-63) and at some sites in Gallia Narbonensis (Brun 1986: 144-146, 199-200, 204-217) that we find archaeological evidence for the use of the famous lever and drum press (which White terms the ‘Catonian press’), in which the wooden lever was lowered and raised over the fruit by means of ropes and a winch around a large (approximately 2.7 m long) wooden drum, held in place by substantial posts. This European press remained in use throughout the Roman period, until Late Antiquity and beyond, the most outstanding example being the oil-producing villa of Barbariga (Istria), which operated until the 6th century. It is also represented in the 4th century mosaic at Piazza Armerina (Brun 2003: 214-215). Outside these regions, yet other press types predominated. A type of press weighted with a single parallelepipedic stone weight, usually about 1 m x 0.6-0.8 m and weighing between 700 kg to 1700 kg was the most common press used in Gallia Narbonensis and Aquitania (Brun 1986: 142-219; Brun 2005: 115-116), Hispania (Peña Cervantes 2010: Table 4) and Greece (Foxhall 2007: 188192). In these, a winch was attached to the stone weight, unlike the Near Eastern presses where several stone weights were raised by a series of winches attached to the lever (Frankel 1999: 99-106). Although this type of press was used until Late Antiquity, lever and screw presses were set up at a number of sites in the late 3rd to 4th centuries, sometimes using the earlier stone weight adapted for this new use by the incision of a central hole for the screw, in Hispania (Brun 2004: 291294, 298-299; Peña Cervantes 2010: 214-215 and Table 8; Galve et al. 2012; Uceda 2012; Martín i Oliveras 2012 and 2013), the Moselle (Brun and Karl-Joseph 2001) and Gallia Narbonensis (Brun 1986: catalogue3). Some rare examples have been dated to the 1st to 2nd century in

Thus, what emerges from archaeological studies is a strong pattern of regional variation in the development and construction of stone (and other) parts of press mechanisms, a feature long noted by archaeologists (Frankel 1999: 176-180). While different types did sometimes occur in the same regions, or similar types in different regions, there are clear distributions of press types which typically predominate in each region. These press types were in general used over many centuries, from at least the start of the Roman period (or in some cases earlier) until Late Antiquity or beyond. In the Near East, the wide diffusion of lever and screw and direct screw presses, side by side with the continued use of lever and weights presses, seems to have mainly occurred in Late Antiquity, a phenomenon echoed at some European sites of the 4th or later centuries AD. Clearly the historiographical picture of an Empire-wide linear progression from lever and drum to lever and screw and, finally, to direct screw press types in the late 1st century BC to mid-1st century AD is violently at odds with the archaeological record.

In Brun’s survey of Roman period sites in the Var, all examples of screw press weights or weights converted to use with a screw at dated sites occur where there is late antique or early medieval occupation: Brun 1986: catalogue 7, 24, 26, 56, 58, 61, 71, 74, 81, 88, 95, 96, 99. It is not possible to precisely date these stones, based on the survey evidence. Brun suggests that the screw weights at sites 24, 26 and 58 may be medieval, as is also possible with regard to those found at sites 61 and 81, occupied till the 15th and 11th centuries respectively: Brun 1986: 149-152, 176-178, 196. A number of the screw weights are earlier (approximately 1st century) lever and weight press stones converted for use with a screw at a later date, e.g. at sites 71, 74, 81, 99: Brun 1986: 176-217. Since one can presume that the presses would have been used for a reasonable period of time before being rebuilt, and – where dated – these types of presses appear in the 4th century elsewhere, a date around the 4th century is suggested here. 3 

3. Textual evidence The explanation for this disparity between archaeological and historical analyses lies in the The interpretation of the press at Posto as a screw press is contested, on the grounds that the weight-stone, although cylindrical, has no central hole for a screw: Cotton 1979: 55; Callot et al. 1986: 139-140 n. 14; Frankel 1999: 105, 111; Brun 2004: p. 26-27. Presses at the large winery of the 1st century BC villa of Settefinestre, Etruria, were originally claimed to be screw presses, but there is no evidence for this and they seem rather to have been lever and drum: Brun 2004: 40-41. 4 

101

Stone Tools in the Ancient Near East and Egypt textual evidence for farming technology, and the way in which this textual evidence has been interpreted. White’s discussions, which have been profoundly influential, were based almost entirely on three written texts: those by Cato (De Agric. 18-19), written in Italy in the first half of the 2nd century BC; Pliny (Nat. Hist. 18.74.316-318), written in Italy in the third quarter of the 1st century  AD; and the Alexandrian writer Hero (Mechanica 3.13-21), of uncertain date, surviving only in an Arabic translation from the original Greek.

In this passage, Pliny begins with a brief discussion of whether one or two presses should be used, and notes that length (of the lever and therefore of the whole press), rather than bulk, is what matters to their effectiveness.6 He then describes the mechanisms of different presses. His first press is described as operated by the ‘antiqui’ by means of ropes, leather straps and hand-spakes. It has been universally assumed (starting with Drachmann 1932: 50) that, even though Pliny does not name its distinctive feature, that is, the wooden drum, which Cato calls a ‘sucula’, this press is equivalent to Cato’s lever and drum press. This is a reasonable assumption, given that Pliny used Cato’s treatise on Italian farming as a source.7 Pliny then apparently describes two types of press which he terms ‘Graecanica’, in which a screw (‘coclea’) was used with the lever. Thirdly, he lists a direct screw press, in which the screw pressed directly on a plate laid over the frails of fruit (see Fig. 3), without a lever.

While Cato’s account is detailed and clear, he describes only one type of press, the lever and drum (or ‘Catonian’) press. As discussed above, the archaeological evidence indicates that this was a distinctively Italian press type, although also appearing in Gaul (where it was still in use in pre-modern times, termed the ‘casse-cou’: Lauvergeon 2004), and in nearby Istria and Dalmatia. Thus, the passage clearly reflects Cato’s own geographic context and cannot be used as a basis for the history of press technology throughout the Roman world.

Hero lists three main categories of press, which are different from Pliny’s types: firstly, he lists a lever and drum press, but one which is quite different from Cato’s and Pliny’s, since he adds a weight-stone; then, he describes a lever and screw press, but unlike the ‘arca lapidum’ mentioned by Pliny, his version uses a counterweight of an Eastern Mediterranean type (Frankel 2009: 10-11); finally, he details a variety of single and double direct screw mechanisms (Drachmann 1963: 110-134).

Pliny describes three main types of press, in a passage worth quoting in full here: premunt aliqui singulis, utilius binis, licet magna sit vastitas singulis. longitudo in his refert, non crassitudo. spatiosa melius premunt. Antiqui funibus vittisque loreis ea detrahebant et vectibus. intra C annos inventa Graecanica, mali rugis per cocleam ambulantibus, ab aliis adfixa arboris stella, aliis arcas lapidum adtollente secum arbore, quod maxime probatur. intra XXII hos annos inventum parvis prelis et minore torculario aedificio, breviore malo in media derecto, tympana inposita vinaceis superne toto pondere urguere et super prela construere congeriem (Pliny Nat. Hist. 18.74.317).

4. Reviewing the nature and implications of the textual evidence Although they have been used as the basis of technological history for nearly a century, Pliny’s and Hero’s descriptions raise several problems of interpretation. It is not immediately clear, for example, whether Pliny describes one or two types of lever and screw presses. The crux of the problem occurs in his statement, ‘… ab aliis adfixa arboris stella, aliis arca lapidum adtollente secum arbore, quod maxime probatur.’8 The second aliis is an editorial intervention

Some people press with single [presses], but two are more useful, no matter how large the immense size of the single ones. It is length in these [presses] that matters, not thickness; the long ones press better. Men of old would drag [the levers] down with ropes and leather straps and also with handspakes. Within the last 100 years ‘Greek-style’ [presses] have been devised, with the threads of the rod running throughout the length of the screw; with the star of the pole being fixed in position by some, in some with the pole raising with itself boxes of stones which is especially favoured. Within the past 22 years there has been devised a way, with small presses and in a smaller press building, with a shorter rod directed toward the middle to press plates placed on top of the marc from above with with all its weight and to build a mass above the presses.5

A sound mechanical point, as a longer beam has more leverage. Pressure per cm2 exerted on the fruit is determined by multiple factors such as both the height and length of the lever, the size and number of weights used (if any), height and surface area of frails in which fruit is placed (which can be wide but piled to a lower height, or of narrower diameter but piled high), and heights and position of anchorage and pressing points in relation to other elements. See Waliszewski 2014: 252-258. 7  Cato is listed in Pliny’s table of contents (Book 1) among the Roman authorities he consulted for Book 18, and indeed he is quoted and cited in the book 20 times, but not, unfortunately, at 18.74.317. Elsewhere, Pliny cites Cato on olive processing: ‘Now Cato’s teachings about olives will be discussed’ (‘Nunc dicentur Catonis placita de olivis’ Nat. Hist. 15.6.23). Pliny’s use of antiqui in the passage, however, may allude to his use of Cato De Agric. 18-19 since the adjective well characterises the men who used the lever and drum press described in Cato’s 250-year-old treatise (see infra). 8  This is the version printed in Mayhoff ’s revision of Jan’s first Teubner edition: Jan and Mayhoff, 1967-1970 (Jan had printed manuscript d’s palis … a palis: Jan 1857; see next n.). 6 

The authors’ translation and interpretation of this difficult (and routinely misunderstood) passage is discussed in detail in Burton and Lewit 2019. 5 

102

T. Lewit and P. Burton: Wine and oil presses in the Roman to Late Antique Near East and Mediterranean by one of the editors of the Teubner edition of the text, Karl Mayhoff, and does not appear in any of the surviving manuscripts of Pliny. Some manuscripts (D, F) read ab alis, ‘by means of wings’; others, including later ones (d, E and the subsequent vulgate tradition), a palis, ‘by means of stakes’;9 and a later scribal annotator (F2) has replaced alis with aliis in F. All but the last, in other words, envision a detailed description of different elements of a single press type. In spite of this, Mayhoff is probably right to print aliis, a locative ablative.10 The restored text makes clear (as understood by Drachmann 1932: 52-56, and followed by most subsequent discussions of presses) that two types of press are being contrasted, one type built by some (ab aliis), and in some (aliis) a different, and preferable, mechanism is used. Unlike Cato, who provides a minutely detailed ‘recipe’ for press construction (Asper 2017: 30-31), Pliny is comparing different mechanisms of pressing – with ropes, with screws, with screws lifting weights and with direct screws – and not giving the details of parts with which the presses are constructed, such as ‘wings’ or ‘stakes’.

course of his life. We can also assume that his treatise was not intended to be used as a practical farm manual by those who would have done so. Presses were built by craftsmen and farm workers in rural estates and villages, such as the stone masons (‘leukurgoi’) and carpenters (‘tektones’) named in the 3rd century inscription at the village of Kafr Nabo in northern Syria, who built the oil press attached to a sanctuary (Waliszewski 2014: 56). Hero’s description of presses is not intended as a practical manual to advise such craftsmen. He was writing in the urban environment of Alexandria within a tradition of specialist literature, with the intention of explaining the mathematical and mechanical principles underlying technical implements to an educated elite and urban audience (Fleury 1996, 2005; Schneider 2009). One recent analysis highlights the role of Book 3 within the Mechanica as a kind of ‘working “down”’ or case study of geometricised mechanical principles, attempting to create a systematic ‘map’ of mechanical knowledge (Asper 2017). It is possible that Hero is not even describing press designs which were in current use, but is rather suggesting inventions of his own, which may never actually have been constructed (Frankel 1999: 88). There is no archaeological evidence for the use of several of Hero’s press types, even in Egypt. His initial lever and drum press in particular is of a unique type, combining an Italian style ‘drum’ with a weight-stone of an Eastern Mediterranean type, no examples of which have ever been identified (Drachmann 1932: 50-51, 63-67, 145-151; Frankel 1999: 178). In fact, in the much-ignored final section of the passage on presses, Hero declares that he is not describing the many types of press which were widely used, because these are ‘inferior’ (Mechanica 3.20), thus indicating that his intention is not to give readers a picture of current farming technology was in use, but rather to invent and improve on current practices.

Pliny thus describes four types of press: Cato’s oldfashioned lever and drum press; a lever and screw press with a star-shaped arrangement of radiating hand-spakes that, along with the base of the screw, is fixed to the ground; a lever and screw press with a counter-weight of stones attached to the screw, which rises from the ground as the screw is turned; and a direct screw press, its top weighted down to counter upward pressures. All these types of presses are ethnographically attested and were used in different parts of the Mediterranean up until modern times (Paton and Myres 1898; Drachmann 1932: 53-56, 170171; Amouretti et al. 1984; Frankel 1999: 110-111, 122123; Lauvergeon 2004). Before balancing the testimony of Pliny and Hero against the archaeological evidence, we should first consider the broader questions of the nature of their writings, their audience, and their authorial intent. Hero was an Alexandrian intellectual, a teacher, scientist and inventor; he was not a farmer, and we can perhaps safely assume that he himself would not have operated a wine or oil press or supervised workers who were doing so as part of the normal

The work of Pliny was similarly directed to an elite audience of upper class land owners, like himself (Long 2001: 36, 39-42). Born in AD 23 or 24, he was a member of the equestrian class, and saw military service throughout Rome’s western provinces, serving as a junior officer in Germania Inferior and Superior, and later holding several procuratorships in Hispania Tarraconensis, Africa and perhaps Gallia Narbonensis and Gallia Belgica (Syme 1979: 742-73; Healy 1999: 1822). He ended his career as the prefect of the Roman fleet stationed at Misenum—a sinecure rather than a post involving a great deal of work. If even a fraction of the villas and estates that Pliny’s nephew and sole heir, Pliny the Younger, inherited once belonged to his uncle (two of them lovingly described at Pliny Epist. 2.17 and 5.6), Pliny the Elder was a very wealthy man, potentially with intimate knowledge of matters

These same manuscripts read palis in place of ab aliis earlier in the sentence, which is nonsense, since stakes would not be used to attach the ‘star’ to the screw. 10  This is to be preferred to F2’s ablative of agent, ab aliis, since the participle, adtollente, is active rather than passive. It should be noted here that the Budé edition unaccountably attributes ab aliis to Mayhoff, and aliis to D2 (it is the other way round), ab alis to d (which reads a palis) and the repeated ab aliis to D2 rather than F2 (Le Bonniec and Le Boeuffle 1972: 162). For the many editorial pitfalls of the Budé Pliny, see Reeve 2007: 115-16. Jungst and Thielscher 1957, 108-109 (cited by Frankel 1999: 86-87) emend Mayhoff ’s ‘ab aliis … aliis’ to ‘malis … a malis’, for which there is no MS authority whatsoever. They thus still envision one press type rather than two. 9 

103

Stone Tools in the Ancient Near East and Egypt agricultural. Frustratingly, however, he never mentions direct observation of or participation in farming in his Natural History, referring instead to works like Cato’s De Agricultura, as noted above. Despite this, it is hard to believe, given his indefatigable curiosity about the world around him, that his agricultural knowledge was solely theoretical; Murphy (2004: 4-5) perhaps exaggerates his bookishness.

an eclipse (Dioptra 35) which has been equated with a lunar eclipse which occurred in AD 62 and which Neugebauer (1938) and Drachmann (1972), followed by others, assumed he observed in Alexandria. More recently, however, it has rightly been emphasised that Hero’s discussion of eclipses does not necessarily imply that he himself had witnessed one. Rather, its wording suggests that he is giving a theoretical example of the time at which a hypothetical eclipse might be observed in different locations, using the examples of Alexandria and Rome, and not describing his own personal experience (Montanari 2005, 246; Masià 2015). The only other indications of Hero’s floruit are his mention of Archimedes, which gives a terminus post quem of the 3rd century BC, and the fact that he is quoted by Pappus, giving a terminus ante quem of the mid-4th century AD (Acerbi 2008). It could be added that his lever and screw press is operated with a weight-stone of a type which has been identified archaeologically in the Levant, where it has been found in 4th-7th century contexts (Frankel 1999: 87-88; Eisenberg and Ovadiah 2009; Waliszewski 2014: 148, 437-42, 465). Thus, it is difficult to rely on a 1st century AD dating of Hero for the chronology of press innovations, or for confirming Pliny’s chronology.

Although in his prefatory dedication to the future emperor Titus, Pliny states that his work was written for a lower-class audience, ‘the lowly mob, the common crowd of farmers and artisans’ (‘humili vulgo scripta sunt, agricolarum, opificum turbae’ Nat. Hist. Praef. 6), the context clearly indicates that he is exaggerating in order to flatter his dedicand. Elsewhere (Pliny Nat. Hist. Praef. 13), he apologises for using rustic, foreign and even barbaric words (‘aut rusticis vocabulis aut externis, immo barbaris’). None of these terms—not even ‘rustic’—can be taken to mean colloquial or plainspoken so as to be better understood by the throngs of farmers and artisans. The vast majority of these lacked the kind of literacy required to read such texts. In terms of chronological development, Pliny describes the press operated with ropes as used by ‘men of old’ (‘antiqui’) and asserts that the lever and screw and direct screw types of press were newer ‘inventions’. He gives precise dates for these, stating that the lever and screw types were ‘invented within the last 100 years’ (‘inventa intra C annos’) and the direct screw ‘within the last 22 years’ (‘intra XXII hos annos’), i.e. in the mid to late 50s AD, 22 years before 77-79, when the Natural History was written. These dates have been almost universally accepted, for example by Schneider, who states: ‘Towards the middle of the first century AD a new type of press emerged, where direct pressure was put on the material to be pressed by the rotation of a screw fitted vertically into a wooden frame.’ (Schneider 2007: 157). Pliny’s use of the very exact number ‘22 years’ is puzzling. Some writers sidestep this problem by translating the text simply as ‘20 years’ (e.g. Humphrey, Oleson and Sherwood 1998: 156). The contrast between the vague ‘100 years’ and the precision of ‘22’ suggests that Pliny may be speaking from direct personal experience, and connecting the adoption of this type of press to an important event in his own life (Le Bonniec and Le Boeuffle 1972: 308; Burton and Lewit 2019). Textual corruption is unlikely since the extant manuscripts unanimously read ‘XXII’.

In spite of these problems, the combination of Pliny’s specific upward chronology and Hero’s praise for his direct screw presses as ‘the most perfect’ (3.18) has encouraged among historians the concept of a teleological progression from inferior to superior technologies, although this has not remained unchallenged by archaeologists (Greene 1994; Mattingly 1996). Most recently, the economic historian Bang has called ‘the screw press’ one of the ‘significant… technological improvements’ which ‘bespeak gains in the technological capacity of ancient society’ (Bang 2009: 200). The accounts given by Pliny and Hero have been interpreted as describing a straightforward linear evolution, occurring in the 1st centuries BC/AD, towards ever-improving presses, each of which became dominant in turn: ‘It is a short step from the lever-andweight press … to the simple lever-and-drum press, which became the most widespread type ... at least through the first century BC … The final development in pressing technology … [was the] direct screw press’; ‘By the mid-first century AD, the beam was being discarded altogether in favour of the screw alone (Pliny NH 18.317)’ (Curtis 2001: 387, 393, 2008: 383).11 However, interpretations of these texts to mean that each of the press types improved upon and widely

Hero’s description of direct screw presses has been taken (e.g. by Thurmond 2006: 97) as verification of Pliny’s date for the introduction of the direct screw press in the mid-1st century AD. However, Hero’s writings are of uncertain date. They have been dated to the mid-1st century AD on the basis that he mentions

Cf. Drachmann 1932: 56: ‘The next step forward was to use the screw directly on the mass to be pressed’ (emphasis added). While Curtis (2001: 386) acknowledges that ‘regional variation … in the development of pressing technology is the rule’ and Schneider (2009: 167) that ‘technological change remained in many cases spasmodic’, both still assume processes of evolution from older technology to the ‘development’ of presses which had been ‘significantly improved’ (Schneider 2009: 149). 11 

104

T. Lewit and P. Burton: Wine and oil presses in the Roman to Late Antique Near East and Mediterranean superseded earlier and inferior types ill-suit the archaeological evidence. According to this evidence, the lever and drum press used by ‘men of old’ was the main type used in Pliny’s own region of Pompeii right up to the time of his death there. Although wine presses and even a wooden screw from a fuller’s direct screw press, as well as evidence from wall-paintings, are well-preserved in this region, not a single wine press employing a screw has been discovered here.12 The wine-press portrayed in detail in the House of the Vettii is a lever and drum press (Brun 2003: 212-13). By using the word ‘antiqui’, therefore, Pliny cannot mean to imply that this type of press was ‘obsolete’ and no longer used, as modern historians have interpreted his terminology to mean. Pliny’s use of ‘antiqui’ is perhaps rather an allusion to Cato’s 250-year-old De Agricultura, which he may have consulted for his own description of the press.

traditional press types were not replaced by ‘improved’ press technology. Once again, as with Pliny, nothing in Hero’s account implies obsolescence of older types: on the contrary, different types, even ‘inferior’ ones, continued to be widely used in Hero’s day (whenever that was). 5. Towards a balance between archaeological and textual evidence These observations about the literary evidence lessen their apparent disparity with the archaeological evidence. The notion of a continuous evolution of technology towards more ‘developed’ and superior forms, which widely replace earlier and simpler types, is a 19th-20th century conception, based on the patterns of industrialisation which occurred following the Industrial Revolution. In pre-industrial societies, however, different forms of technology frequently exist side by side and newer forms do not necessarily replace the old (Greene 1994)—precisely as Pliny and Hero attest. A study of presses in 11th-18th century Corsica has shown that various types of press – including lever and weights and direct screw press – were all used side by side for centuries, based on local village traditions and needs (Casanova 1993: 359-378). Ethnographic accounts of press use in the Near East attest that ancient forms of presses continued to be used, according to tradition, well into the 20th century, and only the development of iron-built presses and new power sources (hydraulic and diesel powered presses) in the late 19th and 20th centuries led, slowly, to the obsolescence of traditional types of press (e.g. Avitsur 1994: 113-117, 140; Forbes 1992; Waliszewski 2014: 252258). Gulick describes a Lebanese village of the 1950s in which a traditional lever and weights press (with three stone weights) was still in use, although much more efficient, machine-driven presses were available, because the new presses were ‘apparently, less fun to operate’ (Gulick 1955: 72-73).

Further, Pliny’s own discussion also makes nonsense of the idea that the newest, direct screw press had superseded the older types with levers. The immediately preceding part of the passage provides a discussion of the importance of the length of the press – a factor which is only relevant if a lever press is still being used, since direct screw presses are roughly square, not long. This discussion would be pointless if a leveroperated press were now obsolete and a direct screw press (without a lever) had become the most common type. The passage is also written in the present tense, which can only mean that such presses were still in use when Pliny was writing. He uses the phrase ‘which is especially favoured’ (‘quod maxime probatur’), also in the present tense, in relation to the century-old counter-weighted lever and screw press, indicating that it was quite popular in his day, and had not in fact been superseded by the fourth, direct screw type. Nor does he say that the ‘especially favoured’ type had replaced the other: in fact, he specifically states that the type with a fixed screw is used ‘by some’. Pliny does not say that a direct screw press has rendered the other types obsolete, but only that it is smaller and requires a smaller building (‘parvis prelis et minore torculario aedificio’).

Vitruvius (De Arch. X.1.5-6) provides some insight into this phenomenon in the Roman world, when he states that ‘innumerable sorts’ of machines such as ‘the drums, levers and hand-spakes’ with which to produce oil and wine are used ‘according to custom’ (‘ad usum consuetudinibus’), attesting the importance of tradition within technological practice.14 An important

Hero does not state that his presses are sequential, but rather that aside from those he describes, ‘There are also many other types of press’, whose ‘use among the multitude has been widespread and longstanding’ (Mechanica 3.20).13 Although he labels these latter presses as ‘inferior’, he clearly indicates that

‘… had not there been provision of drums, levers and hand-spakes for pressing, we could not have had the shining olive oil nor the fruit of the vines for our pleasure. … No less innumerable are the types of machines, about which it is unnecessary to speak since they are at hand every day, [things] we frequently have the opportunity to use according to custom’ (‘… sucularumque et prelorum et vectium si non fuisset torculariis praeparatio, neque olei nitorem neque vitium fructum habere potuissemus ad iucunditatem…’; ‘non minusque sunt innumerabili modo rationes machinationum, de quibus non necesse videtur disputare, quoniam sunt ad manum cotidianae, [ea] quae communes ad usum consuetudinibus habent opportunitates’). 14 

On the absence of the screw, Brun (2003: 61) comments ‘aucun des pressoirs retrouvés dans la région de Pompéi n’en était encore équipé en 79 après J.-C.’ 13  The authors rely on the translations of Hero from the Arabic by Carra de Vaux 1893 into French, Nix and Schmidt 1900 into German and Drachmann 1963: 115-126 into English for the first phrase. Our grateful thanks to A. Treiger and R. Turnbull for their generous assistance with the translation of the second phrase. 12 

105

Stone Tools in the Ancient Near East and Egypt Near Eastern parallel to this comment by Vitruvius has been identified by Frankel (1999: 178, 191-192): the Hebrew text (Tosefta Toh. 10.12), in its discussion of ritual laws for the sabbatical year, refers in turn to: ‘those places where it was usual to place [the olives] in a bd’, ‘those places where it was usual to place [the olives] in a byydhd’ and ‘those places [where a] qtkw was usual’. While it is uncertain exactly which mechanism is meant by each term, the passage makes clear that local customs determined the type used.

force’, but hints that they are less powerful than those using a lever, which he calls ‘among the most powerful’. He also notes an important limitation of a direct screw press, that they require more manual assistance since the pressure needs to be renewed at intervals by turning the screw: ‘they are very powerful but their pressure is not strong by continuation. And therefore it is necessary to repeat from time to time the turning and the pressure’. In their favour, he emphasises that they are ‘easy to work’, easy to transport and ‘put up in any place we want’, and in particular do not require expensive construction materials: ‘there is no need for a long straight beam of a hard nature, nor for a large, heavy stone, nor for strong ropes’ (3.18-19). Pliny makes a similar point when he tells us that the direct screw press requires a smaller space and shorter screw. Although his discussion contains less detail, he also tells us that a lever and screw type with mobile counter-weight constructed of a wooden ‘box’ of stones was considered better than one with a fixed screw.

Archaeological evidence also attests that this adherence to local traditions was a strong force within Roman and late antique society, despite the impetus to change offered by Roman transport and communications, widespread trade, and diffusion of techniques through military, veteran and government presence in the provinces (Burford 1972: 64-65; Greene 1992; Lewit, in press). In the Near East, the picture that may be derived from archaeological evidence is one of pervasive local tradition, moderated to some degree, particularly from the 4th and later centuries, by panMediterranean influences. Distinct forms of presses were used in the Near East in contrast to other regions where different types predominated. The various types of screw press display strong variations, rooted in local technical traditions, including different weight-stone designs and anchoring systems, drainage systems and press-bed designs (Frankel 2012). At least part of the reason for this must have been the fact that the stone elements of presses (in both Roman and later periods) were expensive and highly durable items of equipment, which were seldom if ever transported, demanding the mustering of resources (materials, technical knowledge and skilled labour), and used for long periods of time (even centuries), rather than being replaced (Frankel 1999: 176).15 It is thus crucial to note the distinct geographic context of Pliny in Italy and Hero in Alexandria, regions in which, according to archaeological evidence, quite different presses were used in this period.

These points are supported by archaeological and ethnographic indications. Ancient direct screw presses that have been identified archaeologically seem to occur most often (although by no means always) in urban contexts where space was scarce, for example at Jifna (Zelinger 2009), Theadelphia, Karanis, Akoris (Brun 2004: 166, 177, 181), Barcelona and Valencia (Peña Cervantes 2010: 213). The 18th century French investigator Rozier described a direct screw press used in Languedoc and Provence as ‘very convenient… and it occupies little space’, but also observed that a larger lever and screw press known as a ‘pressoir à Martin’ could press a larger amount at once ‘much better’, although it was more ‘costly’ and ‘bulky’.16 The same author also noted, like Hero, the disadvantage of the direct screw presses in that they required constant renewal of pressure through the repeated manual turning of the screw, and to overcome this disadvantage winches were added to the mechanism used in his time (Rozier 1776: 427). Winch mechanisms were for the same reason also used in the early to mid20th century Near East (Avitsur 1994: 127-128, 149150). Medieval and ethnographic records attest the significant disruptions to work posed by the breaking of press parts and accidents, as described by Hero, and of the advantages noted by Pliny of a lever and screw press with a mobile counter-weight over those with a fixed screw, which tended to jam and break (Drachmann 1932: 55-56; Forbes 1992: 99; Lauvergeon 2004). Records from the Near East and elsewhere

Hero tells us that when a press using a rope and winch was used, the hand-spake could break if the number of workers was increased to exert more pressure on the heavy wooden beam, so that ‘it falls down and they are hurt’. He also describes the tendency of the rope to ‘cause some hindrance’ when it does not slide well on the pulley, and concludes that replacing a rope with a screw makes a press ‘easier … and safer’ to use. He describes his lever and screw press design as ‘powerful, solid and secure in output and without much labour’ (Mechanica 3.15). Hero asserts that his direct screw press designs are ‘very powerful’ and would generate ‘lots of

‘Ce pressoir est très commode… et il occupe peu de place; mais je pense qu’on exprime point assez de pâte à la fois, de qui est un perte de temps et une multiplication de main-d’œuvre; que la force de deux ou trois hommes placés à chaque barre n’est pas assez suffisante… Le pressoir à Martin presse beaucoup plus mieux et une plus grande quantité de pâte à la fois. S’il n’étoit si dispendieux et si volumineux, je le préférerois’ (Rozier 1776: 427). 16 

Schneider (2009: 149) acknowledges these factors and that ‘innovations in agricultural technology were not established throughout the empire, but only in individual regions … Thus it was possible to have various types of press simultaneously in use’. 15 

106

T. Lewit and P. Burton: Wine and oil presses in the Roman to Late Antique Near East and Mediterranean attest the convenience and robustness of a screw, which could be carved from any suitably hard wood with hand tools, and could last for a period of about 20 years before being replaced (Avitsur 1994: 117-119; Amouretti et al. 1984: 401).

Hero’s concern is to demonstrate the mechanical superiority of a screw, not to record the practicalities of farming, and he refuses to describe the inferior types of press commonly used and widespread. Yet these were the presses that were used for centuries, attested by the remains of grooved, pierced and hewn stone devices, which, in Pliny’s words, the labour of the ‘common crowd of farmers and artisans’ brought into existence.

6. Conclusions It is only by balancing and critically analysing both the testimony of textual sources and the full body of archaeological evidence that has recently been brought to light that we can arrive at an accurate understanding of presses in the Roman Empire and Late Antiquity, and those of the Near East in particular. The historiographical picture of Empire-wide technical evolution is completely at odds with the evidence of press use in the Roman and late antique Near East, which reveals that quite distinctive types of press were used for many centuries in different regions and microregions of the Mediterranean. Too much emphasis has been placed on change in the 1st century BC to mid-1st century AD, in spite of the evidence, particularly in the Near East, that substantial change occurred only in Late Antiquity, and indeed that in many regions traditional presses continued to be used up to modern times. The conceptual gloss of post-industrial assumptions of a linear evolution of technology from simpler to superior forms has tempted historians to read more into the serial presentation of types by ancient writers than they actually claim. In fact, neither Pliny nor Hero claims that each press type was better than or superseded the one previously described. Rather, they give us insights into the varied problems and advantages which each type presented to their contemporaries. A close examination of their accounts, moreover, suggests that a variety of press types were in current use when they were writing.

Acknowledgements Our grateful thanks for their kind assistance to Gary Ciddor and Miriam Hoffman, and to the reviewers for their helpful advice and comments. Any errors are our own. Bibliography Acerbi, F. 2008. ‘Hero of Alexandria’, in New Dictionary of Scientific Biography. Detroit, 283-286. Amouretti, M.–C., Comet, G., Ney, C. and Paillet, J.-L. 1984. ‘A propos du pressoir à l’huile: de l’archéologie industrielle à l’histoire’, Mélanges de l’École française de Rome: Antiquité 96: 379-421. Amouretti, M.-C. and Brun J.-P. (eds) 1993. La production du vin et de l’huile en Méditerranée. Oil and wine production in the Mediterranean area. Actes du symposium international organisé par le Centre Camille Jullian (Université de Provence-CNRS) et le Centre Archéologique du Var, Aix-en-Provence et Toulon, 20-22 Novembre 1991) (BCH Supp. 26). Paris. Asper, M. 2017. ‘Machines on Paper. From Words to Acts in Ancient Mechanics’, in M. Formisano and P. van der Eijk (eds), Knowledge, Text and Practice in Ancient Technical Writing. Cambridge, 27-52. Avitsur, S. 1994. ‘Olive Oil Production in the Land of Israel. Traditional to Industrial’, in R. Franḳel, S. Avitsur, and E. Ayalon (eds), History and technology of olive oil in the Holy Land. Arlington, 90-157. Ayalon, E. 2015. ‘Twin-Screw Olive-Oil Presses from Israel’, IEJ 65: 100-110. Ayalon, E., Frankel, R. and Kloner, A. (eds) 2009a. Oil and Wine Presses in Israel from the Hellenistic, Roman and Byzantine Periods (BAR IntS 1972). Oxford. Ayalon, E., Frankel, R. and Kloner, A. 2009b. ‘Oil Presses at Qedumim’, in E. Ayalon, R. Frankel, and A. Kloner (eds), Oil and Wine Presses in Israel from the Hellenistic, Roman and Byzantine Periods (BAR IntS 1972). Oxford, 185-189. Aydınoğlu, Ü. and Çakmak, Ü. 2011. ‘A Rural Settlement in the Rough Cilicia-Isauria Region: Karakabaklı’, Adalya 14: 71-102. Aydinoğlu, Ü., Mörel, A. and Kerem, F. 2015. ‘Olive oil production in Kanytellis, Rough Cilicia, in the light of new excavations’, in A. Diler, K. Şenol and Ü. Aydinoğlu (eds), Olive oil and wine production in eastern Mediterranean during antiquity International

Because historians have been focused on the mostly artificially imposed teleology of their accounts, they have missed something of much greater significance. Pliny and Hero record aspects of presses which are not directly archaeologically attested: their ephemeral parts such as the ropes, the wooden screw, the ‘star’ and ‘box of stones’; the existence of all-wooden direct presses, which have been so rarely discerned in European archaeological contexts; and their understanding of the uses, advantages and disadvantages of different types of press. It is vital to consider the authorship and intention of these accounts: there is no reason to think that either Pliny or Hero intended to write a complete record of the press technology used in farming across the Empire up to their own time. Pliny, an upper-class landowner, writes for fellow members of his own elite group about a variety of presses in current use in his own region, southern Italy – and perhaps even on his own estates. 107

Stone Tools in the Ancient Near East and Egypt Symposium Proceedings 17-19 November 2011 Urla, Turkey. İzmir, 51-63. Baldwin, B. 1990. ‘The Date, Identity, and Career of Vitruvius’, Latomus, 49: 425-434. Bang, P. F. 2009. ‘The Ancient Economy and New Institutional Economics’, JRS 99: 194-206. Ben-David, C. 2009. ‘An Oil Press at ‘Ein Nashut’, in E. Ayalon, R. Frankel and A. Kloner (eds), Oil and Wine Presses in Israel from the Hellenistic, Roman and Byzantine Periods, BAR Int S 1972. Oxford, 71-74. Brun, J.-P. 1986. L’Oléiculture antique en Provence. Les huileries du département du Var. Paris. Brun, J.-P. 1993. ‘Les innovations techniques et leur diffusion dans les pressoirs’, in M.-C. Amouretti and J.-P. Brun (eds), La production du vin et de l’huile en Méditerranée; Oil and wine production in the Mediterranean area. Paris, 539-550. Brun, J.-P. 2003. Le vin et l’huile dans la Méditerranée antique. Paris. Brun, J.-P. 2004. Archéologie du vin et de l’huile dans l’Empire romain. Paris. Brun, J.-P. 2005. Archéologie du vin et de l’huile en Gaule romaine. Paris. Brun, J.-P. and Karl-Joseph, G. 2001. ‘La viticulture antique en Rhénanie’, Gallia 58: 165-179. Burford, A. 1972. Craftsmen in Greek and Roman society. London. Burton, P. and Lewit, T. 2019. ‘Pliny’s Presses: the true story of the 1st century wine press’, Klio 101.2. Callot, O. 1984. Huileries antiques de Syrie du Nord. Paris. Callot, O., Reddé, M., and Vallat, J.-P. 1986. ‘Un contrepoids ou pierre d’ancrage de pressoir à Carinola (Campanie Septentrionale)’, Mélanges de l’École française de Rome. Antiquité 98: 129-140. Carra De Vaux, M. 1894. Les Mécaniques ou L’élévateur de Héron d’Alexandrie, Paris. Casanova, A. 1993. ‘Types de pressoirs et types de productions à partir de l’exemple de la Corse à la fin du XVIIe siècle’, in Amouretti, M.-C. and Brun J.-P. (eds), La production du vin et de l’huile en Méditerranée. Oil and wine production in the Mediterranean area. Paris, 359-378. Chibnall, M. 1975. ‘Pliny’s Natural History and the Middle Ages’, in Dorey, T. A. (ed.), Empire and Aftermath: Silver Latin II. London and Boston, 57-78. Cotton, M. A. 1979. The Late Republican Villa at Posto, Francolise. London. Cresswell, R. 1965. ‘Un pressoir à olives au Liban’, L’Homme 5: 33-63. Curtis, R. I. 2001. Ancient Food Technology. Leiden. Curtis, R. I. 2008. ‘Food Processing and Preparation’ in J. P. Oleson (ed.), Engineering and Technology in the Classical World. Oxford, 369- 392. Dar, S. 1986. Landscape and Pattern: An Archaeological Survey of Samaria 800 B.C.E.-636 C.E. (BAR IntS 308). Oxford. Drachmann, A. G. 1932. Ancient Oil Mills and Presses. Copenhagen.

Drachmann, A. G. 1963. The Mechanical Technology of Greek and Roman Antiquity. Copenhagen. Eisenberg, E. and Ovadiah, R. 2009. ‘Byzantine Wine and Oil Presses at Mevo’ Modi’im’, in E. Ayalon, R. Frankel and A. Kloner (eds), Oil and Wine Presses in Israel from the Hellenistic, Roman and Byzantine Periods, BAR Int S 1972. Oxford, 259-263. Eitam, D. and Helzer, M. (eds) 1996. Olive Oil in Antiquity. Israel and Neighbouring Countries from the Neolithic to the Early Arab Period. Padua. Fleury, P. 1996. ‘Traités de mécanique et textes sur les machines’, in C. Nicolet (ed.), Les Littératures techniques dans l’antiquité romaine. Geneva, 45-76. Fleury, P. 2005. ‘La démonstration d’un système dans les textes techniques latins’, Pallas 69: 277-298. Forbes, R. J. 1965. Studies in Ancient Technology III. Leiden. Forbes, H. 1992. ‘The Ethnoarchaeological Approach to Ancient Greek Agriculture’, in B. Wells (ed.), Agriculture in Ancient Greece. Stockholm, 87-101. Foxhall, L. 2007. Olive cultivation in ancient Greece: seeking the ancient economy. Oxford. Frankel, R. 1996. ‘Oil Presses in Western Galilee and Judaea. A Comparison’, in D. Eitam and M. Helzer (eds), Olive Oil in Antiquity. Israel and Neighbouring Countries from the Neolithic to the Early Arab Period. Padua, 197-218. Frankel, R. 1999. Wine and Oil Production in Antiquity in Israel and Other Mediterranean Countries. Sheffield. Frankel, R. 2001. ‘Cato’s Press, a Reappraisal’, in J.-P. Brun and P. Jockey (eds), Techniques et sociétés en Méditerranée. Paris, 313-325. Frankel, R. 2009. ‘Introduction’, in E. Ayalon, R. Frankel, and A. Kloner (eds), Oil and Wine Presses in Israel from the Hellenistic, Roman and Byzantine Periods (BAR IntS 1972). Oxford, 1-24. Frankel, R. 2012. ‘Ancient Technologies: complete vs conceptual transfer’, Tel Aviv 39: 115-126. Frankel, R. 2016. ‘Oil and Wine Production’, in G. L. Irby (ed.), A Companion to Science, Technology, and Medicine in Ancient Greece and Rome, 1. Oxford, 550569. Galve, S. A., Ros, C. V., and Arantegui, J. P. 2012. ‘El complejo de prensado de la villa romana de La Loma Del Regadío (Urrea De Gaén, Teruel)/The Mill Complex in the Roman Villa of Loma Del Regadío (Urrea De Gaén, Teruel)’, Anales de Prehistoria y Arqueología 27: 219-230. Greene, K. 1992. ‘How was technology transferred in the western provinces?’, in M. Wood and F. Queiroga (eds). Current Research on the Romanization of the Western Provinces. (BAR IntS 575). Oxford, 101105. Greene, K. 1994. ‘Technology and Innovation in Context’, JRA 7: 22-33. Gulick, J. 1955. Social Structure and culture change in a Lebanese Village. New York. Haas, C. 2001. ‘Alexandria and the Mareotis Region’, in T. S. Burns and J. W. Eadie (eds), Urban Centers 108

T. Lewit and P. Burton: Wine and oil presses in the Roman to Late Antique Near East and Mediterranean and Rural Contexts in Late Antiquity. East Lansing, 4762. Hadjisavvas, S. 1992. Olive Oil Processing in Cyprus from the Bronze Age to the Byzantine Period. Studies in Mediterranean Archaeology 99. Nicosia. Healy, J. F. (1999). Pliny the Elder on science and technology. Oxford. Höker, C. 2005. ‘Vitruvius’, in H. Canick and H. Schneider (eds), Brill’s New Pauly. Stuttgart, 479-480. Humphrey, J. W., Oleson, J. P. and Sherwood, A. N. 1998. Greek and Roman Technology: A Sourcebook. London and New York. Jan, L. von (ed.) 1857. C. Plinius Secundus Naturalis historia, vol. III. Leipzig. Jan, L. von and Mayhoff, C. (eds) 1967. C. Plinius Secundus Naturalis historia, vol. III. Stuttgart. Kingsley, S. A. 2002. A Sixth-Century AD Shipwreck off the carmel Coast, Israel. Dor D and Holy Land Wine Trade (BAR IntS 1065). Oxford. Lauvergeon, B. 2004. ‘Les grands pressoirs bourguignons pré-industriels: essai de chrono-typologie’, In Situ. Revue des patrimoines 5. http://insitu.revues.org/2447. Le Bonniec, H. and Le Boeuffle, A. (eds) 1972. Pline l’Ancien. Histoire Naturelle Livre XVIII. Paris. Lewit, T. 2007. ‘Absent-Minded Landlords and Innovating Peasants? The Press in Africa and the Eastern Mediterranean’, in L. Lavan, E. Zanini and A. Sarantis (eds), Technology in Transition A.D. 300-650 (Late Antique Archaeology 4). Leiden, 119-139. Lewit, T. 2012. ‘Oil and wine press technology in its economic context: Screw presses, the rural economy and trade in Late Antiquity’, Antiquité tardive: Mondes ruraux en Orient et en Occident I 20: 137-149. Lewit, T. 2017 in press. ‘Invention, tinkering, or transfer? Innovation in oil and wine presses in the Roman Empire’, in P. Erdkamp, K. Verboven and A. Zuiderhoek (eds), Capital, Investment and Innovation in the Roman World, Oxford Studies on the Roman Economy (OSRE). Oxford. Liverani, P. 1987. ‘Termini muti di centurazione o contrappesi di torchi?’, in Mélanges de l’École française de Rome: Antiquité 99: 111-127. Long, P. O. 2001. Openness, secrecy, authorship: technical arts and the culture of knowledge from antiquity to the Renaissance. Baltimore. Martín i Oliveras, A. 2012. ‘Arqueología del vino en época romana: el proyecto Cella Vinaria y el complejo vitivinícola de Vallmora (Teià-MaresmeBarcelona). Nuevas aportaciones a la investigación. The archaeology of wine in the Roman Period: the project Cella Vinaria and the wine production complex at Vallmora (Teià-Maresme-Barcelona). New contributions to research’, in Anales de Prehistoria y Arqueología 27: 113-140. Martín i Oliveras, A. 2013. ‘The CELLA VINARIA Project and Archaeological Park (Teià, Maremse, Barcelona): A great experimental archaeology laboratory’, in F. W. F. Foulds (ed.), Experimental Archaeology and

Theory. Recent approaches to archaeological hypotheses. Oxford, 67-100. Masià, R. 2015. ‘On dating Hero of Alexandria’, Archive for History of Exact Sciences 69: 231-255. Matijašić, R. 1993. ‘Oil and wine production in Istria and in Dalmatia in Classical Antiquity and the early Middle Ages’, in M.-C. Amouretti and J.-P. Brun (eds), La production du vin et de l’huile en Méditerranée; Oil and wine production in the Mediterranean area. Paris, 247261. Mattingly, D. J. 1990. ‘Paintings, Presses and Perfume Production at Pompeii’, Oxford Journal of Archaeology 9: 71-90. Mattingly, D. J. 1993. ‘Maximum figures and maximizing strategies of oil production? Further thoughts on the processing capacity of Roman olive presses’, in M.-C. Amouretti and J.–P. Brun (eds), La production du vin et de l’huile en Méditerranée. Oil and wine production in the Mediterranean area (BCH Supp. 26). Paris, 483498. Mattingly, D. J. 1996. ‘Olive Presses in Roman Africa: Technical Evolution or Stagnation?’, in M. Khanoussi, P. Ruggeri and C. Vismara (eds), L’Africa romana. Atti di convegno di studio Cartagine, 15-18 dicembre 1994. Ozieri, 577-595. Montanari, F. 2005. ‘Hero’, in H. Canick and H. Schneider (eds), Brill’s New Pauly. Stuttgart. Murphy, T. 2004. Pliny the Elder’s Natural History: The Empire in the Encyclopedia. Oxford. Nix, L. and Schmidt, W. 1900 Heronis Alexandrini opera quae supersunt omnia II Mechanica et catoprica. Leipzig. Peña Cervantes, Y. 2010. Torcularia. La producción de vino y aceite en Hispania. Tarragona. Paton, W. R. and Myres, J. L. 1898. ‘On Some Karian and Hellenic Oil-Presses’, JHS 18: 209-217. Reeve, M. D. 2007. ‘The Editing of Pliny’s Natural History’, Revue d’histoire des textes, n.s., 2: 107-179. Roll, I. and Ayalon, E. 2009. ‘A Byzantine Wine Press at Kh. Duran (Rehovot)’, in E. Ayalon, R. Frankel and A. Kloner (eds), Oil and Wine Presses in Israel from the Hellenistic, Roman and Byzantine Periods, BAR Int S 1972. Oxford, 265-268. Rossiter, J. J. 2007. ‘Wine-making after Pliny: Viticulture and farming technology in late antique Italy’, in L. Lavan, E. Zanini and A. Sarantis (eds), Late Antique Archaeology 4 Technology in Transition A.D. 300-650. Leiden, 93-118. Rossiter, J. J. and Freed, J. 1991. ‘Canadian-Turkish excavations at Domuztepe, Cilicia (1989)’, Echos du Monde Classique/Classical Views 35: 145-174. Rossiter, J. J. and Haldenby, E. 1989. ‘A Wine-Making Plant in Pompeii Insula II.5’, Echos Du Monde Classique/Classical Views 33, 229-39. Rozier, F. 1776. ‘Vues economiques sur les moulins et pressoirs à l’huile d’olives, connus en France ou en Italie’, Observations sur la physique, sur l’histoire naturelle et sur les arts 8: 417-443. 109

Stone Tools in the Ancient Near East and Egypt Schneider, H. 2007. ‘Technology’, in W. Scheidel, I. Morris, and R. Saller (eds), The Cambridge economic history of the Greco-Roman world. Cambridge, 144-172. Schneider, H. 2009. ‘The Relationship between Scientific, Technical and Moral Knowledge in Classical Antiquity’, in G. Hirsch Hadorn (ed.), Unity of Knowledge in Transdisciplinary Research. Oxford, vol. 1. Sordinas, A. 1971. Old Olive Mills and Presses on the Island of Corfu, Greece. An essay on industrial archaeology and the ethnography of agricultural implements. Memphis. Syme, R. 1979. Roman Papers II. E. Badian (ed.). Oxford. Taxel, I. 2013. ‘The Olive Oil Economy of Byzantine and Early Islamic Palestine: Some Critical Notes’, Liber Annuus 63: 361-394. Thurmond, D. L. 2006. A handbook of food processing in classical Rome: for her bounty no winter. Leiden. Uceda, I. H. 2012. ‘Torcvlaria inéditos de dos villas del territorio de Valentia (Conventvs Carthaginiensis)/ Some unpublished torcularia from two villae in the

territory of Valentia (Conventvs Carthaginiensis)’, Anales de Prehistoria y Arqueología 27: 273-281. Waliszewski, T. 2014. Elaion. Olive Oil Production in Roman and Byzantine Syria – Palestine, PAM Monographs 6. Warsaw. White, K. D. 1975. Farm Equipment of the Roman World. Cambridge. White, K. D. 1984. Greek and Roman Technology. London. Zelinger, Y. 2009. ‘A Byzantine Oil Press at Jifna’, in E. Ayalon, R. Frankel, and A. Kloner (eds), Oil and Wine Presses in Israel from the Hellenistic, Roman and Byzantine Periods. (BAR IntS 1972). Oxford, 231237. Zerbini, A. 2012. ‘Landscapes of production in Late Antiquity: wineries in the Jebel al-’Arab and Limestone Massif ’, in A. Le Bihan, P. Blanc, F. Braemer, J. Dentzer-Feydy and F. Villeneuve (eds), Territoires, architecture et matériel au Levant: Doctoriales d’archéologie syrienne. Paris-Nanterre, 8-9 décembre 2011. Beyrouth. http://books.openedition.org/ifpo/2886

110

Raw Material and Manufacture

111

112

Tool marks on Old Kingdom limestone vessels from Abusir – production of canopic jars and model vessels Lucie Jirásková Keywords: Egypt, Old Kingdom, stone vessels, limestone, manufacture

1. Introduction

pre-form (e.g. Willems et al. 2009; Shaw 2010; Heldal, Storemyr, Bloxam and Shaw 2016). Consequently, one should expect such a workshop also at the site of the Old Kingdom quarry at Tura. The occurrence of canopic jars and model stone vessel assemblages only in Memphis, the Old Kingdom administrative centre, clearly locates the final treatment of stone pieces in the Memphite territory. However, no clear traces of a workshop producing limestone model vessels have been found so far. The only archaeological evidence from such an area might be the workshop discovered by A. G. Reisner in the valley temple of Menkaure at Giza. Unfortunately, the excavator gave no details on the archaeological situation and only mentioned the presence of broken statues of Menkaure and Khafre (of unspecified material), which could have been used in the production of model stone vessels of the Fifth and Sixth dynasty (Reisner 1931: 104).

As limestone is the most common stone in Egypt, it is not surprising to find it used not only in architecture but also in sculpture, stone vessel manufacture and other branches of material culture. The advantage of this material resides not only in its widespread and abundant sources but also in its characteristics. Quarries were easily accessible, and the softness of limestone enabled fast and precise working of any intended object or feature. The white colour of the material was also of high importance, since it could symbolise purity in its raw state or could be easily coloured. The Old Kingdom (ca 2592-2118 BC) society was well aware of the advantages of this material and exploited it on a large scale (Klemm and Klemm 2008: 23-145). The kind and provenance of the limestone used depended on the consumer’s intentions. Common architectural blocks were usually made of a locally available limestone, whereas special architectural features, statues, and vessels were made exclusively from the fine-grained limestone quarried in Tura on the east bank of the Nile, opposite the capital and the main cemetery of kings and their officials.

2. The cemetery of Abusir The large cemetery of Abusir is situated south of Giza and north of Saqqara, with which it was historically connected. The large area contains various types of tombs. The royal pyramids are situated in the northern part, surrounded by the mastabas of the kings’ relatives. Abusir South is the area of ‘private’ tombs, i.e. the mastabas and rock-cut tombs of the members of ancient Egyptian administration. The royal pyramids and the surrounding tombs were built in the Fifth Dynasty (ca 2435-2305 BC), like most of the tombs at Abusir South. However, past excavations have also brought to light tombs from the Third to early Fourth Dynasties as well as from the Sixth Dynasty (ca 2306-2150 BC).

Considering the process of quarrying, transport, and subsequent working, it is not easy to decide whether the pieces of limestone used in the production of stone vessels were mere ‘by-products’ of primary working of the stone intended for a larger architectural element, or if they were cut from larger blocks delivered to stone vessel workshops. Where large vessels such as canopic jars are considered, it is more plausible that they were cut from a larger block of limestone. In contrast, the small model bowls might have been crafted from ‘waste’ material, which may explain why their sizes varied in the individual assemblages.

Except for the royal pyramidal structures (Vlčková 2006), all other tombs were quite consistent in the nature of stone vessels. Large functional pieces (i.e. not model, dummy vessels, but with fully-drilled interiors) occur very rarely among stone vessels collected from Old Kingdom officials’ tombs; if they do, they are mostly simple bowls. On the contrary, most of the burial chambers contained canopic jars

The evidence from quarries for other kinds of stone used for the production of vessels indicates a preworking at the quarrying site, where the approximate rough shape of the future vessel was given to the 113

Stone Tools in the Ancient Near East and Egypt and model jars. Two materials were employed in their production: travertine and limestone. Travertine was more prestigious and is found mainly in the tombs of the highest officials and members of the royal family. Travertine is rarely used for canopic jars (found only in tombs of kings, royal mothers and a few high officials), whereas it is more commonly used for model vessels (associated with higher officials and members of the royal family). Limestone was a rather traditional material in the case of the Old Kingdom canopic jars, but it was also frequently used in the production of model vessels, especially during the latter part of the Fifth Dynasty and also in the Sixth dynasty.1

angled or rounded collared rim. Some of them are slimmer, some wider, with a rim part that can be thick or rather tiny. All the jars bear many tool marks both outside and inside. The outer face of the body was usually smoothed with a pebble, which left thin grooves in various directions (mostly vertical). The bodies of the vessels often bear traces of damage and subsequent repair (for a detailed study, see Jirásková 2015). These patched reveal special treatment of the surface. It was roughened, probably using a coarse-grained pebble, so that the patch would hold well in position. The rim part is always smoothed, and it is not clear if it was modelled using a copper chisel or a flint blade. The base part sometimes remains unsmoothed, in which case it bears the cutting marks of a chisel. A well-preserved example is 24/AS37/2007 (Fig. 1). Here the chisel was probably already worn and left a wavy tool mark c. 0.6-0.7 cm wide. Two other jars from the set, 25/AS37/2007 and 26/AS37/2007, also had chisel marks on their bases. The bases were cut by a chisel of the same size, but this time, the cuts are almost flat.

This article aims to provide a detailed study of these two kinds of limestone vessels from the point of view of their manufacture. The author of the study has had an opportunity to examine most of the canopic jars and model vessels that have been discovered at Abusir over the last 15 years. They all proved to bear many tool marks, which provide clear indications of the production process.

The use of a chisel for cutting the large base might seem to be unreasonable. Since this part was supposed to be flat, it should have been much easier to cut with a saw. However, unlike in model vessels, no traces of sawing have been noticed on the bottom parts of the studied canopic jars.

3. Limestone canopic jars from Abusir2 Complete limestone canopic jars discovered in the royal cemetery come from the burial chamber of Princess Khekeretnebty (Tomb AC 15; Verner and Callender 2002: 31-32) and the anonymous Tomb AC 31 (Krejčí 2016: Fig. 9). Fragments were collected in the Tombs of Vizier Ptahshepses (Tomb AC 8; Dulíková, Jirásková and Odler in preparation), Lady Tisethor (Tomb AC 15; Verner and Callender 2002: 31-32), Prince Nakhtsare (Tomb AC 25; Krejčí, Callender and Verner 2008: 51-52), and the anonymous Tomb AC 24 (Krejčí, Callender and Verner 2008: 185).

The interior of the limestone canopic jars from Abusir bear several interesting features. There are only a few well-smoothed pieces on the inside which are of no interest for this study; the others are roughly finished inside, which might be considered evidence for their symbolic function (for a wider discussion about this topic, see Rzeuska 2010 and 2011; Bárta 2016). Iconographic evidence indicates the use of a copper tubular drill in the production of large vessels similar in shape to the canopic jars (e.g. Arnold and Pischikova 1999: Fig. 73). D. Stocks’ reconstruction of the process

The shafts of Abusir South were hiding complete sets of canopic jars of a member of Vizier Qar’s family (Tomb AS 16, Shaft 5; Bárta et al. 2009: 142-143), priest Neferinpu and his relative (Tomb AS 37, Shaft 1; Bárta et al. 2014: 94-96, 104-106), the husband of Princess Sheretnebty (AS 68c, Shaft 1; Jirásková 2015: Figs. 5-6), the wife and a member of the family of Nefer (Tomb AS 68d, Shafts 2 and 4; Jirásková 2014: Obr. 3 and Bárta 2014: 4, Obr. 2), and of the unknown owners of Tombs AS 31 (found outside the tomb; unpublished), AS 32 (Shaft 5, almost complete; unpublished), AS 47 (Jirásková 2015: Fig. 9) and AS 67 (Shafts 1 and 2; Arias Kytnarová, Havelková, Jirásková, Sůvová and Bárta 2013: 90-94). All of these limestone canopic jars have a traditional shape; they are mostly tall, shouldered jars with an A detailed study of the chronology of models stone vessels including the interpretation of material was published in Jirásková 2017b. 2  For a detailed table referring to all canopic jars and their fragments discovered by the Czech Institute of Egyptology at Abusir, see Jirásková 2015: Fig. 2. 1 

Figure 1. Base of the canopic jar 24/AS37/2007 with copper chisel cut marks (photo L. Jirásková).

114

L. Jirásková: Tool marks on Old Kingdom limestone vessels from Abusir of gouging out the vessel’s interior involved a hollowed copper tubular drill in the first step and a figure-ofeight stone drill to enlarge the cavity (Stocks 2003). That seems to be true for the canopic jars of Khekeretnebty. In contrast to other Old Kingdom limestone canopic jars discovered at Abusir, they still show visible, mostly horizontal lines, such as in the cases of P5562a and P5563a. The jars P5564a and P5565a bear horizontal lines only in the lower part, whereas the upper half of the vessels was smoothed using a pebble moved in a vertical direction. Stocks also claimed that the core left by the copper tubular drill was cut out using a chisel. In fact, the tool marks left in some of the Old Kingdom canopic jars discovered at Abusir indicate the use of chisels on a wider scale than only for removing the core. For instance, the jars from the burial chamber of Neferinpu (4-7/AS37/2007) seem to have been drilled only in the lower part, whereas the upper part was enlarged and modelled using only a chisel (Fig. 2). Although the surface was also smoothed by a pebble in the upper part, many chisel cut marks remained. It is possible that no figure-of-eight stone drill at all was

used, at least for some of the pieces. Some of the most poorly worked jars show only chisel cuts inside, such as the set from the other burial chamber in Shaft 1 in the Tomb of Neferinpu (23-26/AS37/2007). These have no traces of horizontal drilling at all, but bear vertical chisel cut marks throughout the jars’ interior (Fig. 3). Interestingly, these four canopic jars were also modelled and smoothed inside using plaster; no vertical lines from a smoothing pebble can be seen inside. The set of canopic jars from Tomb AS 47 are possibly the result of a combination of all the manufacturing processes listed above. One jar (9/AS47/2007) was completely hollowed out using a chisel (probably with the help of a tubular drill at the beginning), another one (10/AS47/2007) was modelled inside with a figureof-eight drill, and the last two (4/AS47/2007 and 8/ AS47/2007) were gouged out using both techniques, i.e. a tubular drill in the lower part and a chisel in the upper part.

Figure 2. Cross-section of the canopic jar 5/AS37/2007. The lower part remained narrow after the application of a copper shaft drill and probably of figure-of-eight stone drills, whereas the upper part was widened using a copper chisel (drawing L. Jirásková).

Figure 3a. Cross section of the canopic jar 24/AS37/2007, which was completely gouged out using a copper chisel (drawing L. Jirásková)

115

Stone Tools in the Ancient Near East and Egypt

Figure 3b. Detail of the interior of 24/AS37/2007 with unsmoothed cut marks (photo L. Jirásková).

Figure 4. Wavy interior of the canopic jar 388_1/AS68/2014, which was modelled by a figure-of-eight stone drill (photo M. Frouz).

The canopic jars from Shaft 2 of Tomb AS 67 were almost perfectly smoothed inside, leaving drill marks only at the very bottom. The interior lowest part of the bottom of most of the studied canopic jars (except for those that were cut only by chisels) was usually modelled in the same way. There are always circular or triangular horizontal drilling marks that indicate the use of a small flint or chert crescent-shaped borer to remove the remains of the core left after the use of the tubular drill. However, the shape of the cavities of the canopic jars from Shaft 2 of Tomb AS 67, especially in its upper part, points to chisel work. Shaft 1 of Tomb AS 67 contained a set of two jars gouged out by a chisel, and two jars probably made using a drill. A figure-ofeight drill borer was also used in the case of the jars found in Shaft 1 in Tomb AS 68c and Shafts 2 and 4 in AS 68d; they were all carefully smoothed inside (Fig. 4).

4. Model vessels To date, a handful of assemblages of model stone vessels have been collected in the necropolis of Abusir.3 There is no complete or almost complete set from the tombs in the pyramid field, but a number of pieces have been found in the Tombs of Princess Khekeretnebty (Tomb AC 15; Verner and Callender 2002: 34-38), Queen Khentkaus III (Tomb AC 30; Krejčí, Arias Kytnarová and Odler 2015: Fig. 5) and in the anonymous Tomb AC 31 (Krejčí 2016: Pl. 1). Only a few bowls or jars come from the burial chambers of Queen Khentkaus II (Tomb AC 14; Jirásková 2017a), Prince Nakhtsare (Tomb AC 25; Krejčí, Callender and Verner 2008: 52-54) and from the anonymous Tombs AC 22 (Krejčí, Callender and Verner 2008: 109-111) and AC 24 (Krejčí, Callender and Verner 2008: 85-86). However, all of these vessels were made exclusively of travertine, which is not the topic of this study.

Although the application of both methods for hollowing out the jar’s interior may seem complicated, it is very reasonable. It must have been the most practical and comfortable way of working the jars, since the lower part was more difficult to access and probably much easier to enlarge using a drill. The width of the vessel in this part was also much smaller than in the upper part, which was wide near the mouth. The upper half could have been well reached by a relatively small chisel with a width of about 0.7 cm. Vessels completely cut inside with such a chisel provide evidence for its length: for instance, the depth of one such jar 9/AS47/2007 was 18 cm.

Abusir South has revealed one of the few examples of intact assemblages. It was discovered in the burial chamber of Neferinpu (Tomb AS 37; Bárta et al. 2014: 147-164). Other large and almost complete sets come from the Tombs of Inty (Tomb AS 22, Shaft J; Jirásková forthcoming), Inty-Pepyankh (Tomb AS 22, Shaft A; A general overview of the assemblages of model stone vessels discovered at Abusir is presented in Jirásková 2017a. 3 

116

L. Jirásková: Tool marks on Old Kingdom limestone vessels from Abusir Jirásková forthcoming), Nefer (Tomb AS 68d, Shaft 1; unpublished) and the anonymous Tombs AS 27 (Jirásková 2017a), AS 47 (unpublished), AS 67 (Shafts 1 and 2; Arias Kytnarová, Havelková, Jirásková, Sůvová and Bárta 2013: 90-94). Limestone was the prevailing material at the officials’ cemetery, whereas travertine occurred only sporadically (AS 22, AS 27 and AS 31).4 Almost all the model vessels mentioned above bear traces left after their production. The tools used have not been determined only on the basis of the tool marks; a series of simple test boring and smoothing has been undertaken. Limestone is very soft and easy to work, and, as in the case of canopic jars, there was not a single method for producing limestone model vessels.

Figure 5. Roughly shaped unsmoothed exterior of the bowl 19_9/AS37/2007 with planes left after the use of a copper adze (photo L. Jirásková).

The first step involved the preparation of the piece of stone. Tool marks left on the bases of the vessels and sometimes on the rims of the less-modelled and less-smoothed pieces indicate the use of saws. In this respect, all the vessels were carved out of small stone blocks cut by saws. Saw marks can be seen in places where the vessels’ walls are unsmoothed and flattish. Here, there are fine regular linear grooves still visible. Many of the vessels still have visible flat areas on the surface of their walls. These were left after the first stage of working of the stone.

have left two different tool marks: circular and/or triangular grooves (Fig. 6). If the bowl was not smoothed properly inside, these lines can be well documented. They are mostly centrally oriented towards the bottom; in some instances, the drilling was done irregularly and

During the second step, the rough shape of the vessel was cut, probably using copper adzes. These left some flattish areas on the vessel walls. In most cases, these flattish areas were further worked; the edges were smoothed and the vessel walls rounded. However, some vessels still display these flat areas (e.g. Tomb AS 37; Fig. 5). These unworked examples demonstrate that the shape of the future vessel was well-defined already at the time of rough cutting. The next stage probably involved small copper chisels and sharp chips of flint,5 which modelled the grooves and other – sometimes only decorative – features of the vessel. These tool marks are also often still visible, especially in the depressions, where the surface could not have been smoothed properly. The final treatment of the outer surface involved smoothing with a pebble, which eliminated sharp edges.

Figure 6a. Depression of the bowl 6_27/AS67/2012 with triangular boring traces (photo L. Jirásková).

The depressions of bowls and inner cavities of tall jars were created in several ways. In the case of bowls, there is evidence of crescent-shaped borers that were turned by hand. It is interesting to note that these tools could For the discussion on travertine versus limestone in the production of Old Kingdom model stone vessels, see Jirásková 2017b. 5  The flint tools of the Early Dynastic period and the Old Kingdom have been recently catalogued by M. Kobusiewicz. Concerning the tools suitable for the production of stone vessels, he discussed crescent-shaped and other borers as well as microperforators (Kobusiewicz 2015: 20-22). 4 

Figure 6b. Depression of the bowl 16_30/AS67/2012 with circular boring traces (photo L. Jirásková).

117

Stone Tools in the Ancient Near East and Egypt

Figure 8. The symbolic shallow interior of the jar 383_16/ AS68/2014 was worked with a copper chisel or a pick (photo L. Jirásková).

Most were probably simply bored using a very narrow flint chip: a process that left regular circular horizontal grooves inside. A tubular drill was not necessary in this case, since limestone is sufficiently soft. There are also no remains of cylindrical cores left inside the cavities, such as in case of the travertine pieces. The only assemblage that may point to the use of a tubular drill comes from the Tomb of Nefer at Abusir (AS 68d). The depressions inside the jars definitely involved copper chisels of a width of approximately 0.6-1.0 cm (Fig. 8). The question is whether the interior of these jars was simply cut out using a chisel and probably a kind of pick, or whether the chisel served as a tool for the extraction of a tiny cylindrical drilling core. The first possibility seems to be more plausible, since the chisel marks are in various directions in a single jar. Chisels also seem to have been used for modelling the cavity without any subsequent smoothing. If the chisel had been used merely to cut out the core, one would expect to see a cut in one direction only.

Figure 7. Depression of the bowl 19_25/AS37/2007 which was not bored but gouged out using a copper chisel (photo L. Jirásková).

had its centre in the wall of the bowl rather than at the bottom. The test boring demonstrated that the circular grooves are left by a firmly held, mostly narrow piece of a crescent-shaped borer that stays in position, whereas the triangular tool marks are the result of a wider borer that turns irregularly out of its axis. Many limestone bowls bear triangular tool marks, which may point to a rather fast production process. Even these tool marks could have been later smoothed by a pebble, but this seldom happened. Another way for making the depression in a limestone bowl involved the use of a copper chisel instead of a crescent-shaped borer. A well-documented example is the assemblage of Neferinpu (AS 37), where chisels of an approximate width of 0.6-0.8 cm were used to hollow out the depressions of the roughly shaped bowls (Fig. 7). The use of chisels to hollow-out the interior of vessels is also documented on vessels from the Levant and the Aegean area (Bevan 2007: 55-56; Sparks 2007: 194; Squitieri 2017: 100-102; Evely 1980). However, the study of the tool marks on various Old Kingdom limestone vessels has shown that the use of chisels was much more widespread in Egypt than was expected and documented in the past.6

5. Conclusions The study of limestone canopic jars and model vessels has shown that there was no unique, generally accepted method for their manufacture. Copper tubular drills, figure-of-eight drills, copper chisels and flint borers were all employed to various degrees, depending probably partly on the personal preference of the workman, partly on the social position of the consumer. Finer canopic jars were always made using tubular drills and figure-of-eight drills. They were perfectly smoothed and modelled outside, such as those of Hathornofret (AS 68d, Shaft 2). They also had a narrow mouth and wider shoulders. In contrast, average canopic jars were roughly shaped and smoothed. Their mouths were wide so that the cavities could be gouged out using a chisel (at least in the upper part).

The symbolic cavities of tall model jars made of limestone bear two kinds of tool marks once again. The copper tools of the Old Kingdom have been recently systematically documented and presented in Odler et al. 2016. 6 

118

L. Jirásková: Tool marks on Old Kingdom limestone vessels from Abusir Higher quality model vessels, such as those from AS 67, Shaft 1, were always perfectly modelled and smoothed. The bowls had a regular thickness and their bottom was rounded. Lower grade pieces usually had a flat basis, and smoothing was mostly omitted (AS 37). The preference for flint or chert borers is clear.

Arnold, D. and Pischikova, E. 1999. Stone vessels. Luxury items with manifold implications, in D. Arnold (ed.) Egyptian Art in the Age of the Pyramids: 121-131. New York: The Metropolitan Museum of Art. Bárta, M. 2014. Objev pohřební komory hodnostáře Nefera. Pražské egyptologické studie XIII: 3-6. Bárta, M. 2016. A new intact case for an Old Kingdom dummy mummification from the Fifth Dynasty Abusir, in R. Landgráfová and J. Mynářová (eds) Rich and Great. Studies in Honour of Anthony J. Spalinger on the Occasion of his 70th Feast of Thoth: 15-17. Prague: Faculty of Arts, Charles University in Prague. Bárta, M. et al. 2009. Tomb complex of the vizier Qar, his sons Qar Junior and Sedjemib, and Iykai. Prague: Czech Institute of Egyptology – Dryada. Bárta, M. et al. 2014. The Tomb of the sun priest Neferinpu (AS 37). Prague: Faculty of Arts, Charles University in Prague. Dulíková, V., Jirásková, L. and Odler, M. In preparation. Burial equipment of Vizier Ptahshepses from Abusir. Heldal, T., Storemyr, P., Bloxam, E. and Shaw, I. 2016. Gneiss for the Pharaoh: Geology of the Third Millennium BCE Chepren’s Quarries in Southern Egypt. Geoscience Canada 43: 63-78. Evely, D. 1980. ‘Some Manufacturing Processes in a Knossian Stone Vase Workshop’, in E. B. French & K. A. Wardle (eds), Problems in Greek Prehistory, Bristol: Bristol Classical Press, pp. 397-415. Jirásková, L. 2014. Kanopy v době Staré říše. Pražské egyptologické studie XIII: 49-55. Jirásková, L. 2015. Damage and repair of the Old Kingdom canopic jars – the case at Abusir. Prague Egyptological Studies XV: 76-85. Jirásková, L. 2017a. Abusir miniature and model stone vessels of the Old Kingdom, in T. Gillen (ed.) (Re) productive Traditions in Ancient Egypt. Proceedings of the Conference held at the University of Liège, 6th – 8th February 2013 (Aegyptiaca Leodiensia X). Liège, 425443. Jirásková, L. 2017b. Model stone vessels of the Old Kingdom – typology and chronology, in M. Bárta, F. Coppens and J. Krejčí (eds) Abusir and Saqqara in the Year 2015. Prague, 145-156. Jirásková, L. Forthcoming. Stone vessels of the tomb of Inty, AS 22, in M. Bárta et al., The tomb of Inty (AS 22). Prague. Klemm, R. and Klemm, D. D. 2008. Stones and quarries in ancient Egypt. London: British Museum Press. Krejčí, J. 2016. Archaeological excavation of tomb AC 31 in Abusir Centre. A preliminary report. Prague Egyptological Studies XVII: 12-23. Krejčí, J., Arias Kytnarová, K. and Odler, M. 2015. Archaeological excavation of the mastaba of Queen Khentkaus III (tomb AC 30), Prague Egyptological Studies XV: 28-42. Krejčí, J., Callender, V. G. and Verner, M. 2008. Minor tombs in the royal necropolis I (The mastabas of Nebtyemneferes and Nakhtsare, pyramid complex Lepsius no. 24 and tomb

The techniques for the manufacture of limestone vessels differed not only at the level of assemblages. As we have seen before in the case of canopic jars, vessels from the same assemblage seem to have been produced differently. Model jars and bowls are usually more consistent in terms of techniques adopted for manufacturing them, but the final stage of manufacture seems to have been done in different ways: some bowls were smoothed, some remained roughly shaped, and others were only basically worked (e.g. Tomb AS 37). The use of different methods and their different results may indicate that several artisans were at work producing the sets of canopic jars or model vessels. An intriguing question is whether there was one, or many, workshops in the Memphis area. In either case, it/they must have been closely connected to the royal administration and control, since the production of canopic jars and model stone vessels was limited to the area of the royal necropoleis in the Old Kingdom. 6. Final remarks Limestone vessels were produced on a large scale in Old Kingdom Egypt, especially in a non-royal context. Almost all canopic jars and many assemblages of model vessels were made of limestone, probably due to its softness, accessibility and colour. The article has focussed on the pieces collected in the cemeteries of Abusir and has presented their production methods based on the tool marks. The manufacture of canopic jars involved mainly copper adzes, copper tubular drills, stone figure-of-eight drills, and very often also copper chisels. Model vessels were made using, in particular, copper adzes, crescent-shaped flint borers and sometimes also copper chisels. In conclusion, both canopic jars and model vessels were made adopting different techniques, which probably indicates that several workshops were active at the same time, or that the demand for the vessels entailed different production techniques. The intended quality of the finished product, personal preference, and training of individual artisans may all have been elements behind the choice. Bibliography Arias Kytnarová, K., Havelková, P., Jirásková, L., Sůvová, Z. and Bárta, M. 2013. Záhady hrobky AS 67. Jedna hrobka pro dvě generace?. Pražské egyptologické studie X: 80-94. 119

Stone Tools in the Ancient Near East and Egypt complex Lepsius no. 25. Prague: Faculty of Arts, Charles University in Prague. Odler, M. et al. 2016. Old Kingdom Copper Tools and Model Tools, Oxford: Archaeopress. Reisner, G. A. 1931. Mycerinus. The Temples of the Third Pyramid at Giza, Cambridge, MA: Harvard Univeristy Press. Rzeuska, T. I. 2010. A wooden chest from West Saqqara and its enigmatic contents. Remarks on mummification process in the Old Kingdom. Études et Travaux XXIII: 90-140. Rzeuska, T. I. 2011. And where are the viscera…? Reassessing the function of Old Kingdom canopic recesses and pits, in N. Strudwick and H. Strudwick (eds) Old Kingdom, New Perspectives. Egyptian Art and Archaeology 2750-2150 BC: 244-255. Oxford – Oakville: Oxbow Books. Shaw, I. 2010. Hatnub: Quarrying Travertine in Ancient Egypt. London: Egypt Exploration Society.

Squitieri, A. 2017. Stone Vessels in the Near East during the Iron Age and the Persian Period (c. 1200-330 BCE). Oxford: Archaeopress. Stocks, D. A. 2003. Experiments in Egyptian archaeology. Stoneworking technology in Ancient Egypt. London – New York: Routledge. Verner, M. and Callender, V. G. 2002. Djedkare’s family cemetery: Prague: Czech Institute of Egyptology, Faculty of Arts, Charles University in Prague – Set Out. Vlčková, P. 2006. Stone vessels from the mortuary complex of Raneferef at Abusir. Prague: Czech Institute of Egyptology. Willems, H. et al. 2009. An industrial site at al-Shaykh Sacíd / Wádí Zabayda’. Ägypten und Levante / Egypt and the Levant 19: 293-331.

120

Raw material variety and acquisition of the EB III ground stone assemblage of Tell es-Safi/Gath (Israel) Jeremy A. Beller, Haskel J. Greenfield, Mostafa Fayek, Itzhaq Shai, and Aren M. Maeir Keywords: Ground stone tools, Tell es-Safi/Gath, Early Bronze Age, southern Levant, provenance

1. Introduction

settlement dated to the EB III of the region (ca. 29002500 BCE). On the eastern spur of the tel, a section of a residential neighbourhood has been exposed. It is composed of approximately four buildings, which are intersected by an alleyway (Shai et al. 2012, 2014; Greenfield et al. 2016).

Until recently, ground stone objects and installations were largely ignored in discussions of material culture or under-studied (Rowan and Ebeling 2008). During the 1990s, their potential to contribute to the understanding of human behaviour was realized and an emphasis was placed on their integration into archaeological research (e.g., Eitam 1996; Elliott 1983; Schneider 1996; Wright 1993). As a result, a body of knowledge has been blossoming regarding the production, distribution, consumption, and discard associated with these objects (e.g., Dubreuil and Savage 2014; Eitam 2009; Frankel 1999; Gluhak and Rosenberg 2013; Wright 2008). In this manner, the nature of raw material procurement and the movement of ground stone tools has been a subject that has received growing inquiry in the Near East (e.g., Gluhak and Hofmeister 2011; Rutter and Philip 2008; Williams-Thorpe and Thorpe 1993). However, such information is lacking for the period when early cities and city-states spread across the Near East and become the dominant settlement pattern – the Early Bronze Age.

The ground stone assemblage from this EB settlement offers the opportunity to investigate issues surrounding the ground stone industry at a large urban centre during the zenith of urbanism and social complexity of the EB III southern Levant. This study describes the raw material of the ground stone objects and provides preliminary insight into the nature of their procurement and acquisition. In doing so, it demonstrates that the residents of Tell es-Safi/Gath were largely reliant on local materials, but were simultaneously wellconnected to intra-regional exchange systems that supplied non-local ground stone commodities. 2. Background and context 2.1 The southern Levant during the Early Bronze Age

This paper will focus on material from this crucial period, in order to help increase our understanding of their importance in daily life in the urban centres of the southern Levant (Fig. 1). In the southern Levant, the Early Bronze Age (hereafter EB) is divided into several phases (I-IV), wherein the nature of ground stone production and distribution systems probably changed over time with the shifting of political and economic conditions. At the zenith of the urban settlement system from this period (the EB II-III), fortified cities spread across the region suggesting difficult political conditions. Nonetheless, the exchange systems that operated to distribute ground stone objects to the various urban settlements continued to operate.

Relative to the Chalcolithic period (4500-3500 BCE), the Early Bronze Age is characterized by a pronounced increase in social complexity (de Miroschedji 2006, 2009; Greenberg 2014). From each EB sub-phase to its successor, there are apparent urban and demographic patterns that are archaeologically discernible. The EB I-II transition is marked by the expansion of polymorphic societies, consisting of mobile groups and small villages, into larger urban centres. Between the EB II and EB III, there is a minor shift where some villages are abandoned and others are erected, ultimately marking the zenith of urbanism during this period where city-states come to dominate each of the sub-regions (Finkelstein 1995; Rast 2001). The EB III of the southern Levant experiences a continuation of the settlement and other cultural patterns found in the EB II (de Miroschedji 2014). As a result, the two subperiods are often grouped together in the literature.

In this paper, we will present recently collected data from our excavations at Tell es-Safi/Gath, Israel. There we have uncovered portions of an extensive early urban 121

Stone Tools in the Ancient Near East and Egypt

Figure 1. Notable sites of the EB II-III southern Levant.

122

J. A. Beller et al.: Raw material variety and acquisition of the EB III ground stone assemblage Among many societal facets, the EB II-III features the aggregation of populations, the emergence of varying economic specializations, the appearance of heavily fortified settlements, the development of redistributive economies (greater surplus production and exchange), and evidence for more extensive long distance trade and exchange (de Miroschedji 2009; Mazar 1990). However, the end of the EB III is characterized by a widespread abandonment and, in some cases, the destruction of numerous urban centres.

Ramon Crater, Israel, supplied residents of the nearby Camel Site with sandstone grinding stones (Abadi-Reiss and Rosen 2008). Furthermore, the surplus of material allowed the residents to generate an exchange system that transported sandstone commodities to distant settlements, such as Horvat Ahdir (Cohen 1999) and Arad (Amiran et al. 1997). 2.3 Tell es-Safi/Gath The archaeological site of Tell es-Safi/Gath is located in central Israel on the border of the Shephelah (foothills) and Coastal Plain. This isolated mound sits roughly midway between the modern cities of Bet Shemesh and Ashdod, and is approximately 25 km inland from the Mediterranean Sea (Maeir 2012). The local landscape that surrounds Tell es-Safi/Gath includes arable fields and the Nahal Ha’elah water source, which together with the height of the formation provided a hospitable and strategic location for long-term residence. The sequential layers of the tell, from modern to Chalcolithic, are situated atop a natural limestone formation (Ackermann and Bruins 2012). However, the EB III settlement is the earliest excavated layer to date and is approximately 24 ha in size (Shai et al. 2016a; Uziel and Maeir 2005). It is surrounded by a massive stone fortification system that encircles the entire top of the tell (Shai et al. 2016a).

2.2 Ground stone material and movement during the Early Bronze Age In general, the raw materials of ground stone assemblages of the EB settlements are dominated by what is locally available (Milevski 2011). In this manner, residents of settlements (e.g., Bet Yerah) proximate to basalt sources would exploit these more heavily in comparison to acquiring ground stone commodities of non-local materials (e.g., sandstone). Quite often, non-local materials are discovered among the ground stone assemblages, though in much lower frequencies than those of local materials (Milevski 2008). However, little discussion is afforded to limestone ground stones, despite the abundance of this raw material. At present, much of the scholarship on EB ground stone tools, their procurement, and movement has focused on basalt artifacts. In particular, the basalt vessel industry, which was prominent during the ChalcolithicEB I, has become a hallmark of these periods (Braun 1990). Several provenance studies of basalt vessels have indicated that their main region of origin is the basalt sources on the eastern side of Dead Sea in Jordan and, to a lesser extent, around the Sea of Galilee (MalloryGreenough and Greenough 2004; Philip and WilliamsThorpe 1993, 2000). Their movement has extended throughout the southern Levant and into Egypt. Households were not the only consumers of the ground stone commodities, as the basalt vessels were recovered from Egyptian tombs and Proto-Canaanite burials (Mallory-Greenough and Greenough 2004). While these artifacts are considered to be ritual or prestige items, the scale of exchange wains during the EB II-III (Rutter 2003). Basalt spindle whorls were also exchanged with Egypt and north-western Mesopotamia during the EB I and are assumed to originate from Levantine sources (Savage 2011).

Extensive descriptions of the EB III occupation at Tell es-Safi/Gath can be found elsewhere (e.g., Greenfield et al. 2016; Shai et al. 2012, 2014), but a brief summary is presented below. The recovery of EB III remains has been concentrated in Area E on the eastern spur of the tell, away from the acropolis, where a residential neighbourhood has been exposed (Fig. 2). The neighbourhood is comprised of four housing complexes aligned on either side of a narrow alleyway. Although many goods of exotic origins are present, there are very few elite artifacts and no evidence of elite structures among the Area E remains (Shai et al. 2014). The housing complexes form part of a non-elite residential neighbourhood, which has been proposed to possibly represent a merchants’ quarter (Greenfield et al. 2012). At present, two stratigraphic layers in Area E have been identified as belonging to the EB occupation before the site is abandoned in c. 2500 BCE at the end of the EB IIIB. From latest to earliest, they are Strata E5 (later half of the EB III) and E6 (earlier half of EB III). Stratum E5 (60 m2) is the more excavated of the two strata and contains three separate occupational phases (E5a, E5b, and E5c). Each phase of E5 (a-c) represents a shift in the architectural organisation of the neighbourhood, meaning that the layout of buildings and rooms alters during the EB III. In addition, a much smaller area of the underlying Stratum E6 has been uncovered (20 m2). To date, seven dwellings (temporally) and four

Unfortunately, no EB ground stone workshops have been discovered, impeding a complete reconstruction of the nature of production and the scale of distribution. However, the discovery of an EB II sandstone quarry operation in the Negev Desert has offered insight into the nature of raw material procurement and preference within the southern Levant (Rosen and Schneider 2001). Here, two adjacent quarries in the 123

Stone Tools in the Ancient Near East and Egypt

Figure 2. Excavation areas of Tell es-Safi/Gath.

complete (Fig. 5). Many (n=20) of the lower grinding stones are fragments that could not be further divided into sub-types. Their counterparts, upper grinding stones, are less numerous (n=13). Similarly, most upper grinding stones are fragments (n=10). However, the sub-type of many can be discerned with loaf handstone as the most frequent (n=6) (Fig. 6). In addition, there is an unidentified grinding stone among the assemblage, which, although it retains a level use face, cannot be identified as an upper or lower implement. A suite (n=18) of perforated stones is present in the form of miscellaneous weights (Fig. 7) and preform disks. The former can be divided based on their size, where smaller weights likely functioned as spindle whorls and larger weights may have served on digging sticks or crank drills. There are two minute fragments of vessels within the assemblage (Fig. 8). These only retain portions of the rim and sides, which permit their identification. All the pounders (n=13) are complete and are small spheroidcuboid objects that are easily operable in one hand (Fig. 9). Similarly, both (n=2) rubbers are complete and are identifiable by the abrasive wear on their use-faces. The mortars/sockets are few in number (n=5), but are similar in form and cavity size to each other (Fig. 10). Pestles (n=2), the likely counter part of mortars, range

dwellings (spatially) have been uncovered within the neighbourhood (Shai et al. 2012; Shai et al. 2014). 3. The ground stone assemblage A total of ninety ground stone tools comprise the EB ground stone assemblage from Area E at Tell es-Safi/ Gath from the years 2004-2013. The observations made thus far are preliminary in nature, as the excavations at Tell es-Safi/Gath are on-going with one excavation season remaining. The number of ground stones will also grow as those from years 2014-2017 are examined. The assemblage is currently stored at BarIlan University, Israel. While the following paragraphs provide their breakdown by type and raw material, a more completed description of the ground stone assemblage can be found in Beller et al. (2016a) and in the forthcoming Tell es-Safi/Gath III volume. 3.1 Typology The ground stone assemblage contains a variety of types and in varied quantities (Figures 3-4; Table 1). The most numerous are lower grinding stones (n=30), of which the majority (n=26) are fractured, leaving four 124

J. A. Beller et al.: Raw material variety and acquisition of the EB III ground stone assemblage

Figure 3. Typology of ground stone assemblage (by count).

Figure 4. Typology of ground stone assemblage (by % frequency).

125

Stone Tools in the Ancient Near East and Egypt

Locus

Basket Phase

114112

1141073

74503 74505 74505 74505 74506 74509 74808 74808 94109 74511 104203 104801 114306 114402 134604 134709 E15AQ15 16E73D02 16E84C04 16E93A06 16E93A06 16E93B08 16E93B09 16E94A03

E5a

745013 E5a 745019 E5a 745029 E5a 745047 E5a 745050 E5a 745077 E5 748110 E5a 748114 E5a 941057 E5 944090 E5-E6 1042042 E5a 1048013 E5a 1143031 E5b 1144013 E5a 1346093 E5c 1347085 E5a E15AQ179 E6 16E73D004 E5 16E84C012 E5b 16E93A050 E6 16E93A049 E6 16E93B070 E5a 16E93B081 E5a 16E94A011 E5

Material

Material comment

Basalt

Vesicular

Basalt Basalt Basalt Basalt Basalt Basalt Basalt Basalt Basalt Basalt Basalt Basalt Basalt Basalt Basalt Basalt Basalt Basalt Basalt Basalt Basalt Basalt Basalt Basalt

Vesicular Vesicular Vesicular Vesicular Vesicular Vesicular Vesicular Vesicular Vesicular Vesicular Vesicular Vesicular Vesicular Vesicular Vesicular Vesicular Vesicular Vesicular Compact Vesicular Vesicular Vesicular Vesicular Vesicular

74505

745036

E5a

Brecciated chert

84105 94505

841014 945033

E5 E5b

Brecciated chert Brecciated chert

94605

946016

E5b

Brecciated chert

94411

1146028

E5a

Brecciated chert

114402

1144066

E5a

Limestone

74602 74808 74808 94605 94605 74505 74504

746018 748121 748122 946024 946028 745066 745009

E5 E5a E5a E5b E5b E5 E5

Limestone Limestone Limestone Limestone Limestone Chalk Chalk

16E83C07 16E83C087

E5a

Basalt

E15AM09 74606 74517 74519 104303 134302 134307 134307 134307 16E83A03 74606 84015 94410 104111 134804

E5c E5b E5 E5 E5a E5 E5c E5c E5 E5 E5b E5a E5a E5 E5c

Chalk Chalk Chalk Chalk Chalk Chalk Chalk Chalk Chalk Chalk Limestone Limestone Limestone Limestone Limestone

E15AM116 746039 745117 745125 1043066 1343006 1343020 1343165 1343067 16E83A023 746085 840072 944094 1041044 1348043

Compact

Type

Type comment

Indeterminate Fragment grinding stone Lower grinding stone Miscellaneous slab/quern Lower grinding stone Slab/quern fragment Lower grinding stone Slab/quern fragment Lower grinding stone Slab/quern fragment Lower grinding stone Miscellaneous slab/quern Lower grinding stone Slab/quern fragment Lower grinding stone Slab/quern fragment Lower grinding stone Slab/quern fragment Lower grinding stone Boulder slab/quern Lower grinding stone Slab/quern fragment Lower grinding stone Slab/quern fragment Lower grinding stone Slab/quern fragment Lower grinding stone Miscellaneous slab/quern Lower grinding stone Slab/quern fragment Lower grinding stone Slab/quern fragment Lower grinding stone Slab/quern fragment Lower grinding stone Slab/quern fragment Lower grinding stone Slab/quern fragment Lower grinding stone Basin slab/quern Lower grinding stone Slab/quern fragment Lower grinding stone Slab/quern fragment Lower grinding stone Slab/quern fragment Lower grinding stone Slab/quern fragment Lower grinding stone Boulder slab/quern Saddle-shaped grinding Lower grinding stone slab Lower grinding stone Slab/quern fragment Lower grinding stone Boulder slab/quern Saddle-shaped grinding Lower grinding stone slab Lower grinding stone Slab/quern fragment Saddle-shaped grinding Lower grinding stone slab Mortar/socket Boulder mortar? Mortar/socket Boulder mortar? Mortar/socket Bowl mortar Mortar/socket Boulder mortar Mortar/socket Boulder mortar Ornament Bead Ornament Pendant Miscellaneous weight Perforated somwehat askew Perforated Preform disk Perforated Miscellaneous weight Perforated Spindle whorl? Perforated Miscellaneous weight Perforated Spindle whorl Perforated Spindle whorl Perforated Spindle whorl Perforated Spindle whorl Perforated Spindle whorl? Perforated Spindle whorl? Perforated Miscellaneous weight Perforated Preform disk Perforated Miscellaneous weight Perforated Miscellaneous weight Perforated Miscellaneous weight

Weight Length Width (g) (cm) (cm) 157,0

5,0

4,0

1418,5 311,0 455,0 657,5 735,0 207,0 762,0 380,0 2578,0 640,5 327,5 418,5 1292,5 469,5 104,0 182,5 57,5 471,5 8457,5 139,0 315,5 123,0 270,5 17579,5

16,5 11,0 9,0 12,5 15,5 6,5 14,5 10,0 23,0 11,0 9,0 13,0 19,0 11,0 5,5 4,5 5,0 11,5 29,0 6,0 7,0 6,5 9,0 42,0

16,0 8,0 8,0 6,5 11,0 5,5 12,0 7,5 21,0 10,0 5,0 10,0 16,0 7,0 4,0 4,0 4,0 6,5 26,0 6,0 7,0 6,0 9,0 22,0

55,0

27,0

1340,0 6676,0

15,0 20,0

13,0 20,0

22866,5

55,0

27,0

16,0

13,0

19085,0

41,0

28,0

7003,5 16088,5 4980,5 17141,5 12431,5 3,5 6,0

28,0 31,0 21,0 36,0 30,0

13,0 24,0 15,0 28,0 23,0

3,0

2,0

5,5 5,0

5,0 2,5

4,0 6,5

3,0 6,0

8,0

5,5

14,5 151,0 857,5 16,5 39,0 19,5 16,0 11,5 132,5 9,0 1741,5 82,5 244,5 1858,0 223,0

Table 1a. Selected characteristics of EB III Tell es-Safi/Gath ground stone catalogue.

126

J. A. Beller et al.: Raw material variety and acquisition of the EB III ground stone assemblage

Locus

Basket Phase

Material

E15AQ02 16E93B03 74606 104406 84025 74606 74606 84503 94409 94605 94605 104204 104803 114313 114505 74808 104902 E15AT04 84025 74505 74610 74808 84810 104116 104202 104601 114503 16E84A03 94410 16E93B09 74808 104801

E15AQ017 16E93B011 746080 1044016 840111 746051 746066 845009 944072 946017 946019 1042011 1048035 1143047 1145023 748124 1049011 E15AT086 840113 745098 746060 748093 848094 1041041 1042024 1046004 1145010 16E84A007 944076 16E93B060 748100 1048008

E5a E5a E5b E5a E5a E5b E5b E5 E5a E5b E5b E5 E5c E5 E5c E5a E5b E5a E5a E5a E5b E5a E5 E5 E5a E5a E5c E5b E5a E5a E5a E5a

Limestone Limestone Chalk Limestone Basalt Brecciated chert Brecciated chert Brecciated chert Brecciated chert Brecciated chert Brecciated chert Brecciated chert Brecciated chert Brecciated chert Brecciated chert Limestone Limestone Limestone Limestone Basalt Basalt Basalt Basalt Basalt Basalt Basalt Basalt Basalt Beachrock Brecciated chert Limestone Sandstone

74517

845008

E5a

Limestone

114402

1144004

E5a

Siltstone

114303 114106

1143010 1141058

E5b E5a

Basalt Limestone

Material comment

Vesicular

Vesicular Compact Vesicular Compact Compact Vesicular Vesicular Vesicular Vesicular Coquina

Arkose Greyblack Finegrained, some fissility Compact

Type

Type comment

Perforated Perforated Pestle Pestle Pounder Pounder Pounder Pounder Pounder Pounder Pounder Pounder Pounder Pounder Pounder Pounder Pounder Rubber/muller Rubber/muller Upper grinding stone Upper grinding stone Upper grinding stone Upper grinding stone Upper grinding stone Upper grinding stone Upper grinding stone Upper grinding stone Upper grinding stone Upper grinding stone Upper grinding stone Upper grinding stone Upper grinding stone

Miscellaneous weight Miscellaneous weight Bipolar cylindrical pestle Multi-purpose tool Spheroid Spherical/irregular Spherical/irregular Spherical/irregular Spheroid Spheroid Spheroid Spheroid Cuboid Spherical/irregular Spheroid Spherical/irregular Spherical/irregular Unifacial Unifacial loaf handstone Bifacial rectilinear/flat Unifacial loaf handstone Irregular handstone Bifacial loaf/triangular Handstone fragment Handstone fragment Unifacial loaf handstone Unifacial loaf handstone Unifacial loaf handstone Handstone fragment Bifacial rectilinear/flat Unifacial loaf handstone

Weight Length Width (g) (cm) (cm) 623,5 583,5 38,0 874,5 221,0 294,0 411,0 552,5 265,5 355,0 256,5 363,5 251,0 411,0 236,0 406,0 2893,5 146,0 314,0 721,0 1004,0 497,0 463,5 1522,5 62,0 174,5 190,5 1160,5 1093,0 447,5 628,5 1176,5

5,5 12,5 5,0 8,0 5,5 8,5 6,0 6,0 5,0 6,5 5,0 5,5 5,0 5,0

5,0 5,5 5,5 8,0 5,0 6,0 5,0 6,0 5,0 5,0 4,5 5,0

6,5

4,0

11,0 12,0 10,0 11,5 16,5 4,5 9,5 7,0 14,0 14,0 8,0 10,0 10,0

9,0 7,0 8,5 9,0 8,5 3,0 5,0 7,5 11,0 7,5 7,0 7,0 15,0

Varia

383,0

10,0

9,0

Varia

1348,5

15,0

10,0

121,0 65,0

6,5 4,5

6,5 4,0

Vessel Vessel

Rim fragment Rim fragment

Table 1a. Continued.

Locus 114112 74503 74505 74505 74505 74506 74509 74808 74808 94109 74511 104203

Perfo- PerfoThick- CircumOuter Inner Perforation % Height Depth ration ration Basket ness ference diameter diameter diameter 1 Condition remai(cm) (cm) diameter degree (cm) (cm) (cm) (cm) (cm) ning 2 (cm) (cm)

1141073 745013 745019 745029 745047 745050 745077 748110 748114 941057 944090 1042042

3,0 4,0 3,5 4,0 5,0 4,0 4,0 3,0 4,0 7,0 3,0 3,0

Fractured Fractured Fractured Fractured Fractured Fractured Fractured Fractured Fractured Fractured Fractured Fractured

Table 1b. Continued.

127

5 20 5 10 10 35 5 35 10 35 10 5

Stone Tools in the Ancient Near East and Egypt

Locus

Perfo- PerfoThick- CircumOuter Inner Perforation % Height Depth ration ration Basket ness ference diameter diameter diameter 1 Condition remai(cm) (cm) diameter degree (cm) (cm) (cm) (cm) (cm) ning 2 (cm) (cm)

104801 114306 114402 134604 134709 E15AQ15 16E73D02 16E84C04 16E93A06 16E93A06 16E93B08 16E93B09 16E94A03 74505 84105 94505 94605 94411 114402 74602 74808 74808 94605 94605 74505 74504 16E83C07 E15AM09 74606 74517 74519 104303 134302 134307 134307 134307 16E83A03 74606 84015 94410 104111 134804 E15AQ02 16E93B03 74606

1048013 1143031 1144013 1346093 1347085 E15AQ179 16E73D004 16E84C012 16E93A050 16E93A049 16E93B070 16E93B081 16E94A011 745036 841014 945033 946016 1146028 1144066 746018 748121 748122 946024 946028 745066 745009 16E83C087 E15AM116 746039 745117 745125 1043066 1343006 1343020 1343165 1343067 16E83A023 746085 840072 944094 1041044 1348043 E15AQ017 16E93B011 746080

104406

1044016

84025 74606 74606 84503 94409 94605 94605 104204 104803

840111 746051 746066 845009 944072 946017 946019 1042011 1048035

3,0 4,0 3,5 4,0 4,0 2,0 4,0 12,0 4,0 4,0 3,5 2,5 10,0 12,0 5,0 11,0 10,0 7,0 10,5 15,0 14,0 10,0 14,0 13,5

5,0

86,0

1,0

0,5

5,0

27,0

1,5

1,5 47,0

5,0 4,5 5,0 6,0 5,0 6,0 5,0 6,0 5,0

4,0 23 face, 18 shaft 17,0 16,5 22,0 26,0 18,0 20,0 18,0 20,0 17,5

6,0 3,0 4,5 10,0 6,5

1,5 5,5 1,5 2,0 2,0 1,5 1,5 1,5 1,5 5,5 2,5 4,0 3,0 3,0 6,5

6,5

15,5 9,5 17,0

12,0 2,0