The Rurual Hinterland of Antipatris from the Hellenistic to the Byzantine Periods 1803275278, 9781803275277

Citation preview

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods

Amit Shadman

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods Amit Shadman

Archaeopress Archaeology

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

ISBN 978-1-80327-527-7 ISBN 978-1-80327-528-4 (e-Pdf) © Amit Shadman and Archaeopress 2023 Cover: Wadi al-Bureid. Aerial view after excavation (Sky View, A. Shadman). Editor: Salome Dan-Goor Translated from hebrew by Viviana R. Moscovich Typesetting and Layout: Ann Buchnick-Abuhav

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

Contents

Introduction...................................................................................................................................................... 1 The Area; Site Names and Map References; Definition and Classification of the Sites in the Rural Landscape, and Finds; The Agricultural Plot Chapter 1 – Environmental Background.......................................................................................................... 5 Topography; Geology; Soils; Climate; Water; Flora; Fauna; Summary Chapter 2 – History of Research..................................................................................................................... 13 Chapter 3 – The Sites...................................................................................................................................... 15 No. 1. Qasr es-Sitt........................................................................................................................................................... 16 No. 2. Wadi al-Bureid (510)........................................................................................................................................... 19 No. 3. H. Te’ena.............................................................................................................................................................. 36 No. 4. Naḥal Shiloh........................................................................................................................................................ 50 No. 5. Naḥal Shiloh-West (740) ................................................................................................................................... 52 No. 6. Wadi al-Bureid-East............................................................................................................................................ 56 No. 7. Kh. Umm al-Bureid............................................................................................................................................. 57 No. 8. H. Dayyar.............................................................................................................................................................. 60 No. 9. H. Yeqavim........................................................................................................................................................... 62 No. 10. Wadi al-Bureid (South).................................................................................................................................... 67 No. 11. Kh. Umm al-Hammam..................................................................................................................................... 69 No. 12. Naḥal Shiloh (Southwest................................................................................................................................. 74 No. 13. Kh. Kesfa............................................................................................................................................................. 76 No. 14. Migdal Afeq........................................................................................................................................................ 83 No. 15. Qurnat al-Haramiya......................................................................................................................................... 89 No. 16. Elevation Point 136........................................................................................................................................... 92 No. 17. Naḥal Shiloh (2)................................................................................................................................................ 93 No. 18. Kh. Tukkim........................................................................................................................................................ 94 No. 19. Kh. Ta‘amur....................................................................................................................................................... 96 No. 20. Naḥal Shiloh (3).............................................................................................................................................. 100 No. 21. Naḥal Shiloh (4) ............................................................................................................................................. 103 No. 22. Khallat es-Sihrij.............................................................................................................................................. 104 Chapter 4 – The Archaeological Survey - Methodology............................................................................... 107 Introduction; Site Definition; Types of Surveys; Development Surveys in the Rosh Ha‘ayin Area; Reliability of Surveys vs. Excavations in Rosh Ha‘ayin; Summary Chapter 5 – Landscape Archaeology............................................................................................................. 111 Introduction; The Question of Dating; Defining the boundaries of the settlements in the area; Boundary Stones; Summary

i

Chapter 6 – Roads and Agricultural Systems................................................................................................117 Roads.............................................................................................................................................................................. 117 International Roads; ‘The Via Maris’; ‘National’ Longitudinal Roads; Latitudinal Regional Roads; Rural Roads; Dating the Rural Roads; Rural Roads and Field Towers; Rural Roads and their Contribution to the Development of the Rural Countryside; Rural Roads and Transportation Means; The Study Area and its Main Rural Roads; Summary Agricultural Installations........................................................................................................................................... 132 Introduction; Agricultural Terraces; Dams; Enclosed Plots; Dividing Walls and Enclosures; Dating the Agricultural Systems; OSL summary of method; The Plots’ Outline as a Dating Marker The Wine Industry . .................................................................................................................................................... 141 Simple irregular winepress; simple asymmetrical winepress; simple winepress with a settling pit; simple winepress with a mosaic; Large winepress with mosaic; The treading floors; The collecting vats; A rock-cut vineyard; Dating the winepresses; The location of the winepresses as a dating tool and as a settlement marker; The location of winepresses within the landscape; Summary Field Towers................................................................................................................................................................. 150 Dating the field towers; Field Towers as Mausolea; Winepresses and Field Towers; Summary Water Cisterns.............................................................................................................................................................. 160 Dating the Cisterns; The cisterns in ‘the farmhouses strip’; The cisterns in ‘the rural towns strip’; Cisterns and rural roads; Summary Stone Clearance Heaps................................................................................................................................................ 162 Stone clearance heaps on farmhouses Cultivation potential of vineyards, olive groves, and grains in the area........................................................... 168 Agricultural Systems - Summary.............................................................................................................................. 171 Chapter 7 – Settlement Patterns....................................................................................................................173 Introduction; Settlement types – 1. Farms and Estates; 2. Satellite Villages; 3. ‘Rural Towns’ - Medium-large-Sized Rural Settlements; 4. Fortresses; 5. Rural Monasteries and Churches; H. Te’ena—A Case Study; Settlement Patterns Through the Periods; The Persian Period; The Hellenistic period; The Roman period; The Byzantine period; The Early Islamic period; Summary Chapter 8 – Comparative Study with the area South of Naḥal Shiloh..........................................................197 Introduction; The Main Research Area—between Naḥal Rabah and Naḥal Shiloh; The Area Between Naḥal Shiloh and Naḥal Mazor; Settlement Patterns; Discussion and Summary Chapter 9 –The Estimate Population in the Region.......................................................................................203 Introduction; The Population Count in Hellenistic Period Farms; The Population Count in the Roman Period; The Population in ‘The Rural-towns Strip’ and The Monasteries at H. Te’ena, Kh. Umm Al-Bureid and Kh. Dayyar during the Byzantine Period; Summary Chapter 10 – Historical Sources and Archaeological Finds...........................................................................209 The Iron Age II; The Babylonian and Persian Periods; The Hellenistic Period; The Hasmonean Period; The Roman Period; The Bar Kochba Revolt; The Byzantine Period; The Early Islamic Period Conclusions....................................................................................................................................................221 Bibliography..................................................................................................................................................223 List of Abbreviations......................................................................................................................................243

ii

List of Figures

Fig. 1.1. Location map of the study area.................................................................................................................................... 2 Fig. 1.2. The subdivision of the study area................................................................................................................................ 3 Fig. 2.1.1. Geological map of the study area (Geological Survey, State of Israel)................................................................ 6 Fig. 2.1.2. Lithostratigraphy of the study area (Geological Survey, State of Israel) .......................................................... 6 Fig. 2.2. The soil groups (Gal and Ramon 2012......................................................................................................................... 7 Fig. 2.3. A section in Terrace 1 (O. Ackermann)....................................................................................................................... 8 Fig. 2.4. A section in Terrace 2 (O. Ackermann) ...................................................................................................................... 8 Fig. 2.5. Map of Flora in the study area (Gal and Ramon 2012)............................................................................................11 Fig. 2.6. A gazelle in the Rosh Ha’ayin excavations project (A. Shadman)........................................................................12 Fig. 3.1. The Sites.........................................................................................................................................................................15 Fig. 3.2.1. Qasr es-Sitt. Areal view of the farm (Courtsey of H. Cohen-Klonymus and E. Emanuelov)..........................16 Fig. 3.2.2. Qasr es-Sitt. Plan of the farm (Courtsey of H. Cohen-Klonymus and E. Emanuelov).....................................17 Fig. 3.3. Qasr es-Sitt. The agricultural plot (Courtsey of H. Cohen-Klonymus and E. Emanuelov)................................18 Fig. 3.4.1. Wadi al-Bureid. Plan.................................................................................................................................................20 Fig. 3.4.2. Wadi al-Bureid. Sections to plan.............................................................................................................................21 Fig. 3.5. Wadi al-Bureid. Room I, looking southeast..............................................................................................................21 Fig. 3.6. Wadi al-Bureid. Room I, the drainage channel........................................................................................................22 Fig. 3.7. Wadi al-Bureid. The entrance from the northern courtyard to Room X, looking south..................................22 Fig. 3.8. Wadi al-Bureid. Room X, looking west......................................................................................................................23 Fig. 3.9. Wadi al-Bureid. The sitting bath, looking north.....................................................................................................23 Fig. 3.10. Wadi al-Bureid. The doorway between Room X and XIII, looking northwest..................................................24 Fig. 3.11. Wadi al-Bureid. A stone altar discovered in Room XI..........................................................................................24 Fig. 3.12. Wadi al-Bureid. Aerial view after the excavation (Sky View, A. Shadman)......................................................25 Fig. 3.13. Wadi al-Bureid. A stone clearance heap containing a built-in hewn cell.........................................................26 Fig. 3.14. Wadi al-Bureid. Mortaria (1-4) and bowls (5-21)...................................................................................................27 Fig. 3.15. Wadi al-Bureid. A krater (1) and cooking pots (2-8).............................................................................................29 Fig. 3.16. Wadi al-Bureid. Holemouth (1) and Jars (2-23)......................................................................................................30 Fig. 3.17. Wadi al-Bureid. Jars....................................................................................................................................................32 Fig. 3.18.1. Wadi al-Bureid. Various vessels............................................................................................................................34 Fig. 3.18.2. Coin from Hellenistic period (Ptolemy I)............................................................................................................34 Fig. 3.19. H. Te’ena. Aerial view (Sky View, A. Shadman).....................................................................................................36 Fig. 3.20.1. H. Te’ena. Plan.........................................................................................................................................................37 Fig. 3.20.2. H. Te’ena. Inscription (A. Peretz)..........................................................................................................................38 Fig. 3.20.3. H. Te’ena. The stable (A. Peretz)...........................................................................................................................39 Fig. 3.21.1. H. Te’ena. The olive press (A. Shadman).............................................................................................................40 Fig. 3.21..2. H. Te’ena. The underground cavity (A. Peretz).................................................................................................41 Fig. 3.22. H. Te’ena. Land distribution.....................................................................................................................................42 Fig. 3.23. H. Te’ena. The agricultural systems north of the site..........................................................................................44 Fig. 3.24. H. Te’ena. Winepress No. 461 (50) . .........................................................................................................................45 Fig. 3.25. H. Te’ena. The threshing floor, a winepress, and an enclosed wall (Skyview, A. Shadman)..........................46 Fig. 3.26. H. Te.ena. Pottery.......................................................................................................................................................48 Fig. 3.27. H. Te.ena. Pottery.......................................................................................................................................................48 Fig. 3.28.1. Naḥal Shiloh. Aerial view of the farm, looking northwest (Skyview, A. Shadman).....................................51 iii

Fig. 3.28.2. Naḥal Shiloh. A 4th century BCE silver coin minted in Sidon..........................................................................51 Fig. 3.29. Naḥal Shiloh. Plan......................................................................................................................................................52 Fig. 3.30. Naḥal Shiloh (West). Aerial view of the farm (Skyview, A. Shadman)...............................................................53 Fig. 3.31. Naḥal Shiloh (West). A Hellenistic silver coin. Left: the head of Hercules; right: Zeus..................................54 Fig. 3.32. Naḥal Shiloh (West). A winepress (Skyview, A. Shadman)..................................................................................55 Fig. 3.33. Naḥal Shiloh (West). A water cistern northwest of the farm..............................................................................55 Fig. 3.34. Naḥal Shiloh (West). Ceramic finds. Iron Age II: hemispheric bowls (1-3), cooking pots (4-5); Persian period: mortaria (6-7), cooking pots (8-10), jars (11-13), jug (14); Hellenistic period: jars (15-16).................56 Fig. 3.35. Wadi al-Bureid (East). The western building, looking southwest (A. Shadman).............................................57 Fig. 3.36. Kh. Umm al-Bureid.....................................................................................................................................................58 Fig. 3.37. Kh. Umm al-Bureid. The reservoir, areal view (A. Shadman).............................................................................59 Fig. 3.38. Kh. Umm al-Bureid. Ceramic finds from Kochavi and Beit Arieh 1994: Site 108. Byzantine period bowls (1-2), a jug (3), a juglet (4); Early Islamic Period: a bowl (5), a jug (6).................................................................................59 Fig. 3.39. H. Dayyar. Plan............................................................................................................................................................60 Fig. 3.40. H. Dayyar. The central room in the western complex, looking west (A. Shadman)........................................61 Fig. 3.41. H. Yeqavim. Aerial view (Skyview, A. Shadman)...................................................................................................62 Fig. 3.42. H. Yeqavim. Plan of a large building on the southwest part of the site............................................................63 Fig. 3.43. H. Yeqavim. Land distribution.................................................................................................................................64 Fig. 3.44. H. Yeqavim. Fragments of a sarcophagus lid’ looking north (A. Shadman) ....................................................66 Fig. 3.45. H. Yeqavim. Ceramic finds from the church area. Byzantine period: bowls (1-4); Early Islamic period: a jar (5) ...................................................................................................................................................67 Fig. 3.46. Wadi al-Bureid (south). Areal view (M. Haiman and A. Shadman).....................................................................68 Fig. 3.47. Wadi al-Bureid (south), looking southeast (A. Shadman)....................................................................................69 Fig. 3.48. Wadi al-Bureid (south). An oil press lower crushing stone (A. Shadman)........................................................69 Fig. 3.49. Kh. Umm al-Hammam. Plan.....................................................................................................................................70 Fig. 3.50. Kh. Umm al-Hammam. A Mamluk/ Ottoman structure (M. Haiman)...............................................................72 Fig. 3.51. Kh. Umm al-Hammam. Threshing floors and sheep pens (Skyview, A. Shadman).........................................73 Fig. 3.52. Naḥal Shiloh (Southwest). Aerial view (Skyview, A. Shadman)..........................................................................75 Fig. 3.53. Naḥal Shiloh (Southwest). Plan (M. Haiman and A. Shadman)...........................................................................75 Fig. 3.54. Kh. Kesfa. Plan (Safrai and Fixler 1993)..................................................................................................................77 Fig. 3.55. Kh. Kesfa. General view, looking south (A. Shadman)..........................................................................................78 Fig. 3.56. Kh. Kesfa. Land distribution.....................................................................................................................................79 Fig. 3.57. Kh. Kesfa. A rock-cut winepress near Rafat-Migdal Afeq’s rural road (A. Shadman)......................................81 Fig. 3.58. Kh. Kesfa. A small quarry east of the site (A. Shadman)......................................................................................81 Fig. 3.59. Kh. Kesfa. Ceramic finds: A Byzantine LRC bowl and a jar...................................................................................82 Fig. 3.60. Migdal Afeq. Land distribution................................................................................................................................85 Fig. 3.61. Migdal Afeq. The ‘kokhim cave’ from the end of the Second Temple period, looking southwest................86 Fig. 3.62. Migdal Afeq. Plan of ‘The Magnificent Cave’ (right) and the ‘kokhim cave’ (left, Tsuk 1993........................86 Fig. 3.63. Migdal Afeq. Ceramic and stone vessels (courtesy of A. Tendler). Iron Age: Bowl (1), jars (2-3); Persian and Hellenistic period: mortaria (4), jars (5-6), jugs (7); Early Roman period: cooking pots (8-10), jars (11-13), juglets (14-15); Byzantine period: LRC bowl (16); Crusader period: glazed bowls (17); Early Roman period: ston vessels (18-21)................................................................................................................................88 Fig. 3.64. Qurnat al-Haramiya. Site plan after the 1995 excavations (Torge 2018)...........................................................89 Fig. 3.65. Qurnat al-Haramiya. The ‘compound wall’, looking south (courtesy of A. Tendler)......................................90 Fig. 3.66. Qurnat al-Haramiya. Ceramic finds from the ‘compound’ )Hadad 2011). Iron Age: bowl (1), chalice (2), cooking pot (3), jar (4); Persian period: jars (5-7); Early Hellenistic period: bowls (8), cooking pots (9-10), jars (11)...................................................................................................................................90 Fig. 3.67. Elevation point 136. Plan...........................................................................................................................................92 Fig. 3.68. Elevation point 136. Areal view................................................................................................................................93 Fig. 3.69. Naḥal Shiloh (2). Plan.................................................................................................................................................93 Fig. 3.70. Kh. Tukkim. Plan.........................................................................................................................................................94 Fig. 3.71. Kh. Tukkim. A pressing bed of an oil press............................................................................................................95 Fig. 3.72. Kh. Ta’amur. Plan (after Safrai 1997a:227).............................................................................................................96 Fig. 3.73. Kh. Ta’amur. The eastern gate, looking west (A. Shadman)................................................................................97 Fig. 3.74. Kh. Ta’amur. Land distribution................................................................................................................................99 Fig. 3.75. Naḥal Shiloh (3). Plan...............................................................................................................................................101 Fig. 3.76. Naḥal Shiloh (3). Ceramic finds. Iron Age: bowls (1-5), holemouth jars (9-10); Persian period: a krater (6), cooking pots (7-8), jars (11-13), a jug (14).............................................................................................................................102 iv

Fig. 3.77. Naḥal Shiloh (4). Plan . ............................................................................................................................................103 Fig. 3.78. Khallat es-Sihrij. Aerial view (photo credit: Gilad Itach – IAA).........................................................................104 Fig. 3.79.1. Khallat es-Sihrij. The entrance to the water reservoir. (photo credit: Gilad Itach – IAA).........................104 Fig. 3.79.2. Khallat es-Sihrij. Water reservoir, looking outside. (photograph: A. Peres – IAA).....................................105 Fig. 4.1. Rosh Ha’ayin. Map of archaeological features found in development survey (courtesy of the Israel Antiquities Authorities)...................................................................................................................109 Fig. 5.1. H. Yeqavim and H. Te’ena-Deir Balut road.............................................................................................................112 Fig. 5.2. Estimated boundaries of the Byzantine settlements. The dotted and full lines represent the boundaries as reconstructed by the author; the colored areas are the boundaries generated by the computer-generated map........................................................................................................................114 Fig. 5.3. Upper right: an ancient boundary marker of an agricultural plot. Upper left: a boundary marker of the 1949 ‘Green Line’. Bottom: an ancient boundary marker of a village (A. Shadman).....................................................114 Fig. 6.1. Agricultural systems in the study area...................................................................................................................117 Fig. 6.2. The main roads in the region (after Tsafrir, Di Segni and Green 1994).............................................................120 Fig. 6.3. Gophna-Antipatris road. A milestone found in the excavations (courtesy of O. Sion)...................................121 Fig. 6.4. Rural and agricultural roads.....................................................................................................................................123 Fig. 6.5. Road No. 1. Naḥal Shiloh – Kh. Dayyar (A. Shadman)...........................................................................................124 Fig. 6.6.1. Road No.2. Plan of the southern section..............................................................................................................124 Fig. 6.6.2. Road No.2. Plan of the southern section..............................................................................................................125 Fig. 6.7. Road No. 2, the southern section (Survey of Israel)..............................................................................................125 Fig. 6.8. Map of the area in the 1940s.....................................................................................................................................126 Fig. 6.9. Road No. 4. Migdal Afeq - H. Yeqavim Road. Looking west (A. Shadman)........................................................126 Fig. 6.10. Road No. 6. Kh. Umm al-Hammam – Road No. 2 (Skyview, A. Shadman)........................................................127 Fig. 6.11. Road No. 7. Wadi al-Bureid - H. Yeqavim Road. Looking north (A. Shadman) ..............................................128 Fig. 6.12. A dammed gully and a terraced slope west of H. Yeqavim (map ref. 198730-664100)..................................134 Fig. 6.13. A dam between H. Te’ena and H. Yeqavim (map ref. 198860-664740; A. Shadman)......................................134 Fig. 6.14. The drainage basin of the region’s streams.........................................................................................................135 Fig. 6.15. Dammed wadies........................................................................................................................................................136 Fig. 6.16. OSL tests. Above: A probe northeast of Migdal Afeq. Below: A probe near a farmhouse at site No. 5.......138 Fig. 6.17. Enclosed plots. The asymmetrical plots (upper right) belong to modern Deir Balut; the symmetrical plots belong to the Byzantine settlement of H. Yeqavim....................................................................140 Fig. 6.18. The distribution of winepresses in the study area.............................................................................................142 Fig. 6.19. Four irregular simple winepresses........................................................................................................................143 Fig. 6.20. A simple mosaic winepress, west of H. Te’ena.....................................................................................................144 Fig. 6.21. Distribution of the winepress according to the dimensions of their collecting vats and treading floors........................................................................................................................................145 Fig. 6.22. Distribution of the winepress according to the capacity of their collecting vats.........................................145 Fig. 6.23. A rock-cut vineyard.................................................................................................................................................146 Fig. 6.24. Distribution of field towers.....................................................................................................................................151 Fig.6.25.1. Plan of selected excavated field towers..............................................................................................................152 Fig.6.25.2. Plan of selected excavated field towers..............................................................................................................153 Fig. 6.26. A rounded field tower east of H. Te’ena...............................................................................................................153 Fig. 6.27. Distribution of the field towers according to their outline and dimensions..................................................154 Fig. 6.28. Distribution of the field towers and winepresses...............................................................................................156 Fig. 6.29. Distribution of cisterns in the study area.............................................................................................................161 Fig. 6.30. Water cisterns and rural roads...............................................................................................................................162 Fig. 6.31. Selected plans of excavated stone clearance heaps............................................................................................167 Fig. 6.32. A stone clearance heaps with chives (A. Shadman)............................................................................................168 Fig. 6.33. Cappadocia. Vines growing on stone heaps.........................................................................................................169 Fig. 6.34. The location of agricultural crops in the study area..........................................................................................171 Fig. 7.1. Settlement distribution in the Persian period.......................................................................................................178 Fig. 7.2. Settlement distribution in the Hellenistic period.................................................................................................179 Fig. 7.3. Settlement distribution in the Roman period.......................................................................................................183 Fig. 7.4. Settlement distribution in the Byzantine period..................................................................................................185 v

Fig. 7.5. The boundaries of the Byzantine settlements in the study area........................................................................186 Fig. 7.6. Distribution of enclosed plots in the Byzantine period.......................................................................................187 Fig. 7.7. The distribution of graves in the study area..........................................................................................................189 Fig. 7.8. Cist tombs west of Kh. Kesfa (E. Marco)..................................................................................................................191 Fig. 7.9. Settlement distribution in the Early Islamic period.............................................................................................193 Fig. 7.10. Site distribution by period......................................................................................................................................194 Fig. 8.1. Schumacher’s map (1918).........................................................................................................................................200 Fig. 10.1. ‘The farm strip’. Black dots: Farms excavated by the author or surveyed by Finkelstein (1978a, 1981). Red dots: Farms established in the Hellenistic period .........................................................213

vi

To my Parents Elizabeta and Benjamin Shadman

vii

This volume is based on my doctoral dissertation, supervised by Prof. Shimon Dar of Bar-Ilan University. I want to thank Prof. Dar for his patience and pleasant nature; his friendly attitude and vast knowledge turned the work into a real pleasure. It was a great privilege to be escorted by him on this long, thoughtful and exciting journey. Amit Shadman

viii

INTRODUCTION

The Area

C. The ‘Rural Towns’ Strip (medium- to large-sized villages). This strip includes, among other things, four relatively large sites, probably founded in the Byzantine period. Due to their considerable size, it is suggested to call them ‘rural towns,’ although these are clearly medium-large villages. These are, from north to south: Kh. Kesfa Kh. Umm al-Hammam, H. Yeqavim, and Kh. Ta‘amur. Also included in this strip are the monasteries of Kh. Dayyar, Kh. Umm al-Bureid, and H. Te’ena, as well as some farmhouses or estates located near the ‘rural towns.’

The region under discussion extends for approximately 30 sq km (Fig. 1.1), and it is a distinct geographical area delimited by the Shiloh river (Naḥal Shiloh) Shiloh on the south, the Rabah river (Naḥal Rabah) on the north, and by the eastern part of the Coastal Plain on the west. On the east, the area is bounded by the ‘Green line’— the pre-1967 border between Israel and Jordan. The geographic characteristics of the region influenced the nature of its settlements (land distribution, road system, administrative units, and such). It is worth noticing that some sites in the study area, discovered both in previous surveys and by the author, were not included in the present study, either due to their poor state of preservation which did not allow their plans to be drawn or because the sites had been utterly destroyed by contemporary works. Nevertheless, important destroyed sites were taken into account in the discussion of each period.

D. The Eastern Strip. This strip includes the sites of Rafat, Deir Balut, and Ez-Zawiya, and to some extent also al-Luban and Rantis. Only limited data is available on these sites although, according to the survey conducted in the region by Kochavi (1972: 231), the first two were settled during most historical periods. It should be noted that this strip of settlements slightly deviates from the boundaries of the present study, but they are an essential component for the overall picture.2

The area was subdivided into four settlement strips, and the sites were defined along these strips according to the following divisions and definitions (Fig. 1.2):1

Site Names and Map References

A. The Western Strip. This strip includes the sites of Migdal Afeq and Qurnat al-Haramiya. Additional sites, located south of the study area, include H. Zikhrin, Mazor, Quleh, Nabalat and Hadid. A conspicuous feature of this strip is the location of the sites just above the alluvial valley of the Coastal Plain.

This study includes twenty-two sites, known by their names in earlier studies, except for site No. 20 which was first discovered during the present work. In some cases, there is a discrepancy between the names appearing in the Survey of Western Palestine and other maps of the 19th century, and in modern maps. For example, the location of the sites of Kh. Dayyar, Kh. Umm al-Bureid and Kh. Umm al-Hammam differs from what is known today. In these cases, we followed Kochavi and Beit Arieh (1994), who surveyed the area in the 1970s. A similar problem was identified with the sites of Qurnat al-Haramiya and Qasr es-Sitt.

B. The Farmsteads Strip. Located east of the Western Strip, this strip includes a series of agricultural farms. Most of these farms have been documented by Finkelstein (1978a, 1981) and dated by him mostly to the Iron Age II, the Persian and the Hellenistic periods. Some farms, excavated by the author, will be presented in Chapter 3.

It should be noted that, in the British maps of the 1940s, the sites bear the same names as today. In several cases, we added a suffix to the name of a site—such as

1  Finkelstein (1978a: 31–33) and Safrai (1995b) have already discussed the subdivision of the region into strips. Finkelstein (1981:341) noted two strips, west and east, with two rows of villages. He named the area between these two strips ‘the intermediate hills’.

2 

A noteworthy feature is the existence, east of this strip, of another important strip of sites which includes the sites of Deir el-Mir, Deir Qal‘ah and Deir Sam‘an (see Chapter 10).

1

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods

Fig. 1.1. Location map of the study area.

2

Amit Shadman: Introduction

Fig. 1.2. The subdivision of the study area.

3

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods

The latter drew the plans of Kh. Kesfa and Kh. Ta‘amur. Other sites were surveyed as part of the present study, and their plans were generated by GPS.

the cardinal direction or a number in parentheses—to differentiate it from others. Map references follow the New Israel Grid, with a precision of 12 digits. Based on the data, several maps were generated presenting the settlements’ distribution through time, with the placement of installations such as field towers, winepresses, burial caves, cisterns and more. On the installations maps, the points of reference were occasionally shifted a few meters to make the map more accessible.

Within the framework of the present research, we conducted an in-depth survey of several key sites within the region, mostly in the ‘rural towns strip.’ The ceramic finds from these sites are presented at the end of the discussion of each site. The finds excavated by the author of this book are presented and discussed as part of the study of site No. 2. It should be noted that the ceramic finds from this site are similar to those from the other farmhouses in the region.

Definition and Classification of the Sites in the Rural Landscape, and Finds

The Agricultural Plot

The sites in the region were classified according to size. Their analysis and dating were based on the 22 sites included in this study and their agricultural precinct (the area of the plot). Sites smaller than 0.5 dunams,3 such as field towers, limekilns, winepresses, and so on, were discussed as part of the agricultural plot of each settlement and were not taken into account in the calculation of the size of the constructed area. The site area, therefore, includes the central complex of structures and rooms. Burial fields, churches, and such, discovered near the settlement, were included in the agricultural plot of the relevant site. The sizes of the sites and their definitions are as follows:

As stated before, the agricultural plot is composed of the agricultural lands surrounding the settlement. The discussion includes the various installations in the agricultural plot: winepresses, cisterns, burial caves, stone clearance heaps, sheep pens, kilns, and so forth. The delineation of the agricultural plots was done based on the topography of the area (Kallai 1960: 89) and the walking distances from the site to the roads, particularly to the agricultural ones. These roads were intended to link the settlement to the fields located nearby and to allow for the transportation of agricultural products to the settlement or the field towers. Most of the roads were similar to the rural roads, and the main difference between them is in their length, with the agricultural roads estimated from only hundreds of meters to about a maximum of 1 km. These roads were numbered with the site’s number and a suffix (for example, the agricultural roads of H. Yeqavim—site No. 9—were numbered as 9a, 9b, and so forth.). Emphasis was made on the analysis of the enclosed plots of each settlement, especially those of the major sites, such as Kh. Ta‘amur, H. Yeqavim, and Kh. Kesfa. The region as a whole is characterized by numerous field walls, agricultural terraces, dams and enclosed areas which raised many questions, including their construction dating.

A small, 1–3 dunams site: Farmhouses or estates. A medium-sized, 3–5 dunams site: Monasteries and villages. A large, 10–15 dunams or larger site: ‘Rural towns’—medium- to large-sized villages. Defining each site was done taking into account other studies of a similar nature. The main settlement patterns in the area are farmhouses, estates, rural villages, castles, monasteries/churches and ‘rural towns.’ The structural remains of some sites had been already surveyed in the past and accompanied by plans, for example during the salvage excavations conducted by the Israel Antiquites Authority (IAA), and in Safrai’s survey (Safrai 1997a).

3 

A dunam is a mesure of land equal to 1000 square meters.

4

CHAPTER 1

ENVIRONMENTAL BACKGROUND

Topography

shaped its landscape to a large extent, and also affected the nature of the ancient settlements in the area (Figs. 2.1.1, 2.1.2). This geographical strip is, in fact, the lowest level of the Ramallah anticline, characterized by limestone and dolomite rocks from the CenomanianTuronian Era (Edelman and Anmer 2014:100). Three main formations, associated with the Judean Group, are found in the study area‫‏‬:

The Judean and Samaria Hills comprise the longest mountainous region in central Israel. This mountain range extends from Mt. Gilbo‘a in the north/northeast to the area south of Hebron in the south/southwest. The Samaria Hills are, in fact, a lower-elevation extension of the Judean Hills. Except for Mt. Ebal, rising to 940 m above sea level, most of the Samaria Hills rise to less than 900 m above sea level (Har-El and Nir 1995). Their elevation decreases toward their western foothills, where the area is characterized by low hills rising to 100–200 m above sea level.

The Ba‘ana Formation (Turonian Era). This 40–90-m thick formation lies on top of Veradim Formation. It includes two different members: the lower member, made of dolomite, limestone and marl layers, which creates a terraced rocky landscape (Roth and Flexer 1977), and the upper member, composed of thin layers of thin-crystal chalk, which creates a terraced landscape. The Veradim Formation (Upper CenomanianTuronian). This 20–100 m thick formation, also known as ‘Mizi Yahudi’ lies on the Kefar Shaul Formation (Roth and Flexer 1977: 10). It is characterized by thick layers of limestone and gray dolomite and medium-sized crystals, forming a rocky landscape. These are exposed along the western and eastern slopes of the Hebron, Ramallah and Fre‘ah anticlines (Roth and Flexer 1977). The Aminadav Formation (Cenomanian Era), about 100-m thick, is usually exposed along the lower fringes of the Judea and Samaria anticlines, creating a rocky landscape. It is composed of thick layers of dolomite and sometimes limestone (Roth and Flexer 1977: 9). This formation is characterized by karstic structures (Frumkin 1992: 169–176), many of these visible within the study area.

This area is known by a number of names geographically defining it: ‘the Samaria Shephelah’ (Safrai 1997a); ‘The Samaria Foothills’ (Nir 1989: 83); ‘the Intermediate Hills Strip’ (Finkelstein 1978a: 3) and ‘the Mountain’s Foothill.’ The name ‘The Samaria foothills,’ representing a 10 km wide and 25 km long land strip, will be used throughout the present study. This geographical area is bounded by Naḥal Qana to the north, Naḥal Natuf to the south, the Coastal Plain to the west (longitude 195), and longitude 200 to the east. Unlike the Judean Hills, the Samaria hill country is not a homogeneous mountainous bloc, but rather a system of hills bisected by valleys and streams oriented east–west, creating mountainous segments (Har-El and Nir 1995). Five large streams cross the region from east to west: Naḥal Qana (on the north), Naḥal Rabah, Naḥal Shiloh, Naḥal Beit ‘Arif, and Naḥal Natuf. These streams are characterized by steep slopes on their eastern part which become more moderate as they descend westward towards the alluvial Coastal Plain. Geology Since the withdrawal of the sea, the mountain regions have undergone several uplifting and alignment periods (Dan, Fine and Lavee 2007: 269). Mountain formation was accompanied by erosion processes that caused the removal of the young strata, exposing rocks from the Cenomanian-Turonian Era and even from the Lower Cretaceous Era. Samaria’s foothills is characterized by several geological formations which

Thus, the Samaria foothills, composed mainly of the Ba‘ana Formation, extends between Naḥal Rabah and Naḥal Shiloh. The Veradim Formation is visible in the vicinity of Naḥal Shiloh, as well as in the eastern part of the study area. The Aminadav Formation is located mainly north of the study area, in the vicinity of Naḥal Rabah, where many karstic cavities can be discerned. 5

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods

Fig. 2.1.1. Geological map of the study area (Geological Survey, State of Israel).

Fig. 2.1.2. Lithostratigraphy of the study area (Geological Survey, State of Israel).

6

Amit Shadman: Chapter 1, Environmental Background

at a large distance from the desert. Clay soils are also formed by dust originating in chalk, with dust being washed as a result of the relative humidity prevailing in the region (Dan, Fine and Lavee 2007: 271). Most of the soils in the mountainous regions were created on carbonate rocks, mainly hard limestone, dolomite, chalk, and Nari (Dan 1977: 14). Three main types of soils are found in the west Samaria Hills (Fig. 2.2):

Soils Soil formation is influenced by the climatic conditions prevailing in its environment (Ravikovitch 1981), with dust playing a crucial role in the Land of Israel and its mountainous regions. Thin dust components containing a large quantity of clay have deposited in the Samaria Hills and other regions of the Land of Israel, located

Fig. 2.2. The soil groups (Gal and Ramon 2012).

7

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods

– Terra Rossa. Red or red-brown mountainous soil generated by Mediterranean climatic conditions. – Rendzina. Gray or brown-gray mountainous soil generated by the Mediterranean or Mediterranean-desert-plain climate (Ravikovitch 1981). – Grumosol. Brown and red-brown soil, usually found on plains and plateaus (Dan and Raz 1970: 35). The Sharon Plain, west of the Samaria foothills, is characteristic by Ḥamrah (Dan 1977: 15), while the rocky slopes of the Samaria Hills are usually composed of Terra Rossa and Rendzina. The latter are shallow soils, difficult to cultivate. The study area is partly characterized by Terra Rossa, which accumulated in the plains, pocket valleys and wadis (Yaalon 1995: 23). This fertile soil is rich in organic matter (Ravikovitch 1981:73) that improves the soil’s physical-chemical composition and give it a fluffy nature. These conditions enabled the soils to absorb the rainfall, thereby facilitating their use in agriculture.1 The results are as follows: Fig. 2.3. A section in Terrace 1 (O. Ackermann).

Terrace 1 (Fig. 2.3)—filled with 120 cm of soil/sediment on top of the bedrock. From the surface to a depth of 10 cm the soil is dark brown (7.7YR 3/3, dry), with a granular to crumb structure, and a gradual transition to the lower layer/horizon. From 10 to 120 cm in depth, the soil is dark brown (7.7YR 3/3, dry), with a dense prismatic structure with a few angular rock fragments. Signs of secondary calcium carbonate sedimentation appear at 10–40 cm deep. The structure loosens at 90– 120 cm deep. The transition to the bedrock, at 120 cm deep, is abrupt. Terrace 2 (Fig. 2.4)—filled with 90 cm of soil/sediment on top of the bedrock. From the surface to the depth of 5 cm, the soil is dark reddish brown (5YR 3/3, dry), with a crumb structure and a clear transition to the lower layer/horizon. From 5 to 90 cm in depth, the soil is dark reddish brown (5YR 3/3, wet), with a hard prismatic to subangular blocky structure and a few angular 2-cm long rock fragments. Signs of secondary calcium carbonate sedimentation appear between 5 and 35 cm in depth. The transition to the bedrock, at 90 cm deep, is abrupt.

1  Throughout the study, we carried out a geomorphological/ pedological examination of deep soils and sediments in two representative sections cut in agricultural terraces northeast of Migdal Afeq (Figs. 2.2, 2.3). I wish to thank O. Ackermann for carrying out these examinations.

Fig. 2.4. A section in Terrace 2 (O. Ackermann).

8

Amit Shadman: Chapter 1, Environmental Background

Climate

The winds system affected the location of agricultural and industrial installations within the settlements. The by-products of these installations often constituted a nuisance of stench, dirt and air pollution. The importance of the wind system is also demonstrated by the location of kilns and the way they were operated, with their openings facing the blowing wind’s direction (Spanier and Sasson 2001).

Rainfall The rainy season and the seasonal rainfall are a significant factor in selecting crop areas, especially in antiquity (Langgut, Finkelstein and Litt 2013:149– 175). The water system in the Land of Israel was largely dependent on precipitation, besides fossil water (subterranean water of ancient geological eras unrelated to current rainfall). Several factors affect the amount of precipitation and its dynamics:

Dew

Temperatures The location of the Land of Israel in a subtropical region, as well as the scarcity of clouds, cause considerable exposure to solar radiation (Goldreich 1998). Consequently, the region is characterized by a Mediterranean climate—hot and dry summers and rainy winters (Kalbanner 1999:19). The proximity to the sea, the north–south effect, and the elevation of the mountains also affect temperature differences throughout the land. In the summer days, the temperature in the Samaria foothills reaches 32 c, while on the coastline it is about 2 degrees lower. The rising of the mountains eastwards lower temperatures. On summer nights, the temperature drops to 20 c. In winter, the temperature in the Samaria foothills is about 17–18 c. In the higher mountains, winter temperature is about 10 c. or less by day, and about 4–6 c. by night (Kalbanner 1999).

– The North–south effect—the amount of precipitation decreases toward the south and increases toward the north. – 0 Mountains—the amount of precipitation is greater in the higher mountains. Orographic rainfall in the Samaria Hills increases from west to east because of the increasing elevation of the mountains (Gat and Karni 2003). – 000 Wind—wind affects the thickness of the rain and the course of the drops, resulting in a different spatial distribution that affects precipitation in certain areas (Goldreich 1998). – 0000 Distance from the sea—generally speaking, rainfall in the Land of Israel decreases the farther away from the sea (Gat and Karni 2003: 28). This reduction does not begin at the coastline, but only 3–5 km to the east, where the terranean effect is stronger (Goldreich 1998: 76). The Samaria foothills receive 400–600 mm of rainfall annually (November–April). The daily rainfall averages about 12 mm (Ashbel 1968:14).

Water A most prominent feature of the region is the lack of stable water sources. Except for the large streams which flow during the winter, such as Naḥal Rabah and Naḥal Shiloh, no stable water source is present in the entire area. Springs can be found in the sources of Naḥal Yarkon (see below) and in the eastern part of the Samaria mountain, but not in the study area. The population of the area in antiquity utterly depended on rainwater collected in cisterns and reservoirs, either built or hewn (e.g., Site 7: Kh. Umm al-Bureid).

The amount of dew in mountainous regions is limited (Katsenelson 1966). However, it is more common in the Samaria and the Judean Hills because of the colder temperatures and higher elevation, while in the lower western and eastern Samaria Hills, the amount of dew drops. Larger amounts of dew are common during the summer, as wind gusts from the sea. In winter, when easterly winds blow, the amount of dew falls (Gat and Karni 2003).

The annual rainfall in Judea and Samaria is c. 500 mm. The rainfall may infiltrate the soil and supply the groundwater or evaporate, or, in turn, be washed as runoff water in streams. Precipitation east of the study area and as far as Shechem is drained into Naḥal Yarkon (Boneh and Baida 1977: 34).

Winds The synoptic system and the marine breeze cycles determine the wind system in the Judean and Samaria Hills, especially during the summer (Gat and Karni 2003: 37), when most winds come from the west, from the Mediterranean Sea. During the transitional seasons, especially in spring, east winds blow, generated by low barometric pressure. The strongest gusts blow in the afternoon and early evening. Winds in the west Samaria Hills blow from the east and the northwest. In winter, the wind may also blow from the southwest.

Rainfall in the Samaria Hills causes water runoff, an element of great importance in the water supply system in antiquity. Three types of water runoff were discerned: surface, intermediate and indirect (Ben-Zvi, 1990: 103). Surface runoff flows into the streams after rain or slightly later. Intermediate runoff occurs after the rainwater infiltrates the ground, emerging from it after a while. The indirect runoff is water that flows through springs. The karstic nature of the Samaria Hills 9

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods

results in a small amount of surface water runoff—only about 1%–2% of the total rainfall (Boneh and Baida 1977: 36). Data regarding surface runoff quantities in the Samaria and Judea hill country are limited, but scholars estimate it reaches c. 20–25 million cubic meters annually.

likely that deforestation occurred during these periods in favor of agricultural areas (Lev-Yadun 1997: 85–102). In several locations, where the natural woods have not been destroyed, one can still find Common Oaks and Palestinian Terebinth (Liphschitz 1991: 27). Ceratonia siliqua and Pistacia lentiscus are also distributed in the area (Waisel, Polack and Cohen 1978). Poterium and other plants dominate the scrubland that characterized the area. Other species (Fig. 2.5) are common in the northern part of the Samaria foothills.

The Yarkon River The Yarkon River is the only stable water source adjacent to the study area. This and the Jordan River are the largest and most abundant river in the Land of Israel. Naḥal Qana, Naḥal Rabah, Naḥal Ayalon, and Naḥal Shiloh are tributaries of the Yarkon River and provide it with water during the winter (Avitsur 1957:13). However, about 2000 springs in the source of the Yarkon River (west of Tel Aphek) make up the main water supply of the river. The Yarkon drainage basin covers a c. 1750 sq km area, but it extends, from Rosh Ha‘ayin to Tel Aviv, for no more than 30 km (Safra 1983: 36). The Yarkon River is divided into three sections:

The damage caused to the natural forest resulted in soil erosion in many parts of the Samaria Hills. Consequently, many areas turned into an impoverished rocky landscape (Liphschitz 1991; Lev-Yadun 1997: 90). Nevertheless, natural flora remains were traced through dendro-archaeological studies conducted in several sites of the Samaria Hills. These included common oak, Palestinian terebinth and olive (Olea europaea; Liphschitz 1984; 1989; Liphschitz and Waiesel 1980; Liphschitz 1986–1987).2 Other species, such as Almond, Ziziphus, Mytrus and Jerusalem Pine were also identified (Liphschitz 1991: 28).3

– The upper section—from Rosh Ha‘ayin to the mouth of Naḥal Qana. – The central section—from Naḥal Qana to the ‘Seven Watermills.’ – The lower section—the Yarkon River’s estuary to the Mediterranean Sea.

Fauna The development of the Land of Israel occasionally shoved animals aside, reducing their living environment. Sometimes, certain species have evidently become extinct because of human activity. The development of the city of Rosh Ha‘ayin southwards and eastwards, and of other settlements in the Samaria foothills, pushed animals to the east, where the ‘Green Line’ fence closed almost completely the animals’ living space. The environmental memorandum (Gal and Ramon 2012) also includes field observations that might reflect the fauna condition of the Samaria foothills in antiquity.

The depth of the river’s stream varies from two meters in the upper section to three–four meters in the lower section. The stream is 12–16-m wide in the upper section and up to 20-m wide in the lower section (Zohar 1991). Flora The Samaria Hills’ flora is mentioned in the Bible, described in the book of Joshua as a forest region (Joshua 17:14–18). Today, the Samaria Hills are quite bare of natural vegetation, the area having been exploited for agriculture throughout history, until the present day. Vineyards and groves of olives, figs, pomegranate, and others have replaced the natural vegetation. The Samaria foothills being a rocky area with shallow soil, it was less cultivated than the high Samaria mountains to its east. However, here too, natural vegetation has been damaged because of overgrazing and tree-clearing. Trees were usually used for building or as burning materials (e.g., kilns). It is difficult to say when the destruction of the natural vegetation began, but it is likely that each succeeding period contributed to this process.

Mammals Several mammals were identified during field observations, deer being the most prominent. During the excavations project at Rosh Ha‘ayin (see below), numerous deer bones were observed in all excavation areas and east of them (Fig. 2.6). Jackals, porcupine, badger, Herpestes, and hyenas were also observed. Birds Common birds in the Samaria foothills are the warblers, larks, and Pycnonotus, visible in the bushes; the turtle2  The name of the site Deir Balut may point to the vegetation in the area. 3  The examined sites are located outside the study area, but similar flora existed, presumably, in the Samaria foothills.

It is generally accepted that the Roman, and especially the Byzantine period, witnessed a significant increase in the number of settlements. Thus, it is most 10

Amit Shadman: Chapter 1, Environmental Background

Herbaceous Herbaceous Herbaceous Herbaceous Herbaceous Herbaceous Herbaceous Herbaceous

Vineyards Vineyards Vineyards Miscellanus Miscellanus Miscellanus

Forests Forests Forests Forests Forests Forests Forests Forests Forests Forests Forests Forests

Groves Groves

Fig. 2.5. Map of Flora in the study area (Gal and Ramon 2012).

11

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods

dove, present in the forests and bushes; the partridge, present throughout the region and especially in the rocky areas; the bee-eater, nestled in Naḥal Shiloh; and pigeons in the cliffs, quarries and caves. Growing pigeons was important for agriculture in antiquity, and columbaria were built or rock-cut for these (Bodenheimer 1956; Tepper 1986: 170–196; Zissu 1996: 100–106). It is generally accepted that pigeons were consumed for their meat, while their droppings served as fertilizer (Zissu 1996: 100).

The hard-rock formations have often dictated the nature of the constructions. The local inhabitants in antiquity distinguished between the various types of rocks and knew how to exploit them (Roth and Flexer 1977: 12). Changes occurred only with the introduction of advanced technology and cutting tools. East of the study area, chalky rocks (‘Mizi Hilu’) are known for their thin-crystal composition. These rocks were and are still used today for construction, as reflected in the presence of both ancient and modern quarries. The settlers knew how to exploit the land for cultivation. Since most agriculture in the Samaria Hills was dry-farming, it required soils 1.5 m deep for optimal cultivation (Dan 1977: 16). The construction of dams across the streams contributed to the optimal utilization of the land. In the many excavations conducted in the area, it became evident that alluvium had accumulated between the agricultural terraces and dams up to 1.0–1.5 m.

Raptors Among this species, falcons were mainly observed, followed by the Short-toed snake-eagle, nesting on trees; the lesser kestrel and long-legged buzzard, mostly seen in cliffs and quarries; and hawks, present in pine forests. Summary

In light of the absence of springs in the study area, the only stable water sources near it are the adjacent springs of the Yarkon River, west of Rosh Ha‘ayin. Despite the strategic location of the Yarkon River and its stability, the inhabitants of the ancient settlements of the Samaria foothills seem to have relied on rainwater and runoff water, as evidenced by the many cisterns and reservoirs exposed in excavations and surveys. The winter rainfall in the area sufficed for growing grains, vineyards and olive groves—the main crops in the area.

The Samaria foothills area is sometimes perceived ecologically inferior. However, even though the entire area has no available water sources, it should be considered as relatively convenient for agricultural activity. The western part of the area is characterized by a leveled plateau, moderate hills and small pocket valleys with alluvium deposits. These features made the area, in antiquity, apt for crops. Moreover, while the Coastal Plain to the west suffered from marshes, the Samaria foothills and the mountains remained a preferable region.

Fig. 2.6. A gazelle in the Rosh Ha‘ayin excavations project (A. Shadman).

12

CHAPTER 2

HISTORY OF RESEARCH

The area was first visited in the second half of the 19th century by Guérin, who described villages and ruins in it, among them H. Te’ena, Qasr es-Sitt, Migdal Afeq, Kh. Umm al-Hammam and Kh. Dayyar (1984: 90–91). In their Survey of Western Palestine, Conder and Kitchener also detailed some major ruins in the area (1881–1883, Vol. II: 286, 360–361).

al-Hammam, Kh. Kesfa, Kh. Ta‘ammur. He tried to trace the nature of the rural settlement in the area, among other things by drawing plans of the main sites with their agricultural hinterland. These studies are a milestone for the present research. Additional surveys were also conducted in the area prior to development endeavors (Haiman 1998: 45–46; 1999), aiming at locating archaeological sites likely to be damaged due to infrastructure works. ‘Ad (2000) studied settlement distribution in the Samaria foothills during the Byzantine period. Taxel (2005) studied the Byzantine and Early Islamic periods, partially based on the findings of Fischer’s excavations at Kh. Zikhrin, south of Naḥal Shiloh (Fischer 1983, 1984; 1989). Taxel (2010) studied the area as part of his research on the rural settlement of Palaestina Prima, the largest province in Byzantine Palestine. Torge (2007) investigated the area spreading between the sites of ‘Izbet Sarta on the north and Tel Hadid on the south during the Iron Age.

The proximity of this region to the pre-1967 border between Israel and Jordan (the Green Line) has prevented scholars from conducting archaeological studies in it. However, several researchers carried out surveys there in the 1970s and 1980s. Kochavi and Beit Arieh (1994) inspected the area within the framework of the Map of Rosh Ha‘ayin (78) survey. This map included two different geographic regions: one on the east—the Samaria foothills (part of the current study area)—and part of the Coastal Plain, on the west. In the eastern section, Kochavi and Beit Arieh uncovered many archaeological sites, some of them presented in this work. Besides the large sites already discovered in the 19th century, they recorded dozens of field towers, winepresses, rural roads, and more.

An important work for the present research, both archaeologically and historically, was conducted by Zelinger (2009), who studied the rural settlement in the Lod Valley during the Hellenistic and Early Roman periods through the Bar Kokhba Revolt. Many additional surveys and excavations were conducted south of the study area before the construction of the Cross-Israel Highway (Route 6; Dagan 2011a). 1

Following this survey, Finkelstein, who had also taken part in it, carried out extensive studies in the region (1978a; 1981:331–348). He examined, among other things, 25 farmhouses situated between Naḥal Qana on the north and Naḥal Ayalon on the south. Finkelstein argued that two factors supported the establishment of these farms: a shortage of lands prompting the inhabitants to seek additional areas—including inferior ecological ones—and stable security conditions which enabled the construction of the farms (Finkelstein 1978a:70). He also found that most farmhouses had been initially founded in the Iron Age II and continued through the Persian and Hellenistic periods. Fixler (1991; 2001) conducted an interesting and pioneering work, interpreting extensive areas and ruins based on aerial photographs. He studied, among other sites, Kh. Kesfa, H. Yeqavim, and Kh. Ta‘amur.

Archaeological Excavations Until the present study, archaeological excavations had been carried out mainly in the northern and western parts of the area, prior to the construction of new neighborhoods in the town of Rosh Ha‘ayin. Although the site of ‘Izbat Sarta, north of Naḥal Rabah, is not included in the current study area, it should 1

Other studies are also worth noticing: Clermont-Ganneau 1896:340– 341; Rey 1883; Albright 1923; Beyer 1951; Loewenstamm 1950: 501– 503; Press 1951–1955, 3: 527; Prawer 1963, A: 306, 380, 446, 470, 540, 550, B: 77; Lange 1965; Benvenisti 1970; 1984; Vilnay 1977a, b; Hoade 1978; Bagatti 1983; 2002: 155–156; Hartman 1987; Khalidi 1992; Ellenblum 1992; Tsuk 1993: 67–76; Kennedy 1994; Tsafrir, Di Segni and Green 1994: 64; Pringle 1994; 1997: 67; Raviv and Elinson 2014; Sasson 2014.

In his works, Safrai (1995b, 1995c:104–106; 1997a, 1997b, 1998a; Safrai and Fixler 1993) examined several sites in the western Samaria Hills, including Kh. Umm 13

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods

be nevertheless mentioned due to its importance. Three strata were identified in this site, excavated by Kochavi and Finkelstein in 1976–1978 (Finkelstein 1978a). Stratum III, dating to the 13th century and the beginning of the 12th century BCE, consisted mainly of a row of peripheral rooms surrounding a courtyard. Stratum II yielded a four-room house and other buildings with silos surrounding them. This stratum was dated to the 11th century BCE and lasted only for a short period, after which the site was abandoned.

installations and burial caves (Haddad 2007:66, 2011:53– 63; Eliaz 2002: 112; ‘Ad 2007; Elisha 2010; Tendler 2013a, 2013b, 2015; Drezner 2014). Magen carried out several excavations in the eastern part of the area, mainly in sites along the Green Line. The most important excavation was conducted at Kh. Kesfa, where several impressive olive oil presses were exposed (Magen 2008b). The extensive pre-development surveys carried out in Rosh Ha‘ayin yielded hundreds of archaeological elements, all excavated by the author of this book. The preliminarily published results of some of these excavations (Shadman 2014; Shadman, Tendler and Marcus 2015: 84–88; Tendler and Shadman 2015: 193– 207; Haddad et al. 2015: 50–68) will be fully discussed in the present study.

Settlement remains dating from the Iron Age through the early Islamic period were exposed during a large salvage excavation conducted by Avner-Levi and Torge (1997; 2018) in the western part of the study area. The ongoing development works led to additional excavations, revealing, mainly, agricultural

14

CHAPTER 3

THE SITES

Fig. 3.1. The sites.

15

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods

Site No. 1 Qasr as-Sitt Map reference: 197750/666000

and the northern walls (W09, W11) were relatively close to the building, while the eastern wall (W10) made the corner with them to create a large open courtyard at a level lower than the building. A tabun was discerned in a probe near W10. A small rectangular room of an unclear function was excavated north of the courtyard. Northeast of this room, another building was uncovered, built of large-sized fieldstones, as well as two winepresses. A winepress, replacing a smaller one to its east, comprised a treading floor (3.6 × 3.9 m), a settling pit (0.65 × 0.85 m), and a collecting vat (1.06 × 1.50 m). 

An agricultural farm on the eastern slope of a high hill, 1.5 km northeast of Migdal Afeq. The farm comprises a building with a courtyard adjacent on the east (Fig. 3.2.1). Based on the ceramic finds, the farm operated mainly during the Iron Age II, but also through the Persian and Hellenistic periods. The building. A square structure, built in dry construction of medium- and large-sized fieldstones and ashlars, with traces of plaster and preserved to a height of three courses (Fig. 3.2.2). The southern and the western walls (W01, W02) were better preserved than the northern and the eastern ones (W03, W04). W01 and W02 were flanked by two outer walls on the west and south respectively (W05, W06), forming doublestepped walls. Several rooms identified inside the building were not excavated. This building apparently served for residential purposes.  

The Agricultural Plot This farm owned a c. 400-dunam area extending to its south, west and east (Fig. 3.3) comprising gentle ridges cultivated since the area’s earliest occupation. A rocky slope bordering a gully, located north of the farm, makes it unlikely for the settlement to have spread beyond this line. Further north was a large area that apparently belonged to Kh. Taha. On the west, Qasr asSitt was bounded by a deep gorge probably bordering Qurnat al-Haramiya (Site No. 15). East of the settlement was a large area, possibly its main cultivated area, and the site of Wadi al-Bureid (Site No. 2). Agricultural systems and several installations were recorded nearby.

The courtyard. A square, open space, with walls preserved to a height of two courses built in dry construction and founded on bedrock. Its southern

Fig. 3.2.1. Qasr as-Sitt. Aerial view of the farm (Courtsey of H. Cohen-Klonymus and E. Emanuelov).

16

Amit Shadman: Chapter 3, The Sites

Winepresses. Nine simple winepresses were discovered in the area, two of them adjacent to the farm (Fig. 3.3.2). Set atop the bare bedrock, most winepresses included a treading floor and a collecting vat. Some, such as No. 28, had an irregular treading floor and collecting vat, attesting to its possible Iron Age origins. The relatively large number of winepresses shows that wine industry played an important part in the local economy.

center of the upper depression, was probably intended for extracting the oil, and a rock-cut channel, aimed at pumping the liquid into a larger basin, was found between the two depressions. Finkelstein (1978a:10a) recorded an oil press near this farmhouse which has since disappeared. Field towers. A single square field tower (c. 2 × 2 m), with an opening in its northern wall, was found north of the site, between two terraces and overlooking the wadi. The walls of the tower were built on the westwardsloping bedrock of very large-sized stones. The bedrock below the tower’s western wall was leveled with small stones. Presumably, additional field towers existed in the area but did not survive.

Oil presses. Only a very small number of simple installations (bodedah in Mishnaic Hebrew), perhaps associated with the oil industry, were found in direct relation to this farm, all of them comprising hewn depressions (diam. c. 0.50 m; depth c. 0.20 m). Installation No. 905B included an upper (diam. 0.36 m) and a lower depression (diam. 0.47 m). A basin, in the

Fig. 3.2.2. Qasr as-Sitt. Plan of the farm (Courtsey of H. Cohen-Klonymus and E. Emanuelov).

17

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods

Fig. 3.3. Qasr as-Sitt. The agricultural plot (Courtsey of H. Cohen-Klonymus and E. Emanuelov). 

Water cisterns. Seven cisterns were discovered near the site, and several of them were excavated. Cistern No. 8 had an irregular opening and a shaft (1.4 m deep) leading to a pit at least 4-m deep. Three rock-cut steps descended into the 1.5-m deep shaft of Cistern No. 24, and remnants of gray plaster were discovered on its walls. Four cisterns were excavated around Cistern No. 30, and two of these were later converted into limekilns (perhaps in the Ottoman period). Two additional cisterns (Nos. 903–904) were recorded a few hundred meters south of the site, together with an agricultural road built near them.

fieldstone walls. Segments of the cave have collapsed, but it initially extended for a length of at least 2 m.  Summary The site, one in a series of farms, was first established during the Iron Age II and continued uninterruptedly to the Persian and Hellenistic periods. Based on the numerous winepresses present at the site, it seems that wine production was not intended solely for local consumption. The area was not settled in the Byzantine period and was apparently part of the agricultural hinterland of Migdal Afeq. During the Ottoman period, a large limekiln was built nearby.

Stone-clearance heaps. Fourteen heaps (Nos. 11, 13–14, 34, 49, 79, 60, 62, 65, 78, 94, 100–101, 653; 4–8 m diameter) were excavated south of the farm, all seemingly dated later to it. The stone-clearance heaps usually consisted of an outer wall, one to two courses high, built of mediumand large-sized fieldstones. The heaps included small, fist-sized stones with earth in between.

Guérin:  “I saw the remains of a small compound that was reinforced by small towers which are now destroyed, except for one where some of its lower courses are still standing.” (Guérin 1984: 95). PEF:  “This ruin consists of small enclosure with towers, now destroyed, with the exception of one, whose lower courses still remain. It was seen by Guerin close to the Mejdel Yaba. It is perhaps the place marked ‘Tombs’

Caves. The area is characterized by karstic cavities, most of them apparently unexploited by humans. However, Byzantine potsherds discovered in Cave 95 attest to human activity. The entrance to this cave (0.90 m wide) was found between two c. 0.50 m high 18

Amit Shadman: Chapter 3, The Sites

or Sheikh Bazar ed-Din on the map.” (Conder and Kitchener 1882, II: 336).

6.0 m; Fig. 3.5) was probably in its southeastern corner. Depressions in the bedrock floor of the room were filled with earth and topped with a fieldstone bedding to level it. Room II was a broad room (3.5 × 10.0 m), the largest in the northern wing, entered through a doorway in the middle of its southern wall (W7032). A drainage channel (Fig. 3.6), slightly slanting northwards, was exposed in the eastern part of the room. The walls of this channel, unplastered, were built of two rows of stones laid on their long side. The channel passed through a narrow opening in the northern wall (W7007) and continued northwards, probably leading to a water cistern or a drainage basin that has not yet been exposed. Rooms III–V were interconnected through openings in their partition walls and probably represented a single unit.

Ceramic finds: Iron Age II; Persian, Hellenistic, Byzantine and Ottoman periods. Previous studies: Conder and Kitchener 1882, II: 336; Finkelstein 1978a: 10, Pls. 4–5; 1981: 332, 336–337; Guérin 1984: 95; Kochavi and Beit Arieh 1994: Site 102. Excavations: A. Cohen-Klonymus and E. Emanuelov on behalf of Tel Aviv University. Site No. 2 Wadi al-Bureid  Map reference: 198300/665800 

The central wing. This wing (15 × 30 m) was probably used as a residential quarter and hospitality area. It comprised eight rooms arranged in two adjacent east– west oriented rows: Rooms VII, VIII, IX, XI, XII (the southern row), and Rooms X and XIII (the northern row). A single room (VI), adjacent on the northeast to Room X, was constructed in the southeastern corner of the northern courtyard, and could only be accessed from it. The entrance (1.5 m wide) to the central wing was in the northern W7202 of Room X, in the northern row, which faced the courtyard (Fig. 3.7). South of the entrance, within the room, most probably serving as a reception hall (Fig. 3.8), was an entrance area (4 × 17 m; L7094) paved with small- and medium-sized fieldstones, while the bedrock was used as a floor in its western part.

A farmhouse (31 × 65 m; Fig. 3.4.1, 3.4.2) on a spur to the west of a wadi descending toward Wadi Susi. Three parallel ramparts (height 1.5–2.0 m), made of small fieldstones and oriented east–west, were discerned prior to the excavation. A limekiln, probably from the Ottoman period, was recorded on a fourth, E-shaped north–south oriented rampart located west of the three parallel ones. Beneath the stone ramparts, the excavations revealed rows of east–west oriented rooms making up a carefully constructed rectangular complex. The northern row made up the northern wing of the building, while a double row of rooms made up the central wing (Fig. 3.4.1) and the southern row the southern wing. An open courtyard (14 × 17 m) connected the northern and central wings, while a large courtyard (25.5 × 26.5 m) was found between the central and southern wings. For some unclear reason, the central wing was isolated from the southern courtyard, but its dimensions indicate that it was an important part of the complex. On the west, the three wings and the courtyards were bounded by a common wall (W7131). Since no openings were traced in this wall, the courtyards were most probably approached through the poorly preserved eastern wall (W7116) of the complex. Each courtyard separately served its own wing(s), which could be accessed only from it: The northern courtyard served the northern and central wings while the southern courtyard served the southern wing.

Two piers, built of roughly hewn fieldstones, were found in the room’s eastern part: the northern pier is preserved two courses high and the southern one only one course high above the floor level. These piers formed a separate space that served as a passage to Room VII (3 × 3 m), located to its south and isolated from Room VIII to its west on the same row. Equipped with a plastered sitting bath and with remnants of plaster on the southern and western walls, this room certainly served as a bathroom (Fig. 3.9). The bath was filled with potsherds dating from the Persian and Early Hellenistic periods. Following the abandonment of the building, the bath was seemingly turned into a refuse pit for potsherds.

The complex was built of solid walls (1.1–1.3 m thick) with two faces of hewn and semi-hewn stones and a fill of fieldstones and earth. It is quite possible that certain parts of the building had a second floor. 

On the western end of the southern row of rooms, a doorway in W7130 led from Room XIII (Fig. 3.10) to Room XI (3.0 × 5.5 m). A stone lintel was found above the entrance area. A square fieldstone installation, of an unclear function, was exposed in the southeastern corner of Room XI, and a small stone altar with a small depression atop (Fig. 3.11) was located on the eastern side of the room. This room was connected through another opening with Room IX (4 × 5 m) to its east, where a beaten-earth-and-fieldstone floor was revealed.

The northern wing. This wing consisted of five rooms of various sizes (I–V; 30 m total length). The rooms were found filled with large stone debris caused by the collapse of the building after its abandonment, making it very difficult to trace the floors, which were found only in isolated limited points. The entrance to Room I (3.5 × 19

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods

Fig. 3.4.1.. Wadi al-Bureid. Plan.

20

Amit Shadman: Chapter 3, The Sites

Fig. 3.4.2. Wadi al-Bureid. Sections to plan.

Fig. 3.5. Wadi al-Bureid. Room I, looking southeast.

21

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods

Fig. 3.6. Wadi al-Bureid. Room I, the drainage channel. 

Fig. 3.7. Wadi al-Bureid. The entrance from the northern courtyard to Room X, looking south.

22

Amit Shadman: Chapter 3, The Sites

Fig. 3.8. Wadi al-Bureid. Room X, looking west.

Fig. 3.9. Wadi al-Bureid. The sitting bath, looking north.

23

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods

Fig. 3.10. Wadi al-Bureid. The doorway between Rooms X and XIII, looking northwest. Note the lintel.

An entrance to Room XII (2.8 × 3.0 m) was located in the southwestern part of Room XIII, in the northern wall of Room XII (W7082), which was built on bedrock. A hewn oval shaft (c. 0.5 × 0.8 m) was found south of this wall and was excavated to a depth of 0.5 m, its excavation being terminated for safety issues. It was full of medium-sized fieldstones and potsherds dating from the Persian and Hellenistic periods, and it seems to have been a water cistern used by the farm. All the openings in the eastern part of the central wing, including the main entrance from the courtyard to the reception hall, were deliberately blocked. The Southern wing. Three rooms, oriented east– west, were exposed in this wing. Rooms XIV and XV were entered from the southern courtyard. Passage between the rooms was possible through an opening in the partition wall (W7236). No openings were exposed in Room XVI (2.0 × 3.3 m), and an inner partition wall divided it into two small cells (Fig. 3.12). The floor of the southern cell was partly bedrock and partly paved with small-sized fieldstones.

Fig. 3.11. Wadi al-Bureid. A stone altar discovered in Room XI.

24

Amit Shadman: Chapter 3, The Sites

Fig. 3.12. Wadi al-Bureid. Aerial view after the excavation (Sky View, A. Shadman).

25

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods

The Agricultural Plot

Caves. Four caves were discovered in the region, two of them natural and two man-made. The natural caves were discovered after descending some bedrock steps (depth 1.0–1.5 m). The excavation of the caves yielded mostly a natural alluvium fill, small pebbles, and worn potsherds, but in one cave, the remains of a hearth were found, possibly contemporary. One of the manmade caves was eight meters wide, but most of it had collapsed.

This site had a fairly convenient agricultural area, with a level plateau with alluvial soil (max. depth 0.5 m) to its south and west. Several rock-cut winepresses were excavated on the rocky spur northwest of the site (see below), and a built road probably dated to a later period was recorded to its east. Winepresses. Ten winepresses were found, most of them located on a rocky spur northwest of the site. The high number of presses in relation to the size of the site (two dunams) may indicate the possible use of some of these winepresses also by the inhabitants of the site of Qasr as-Sitt. This is the densest concentration of winepresses discovered in the entire study area, possibly due to the extensive areas extending south and north of the winepresses’ area probably used for vineyards and grain fields. Most of the winepresses were of the simple type, with irregular treading floors and collecting vats, with no connection channels between them. A rock-hewn cupmark, apparently intended for collecting waste, was observed in the treading floor of several winepresses. Also discovered were a few more complex winepresses that included a square treading floor, a settling pit, and a well-cut collecting vat. These presses had hewn channels and showed remnants of plaster on their walls.

Stone-clearance heaps. Four heaps were identified and excavated in this area, two with a circular outline and the other two between two walls that formed a corner. A hewn cell of an unclear nature was found in one of the heaps (Fig. 3.13). The Pottery — Hagit Torge Bowls (Fig. 3.14). The vessels in Fig. 3.14:1–2 are mortaria with an everted curved rim. Fig. 3.14:3–4 are heavy bowls characterized by a very thick wall and mostly used for grinding. The bowl in Fig. 3.14:3 has a fairly straight everted shelf rim; the bowl in Fig. 3.14:4 has a rounded rim and a straight wall. These vessels, dated to the 5th–2nd centuries BCE, were very common throughout the country. Petrographic examination of the Mortaria from Tel Michal showed that they originated in the Aegean region (Gorzalczany 2006: 58). Seven imported bowls were discovered in the excavation, most of them Attic bowls, their bodies either wholly or partially slipped with a glossy black slip: The bowls in Fig. 3.15:5, 6, 8–11, dated to the 5th–4th centuries BCE, had a

Water cisterns. Five rock-cut cisterns were discovered in this area, although the function of two of them is unclear. A hewn channel drained the water into the opening of one of the cisterns, which was closed with a lid (possibly contemporary). 

Fig. 3.13. Wadi al-Bureid. A stone clearance heap containing a built-in hewn cell.

26

Amit Shadman: Chapter 3, The Sites

hemispheric body and ring base, and the bowl in Fig. 3.14:7 is a ‘fish bowl’ dating from the end of the 6th century BCE. Imitations of this vessel continued to appear in the Hellenistic period until the 2nd century BCE. Bowl 3.14:6, of the ‘Eastern Greek’ type, with an inverted rim and a rather low body, dates from the end of the 6th to the 4th century BCE. These bowls were produced throughout the Aegean region.

Fig. 3.14:18 depicts a bowl with an everted thin rim, and Fig. 3.14:19 represents a hemispheric bowl. Although these two bowls are not slipped, they are also considered as local imitations of the Attic bowls, likewise dating from the 5th to the 3rd centuries BCE. The bowls in Fig. 3.14:20, 21 are characterized by an irregular incised decoration inside the vessel and date from the Persian period.

The bowls in Fig. 3.14:12–17 are hemispheric and slipped either completely or on their interior face. These vessels are a local imitation of the Attic bowls and were common in the 5th–3rd centuries BCE.

A cooking krater and cooking pots (Fig. 3.15). The cooking krater (Fig. 3.15:1) is characterized by an everted shelf rim, usually with horizontal handles, although vertical handles can sometimes be present

2 1

4

3

7

6 5

8

9

10

11

12 13

14

16

15

17

18

19

20

21

0

10

Fig. 3.14. Wadi al Bureid. Mortaria (1–4) and bowls (5–21).

27

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods

◄ Fig. 3.14 No.

Locus

Basket

Description

Parallels

1.

7094

70177/1 Collapse in the central room.

Light brown clay, many white and black inclusions of various size.

Kapitaikin 2006: Fig. 3: 12.

2.

7186

70353/2 Floor beneath debris.

Light greenish clay, many white and black inclusions.

Kapitaikin 2006: Fig. 3:13.

3.

7205

70360/3 A room east of W7177.

Light greenish clay, many white and black inclusions.

Kapitaikin 2006: Fig. 2:10.

4.

7053

70096/3 collapse

Pinkish clay, many white and black inclusions of various size.

Tal 1999: Fig. 4.21:17.

5.

7014

79016/5 Surface.

Well levigated pinkish clay, few white inclusions.

Kapitaikin 2006: Fig. 13:3.

6.

7259

70486/2 Bathroom.

Well levigated pinkish clay, few white inclusions.

Kapitaikin 2006: Fig. 12:2.

7.

7053

70096/4 Clearing debris.

Well levigated pinkish clay, few white inclusions.

Kapitaikin 2006: Fig. 1:14.

8.

7052

70196 Foundation trench.

Greyish clay, few black and white inclusions.

Kapitaikin 2006: Fig. 13:2–5.

9.

7187

70356/1 Collapse in the room.

Well levigated pinkish clay, few white inclusions.

Kapitaikin 2006: Fig. 13:3.

10.

7098

70343/2 Surface.

Well levigated grayish clay.

Fischer and Tal 1999: Fig. 5.7:7.

11.

7199

70340/2 Collapse in the room.

Well levigated pink clay, well fired.

Segal Kletter and Ziffer 2006: Fig. 12:12

12.

7187

70350/1 Collapse in the room above floor.

Pinkish clay, white and brown inclusions.

Kapitaikin 2006: Fig. 1:13.

13.

7259

70508/4 Bathroom.

Pinkish clay, few white inclusions.

Kapitaikin 2006: Fig. 13:1.

14.

7205

70383/1 A room east of W7117.

Pinkish clay, few white and brown inclusions.

Kapitaikin 2006: Fig .1:13.

15.

7263

70516/2 Bathroom.

Pinkish clay, few white and brown inclusions.

Kapitaikin 2006: Fig. 13:1.

16.

7205

70383/2 A room east of W7117.

Pinkish clay, white and brown inclusions.

Kapitaikin 2006: Fig. 1:13.

17.

7259

70490/2 Bathroom.

Pinkish clay, few white and brown inclusions.

Kapitaikin 2006: Fig. 13:1.

18.

7052

70106 Foundation trench.

Orang clay, black and white inclusions.

Kapitaikin 2006: Fig. 1:10.

19.

7186

70315/1 Floor beneath collapse.

Orang clay, black and white inclusions.

Fischer and Tal 1999: Fig. 5.12:15.

20.

7032

7232

Light brown clay, white inclusions of various size.

Baruch 2006: Fig. 12: 12.

21.

7205

7205 Room east of W7117.

Light brown clay with dark core, white inclusions of various size.

Baruch 2006: Fig. 12: 11.

(Stern 1982: Type 2-3). These vessels are characteristic of the Persian and Hellenistic periods. Among the cooking pots, the first (Fig. 3.15:2), dated to the Persian period, has a rather short neck, a wide mouth, and a

sharp rim. The cooking pot in Fig. 3.15:3, dated to the 3rd–2nd centuries BCE, has an everted thin rim, a high neck and a sharp incline between the neck and the body. 28

Amit Shadman: Chapter 3, The Sites

1

2

3

4

5 6

8

7 0

10

Fig. 3.15. Wadi al-Bureid. A krater (1) and cooking pots (2–8). No.

Locus

Basket

Description

Parallels

1.

7187

70316/5 Collapse above floor.

Pinkish-orange clay, few large white inclusions.

Segal, Kletter and Ziffer 2006: Fig. 8:1.

2.

7186

70353/1 Floor beneath collapse.

Orange clay, few white inclusions.

Stern 1995: Fig. 2.4:6.

3.

7228

70489 Collapse north of W7212.

Brown clay, dark core, many black and white inclusions of various sizes.

Smithline 2013: Fig 8:4.

4.

7259

70508/2 Bathroom.

Dark brown clay, many black and few white Baruch 2006: Fig. 7:4. inclusions.

5.

7205

70370/2 A room east of W7117.

Brown clay, dark core, many small black and few large white inclusions.

Segal, Kletter and Ziffer 2006: Fig. 8:9.

6.

7205

70370/1 A room east of W7117.

Brown dark clay, many black and few white inclusions.

Baruch 2006: Fig. 7:6.

7.

7205

70370/4 A room east of W7117.

Brown dark clay, many black inclusions.

Baruch 2006: Fig. 7:6.

8.

7215

70382/1 courtyard.

Brown clay, many black and white inclusions of various size.

Berlin 2012: Fig 2.1: 5

Fig. 3.15:4 shows a well-fired cooking pot with thin walls, a convex neck, and an upright rim, dated to the Persian period. Fig. 3.15:5 represents a cooking pot with an everted rim, a pair of handles from the rim to the shoulder, and a wide and upright neck. It is common in assemblages dating from the Persian period (Stern 1982: Type C). This type is broadly dated to the 5th–4th centuries BCE, although it seems to have appeared already at the end of the 6th century BCE (Gitin 1990: Pl. 31:12, Type 154a).

were found at Gezer (Gittin 1990: Fig. 37:15). In cooking pot No. 8, of a type usually dated to the Hellenistic period (Berlin 2012: 10), the handles protrude from the rim. Jars (Figs. 3.16–3.17). Storage jars made up approximately 70% of the pottery vessels recovered from this excavation. Since most of them belonged to the same types, only a sample is presented here, based on their stratigraphic value. Fig. 3.16:1 is a holemouth jar with a thickened, short, rather flat and outward folded rim. These jars date from the Persian period.

The cooking pots in Fig. 3.15:6, 7, of a similar type and well-fired, have a slightly everted upright rim. Similar cooking pots dated to the 3rd–2nd centuries BCE

The jars in Fig. 3.16:2–7, with a wide cylindrical neck with an everted rounded rim and a prominent 29

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods

2

1

3

5

4

10

9

8

12

16

7

6

13

11

15

14

17 19

18

20

22

21

23 0

10

Figs. 3.16. Wadi al-Bureid. Holemouth (1) and Jars (2–23).

depression below it, are common in the excavation’s assemblage. These vessels were widespread throughout the country in the 5th–4th centuries BCE, and continued, with minor changes (such as a shorter neck, Fig. 3.16:8, 9), into the Hellenistic period. Another type is the ‘torpedo’ jar, divided into four subtypes:

neckless or have a very short neck, and are dated to the Persian period, from the end of the 6th century and the beginning of the 5th century BCE to the 2nd century BCE. At Tel Michal (Singer-Avitz 1989: Fig. 9.12: 9) and Tel Hesi (Bennett and Blakely 1989: Fig 13: 13), this type of jars was dated to the Persian period, and, in Samaria, from the beginning of the 5th century to the 2nd century BCE (Kenyon 1957b: 233, Fig 42: 11).

[1] Jars with a long and almost straight rim (Fig. 3.16:16), most common in assemblages dating from the Persian period (Stern 1982: Type G). Parallels were found at Dor in Str. 6B–6A, dated to 538–400 BCE (Stern 1995: Fig 2.9: 1, 2).

Bag-shaped jars comprised three different subtypes: [1] Jars with an everted thickened rim with a square profile, a depression below it, and a cylindrical neck extending towards the body (Fig. 3.16: 10–13), a type dated to the Persian period. The jars in Fig. 3.16:14, 15, also dated to the Persian period, are similar in shape but have a thinner rim with a less angular, almost triangular profile, a marked depression below it, and a cylindrical neck.

[2] Jars with a shorter and more inclined rim (Fig. 3.16: 17, 18), widespread throughout the country and dated, in Jerusalem, from the Persian period (Ben-Arieh 2000: Fig 15: 3, 4) to the beginning of the 2nd century BCE (Lapp 1981: Pl. 73: 21). [3] Jars with a triangle thickened rim with a slightly raised angle (Fig. 3.16: 19–23), dated to the 6th–4th centuries BCE (Stern 1982: 105–106). [4] Jars with a folded rounded rim, a hollow depression below it and a cylindrical body extending toward the center of the vessel (Fig. 3.17: 1–8). These vessels are either

[2] Jars with a rounded everted rim and a short or medium cylindrical neck (Fig. 3.17: 9–15). These are common mainly in the Coastal Plain and are defined in Dor as Type 1F (Stern 1982: 104–105), dated to the end of the 4th century BCE. The jar in Fig. 3.17:16, dated to the 30

Amit Shadman: Chapter 3, The Sites

◄ Fig. 3.16 No.

Locus

Basket

Description

Parallels

1.

7091

70292 Collapse in the northern room, tower.

Brown-orange clay, gray core, many white inclusions of various size.

Baruch 2006: Fig. 14:1.

2.

7204

70357 Shaft in the northern room.

Buff clay, black and white inclusions.

Berlin 2012: Fig. 2.2:3.

3.

7202

70446/1 Wall cleaning.

Pinkish-orange clay, white inclusions of various size.

Segal, Kletter and Ziffer 2006: 6:10.

4.

7202

70446 Wall cleaning.

Light buff clay, black and white inclusions.

Tal 1999: Fig. 4.13:29.

5.

7259

70455/3 Bathroom.

Orange clay, white inclusions of various size.

Tal 1999: Fig. 4.13:29.

6.

7052

70106/3 Foundation trench.

Brown-orange clay, gray core, many whitr inclusions of various size.

Baruch 2006: Fig. 6:8, 9.

7.

7215

70382 The northwestern corner of the southern courtyard.

Dark gray clay, white inclusions of various size.

Tal 1999: Fig. 4.36:9

8.

7014

79016/3 Surface.

Light buff clay, black and white inclusions.

Fischer and Tal 1999: Fig. 5.15:8.

9.

7187

70316/4 Collapse on floor L7211.

Pinkish clay, white inclusions.

Fischer and Tal 1999: Fig. 5.15:10.

10.

7215

70382/3 corner of the northern courtyard.

Grayish clay, few white and black inclusions of various size.

Alexandre 2006: Fig. 63:4.

11.

7259

70440/4 Bathroom.

Light greenish clay, many black and white inclusions.

Torge 2012a: Fig. 1:9.

12.

7007

70267/1 Wall cleaning.

Light greenish clay, many black and white inclusions.

Segal, Kletter and Ziffer 2006: Fig. 9:17.

13.

7193

70344/5 North 0f W7195.

Pink-orange clay, few large white inclusions.

Segal, Kletter and Ziffer 2006: Fig. 9:17.

14.

7259

70496/2 Bathroom.

Pinkish clay, white and black inclusions.

Tal 1999: Fig. 4.36:9.

15.

7187

70316/2 Collapse above floor L7211.

Pinkish clay, large white inclusions.

Tal 1999: Fig. 4.36:9.

16.

7194

70337 west of W7106.

Light buff clay, many small black and white inclusions.

Tal 1999: Fig. 4.22:11.

17.

7203

70352 Southern cell, fill under L7199.

Yellowish clay, many small black and white inclusions.

Segal, Kletter and Ziffer 2006: Fig. 9:17.

18.

7052

70096 Foundation trench.

Greenish clay, many small black and white inclusions.

Berlin 2012: Fig. 2.3: 10.

19.

7224

70539 Collapse between W7105/7028.

Pinkish-orange clay, many white inclusions.

Segal, Kletter and Ziffer 2006: Fig. 9:20.

20.

7211

70371 Dismantle living floor.

Pinkish clay, many small black and white inclusions.

Alexandre 2006: Fig. 53:16.

21.

7259

70490/1 Bathroom.

Pinkish clay, many large black, brown and white inclusions.

Alexandre 2006: Fig. 53:16.

22.

7203

70352/3 Southern cell, below L7199.

Pinkish clay, many small black and white inclusions.

Alexandre 2006: Fig. 60:14; Segal, Kletter and Ziffer 2006: Fig. 9:17.

23.

7052

70229 Foundation trench.

Pinkish clay, many very large black, brown and white inclusions.

Alexandre 2006: Fig. 53:13; Segal, Kletter and Ziffer 2006: Fig. 9:20.

31

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods

2

1

3

6

5

4

7

8 10

9

13

12

11

14

18

16

15

17

19

20

21

22

24 23 0

10

Figs. 3.17. Wadi al-Bureid. Jars.

1995: Fig 2.29: 8). A jar with a square rim, a narrow mouth, and prominent neck, dated to the second half of the 4th century BCE, is depicted in Fig. 3.17:24. However, there are also jars of a similar type with a shorter neck.  

5th–3rd centuries BCE, has a slightly everted thickened rim (Dor — Guz-Zilberstein 1995: 309; Tel Michal — Singer-Avitz 1989: Fig 9.10: 9). The jar in Fig. 3.17:17 has a rounded, rather thin rim, and a relatively long neck. Similar vessels have been dated to the Persian period. Jars with a thick, everted rounded rim, with two depressions on the lower part of the rim and a rather low neck (Fig. 3.17:18, 19) are common throughout the country and are dated to the 6th–5th centuries BCE (Stern 1982: Type F1-4).

Skyphoi/goblets (Fig. 3.18:1, 4). The skyphos in Fig. 3.16:1, with two wishbone handles, is well-fired and red-slipped. A black-slipped skyphos, similar to Attic vessels, is depicted in Fig. 3.18:4. These skyphoi are dated from the Hellenistic period to the middle of the 2nd century BCE. 

[3] Jars with a high, concave neck and a grooved rim for a lid (Fig. 3.17: 20–23), dated to the 6th–5th centuries BCE. Parallels have been found in Tel Hesi (Bennett and Blakely 1989: Fig 143: 14) and Dor (Stern

Jugs (Fig. 3.18:2, 3, 5, 6). Bases of imported black glossy slipped Attic vessels—jugs, juglets or goblets—were

32

Amit Shadman: Chapter 3, The Sites

◄ Fig. 3.17. No.

Locus

Basket

Description

Parallels

1.

7263

70528, bathroom

Yellowish clay, many black and white small inclusions.

Torge 2012a: Gig. 3:27

2.

7228

70478/5 debris north of W7212

Orange-brown clay, many black and white Alexandre 2006: Fig 50:8 l inclusions of various size.

3.

7205

70559, a room east of W7117

Greenish light clay, many black and white inclusions.

Alexandre 2006: Fig 60:11

4.

7079

70559, wall cleaning

Buff clay, black and white inclusions.

Kapitaikin 2006: Fig 6:4

5.

7052

70106/2, hewn foundation trench

Pinkish clay, many small black and white inclusions.

Baruch 2006: Fig. 6:9

6.

7186

70315/2, a floor beneath debris

Greenish light clay, many black and white inclusions.

Alexandre 2006: Fig 50:8

7.

7278

70555/2, Exposing the floor

Greenish light clay, many black and white inclusions.

Alexandre 2006: Fig 60:18

8.

7193

70344/3, north of W7195

Greenish light clay, many black and white inclusions.

Alexandre 2006: Fig 60:5

9.

7186

70336, a floor beneath debris Pink-orange clay, few large white inclusions.

Segal, Kletter and Ziffer 2006: Fig. 10:3

10.

7259

70486/4, bathroom

Light buff clay, black and white inclusions.

Alexandre 2006: Fig 53:17

11.

7091

70231/2, a collapse in the northern room

Light buff clay, black and white inclusions.

Alexandre 2006: Fig 60:7

12.

7193

70344/4, north of W7195

Light buff clay, black and white inclusions.

Alexandre 2006: Fig 60:7

13.

7128

70265/1, eastern wall enclosing large courtyard

Pink-orange clay, many very large white, black and brown inclusions.

Alexandre 2006: Fig 60:5

14.

7203

70352/2

Brown-orange clay, many white and black inclusions of various size.

Alexandre 2006: Fig 53:16

15.

7205

70370/5, a room east of W7117

Pinkish clay, many very large white, black and brown inclusions.

Alexandre 2006: Fig 60:7

16.

7259

70508/3, bathroom

Orange-brown clay, many black and white Alexandre 2006: Fig 48:8 gritz, of various size.

17.

7259

70488, bathroom

Light brown well fired clay, few white and Baruch 2006: Fig. 14:11 brown inclusions.

18.

7128

70248/1, eastern wall of a courtyard

Light buff clay, many black and white inclusions.

Alexandre 2006: Fig 53:12

19.

7186

70353/2, a floor beneath debris

Light greenish clay, many white inclusions.

Segal, Kletter and Ziffer 2006: Fig. 10:4

20.

7107

70206, drainage channel

Light greenish clay, many white inclusions.

Segal, Kletter and Ziffer 2006: Fig. 10:13

21.

7205

70397/1, a room east of W7117

Greenish clay, many white and black inclusions.

Segal, Kletter and Ziffer 2006: Fig. 10:13

22.

7205

70383/3, a room east of W7117

Buff-greenish clay, many white inclusions.

Segal, Kletter and Ziffer 2006: Fig. 10:14

23.

7205

70360/5, a room east of W7117

Light greenish clay, many white and black inclusions.

Segal, Kletter and Ziffer 2006: Fig. 10:13

24.

7259

70495/1, bathroom

Pinkish clay, very large white inclusions.

Kapitaikin 2006: Fig 6:11; Segal, Kletter and Ziffer 2006: Fig. 9:19

33

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods

1 2

4

3

8 7

6

5

9

10

12

11 0

2

2

0

0

0

13 2

10

Fig. 3.18.1. Wadi al-Bureid. Various vessels.

found in the excavations. The first (Fig. 3.18:2) has a high, molded, trumpet base and is decorated on its outer side with designed ridges; the second base (Fig. 3.18:3) has a wide and low trumpet base. The jugs in Fig. 3.18:5, 6 have a very prominent ridged body. The first has a very wide mouth and a gently everted rim. The second is smaller, with a straight rim. Most jugs of this type have a narrower body. These vessels are dated to the Persian period.

Fig 2.11:2, 3), where they are dated to the 5th century BCE.

Juglets (Fig. 3.18:7–9). The juglet in Fig. 3.18:7 has a rather thickened, folded down rim with a sharp depression at the bottom of the fold. The neck is narrow, and the handles are attached to the rim. These vessels are common throughout the country and are dated to the 5th–4th centuries BCE. A juglet with a narrow rim with prominent ridges is depicted in Fig. 3.18:8. At Tel Mevorakh (Stern 1978: Fig. 9:14) and En Gedi (Mazar, Dothan and Dunayevsky 1966: Fig. 33: 4), similar vessels were identified as perfume juglets dated to the 5th–4th centuries BCE. The juglet in Fig. 3.18:9, of the flat-base type, is similar to juglets found in Dor (Stern 1995: 65,

Discussion. Most of the pottery found at the site dated from the Persian period (5th–4th centuries BCE), some vessels dating from the 6th century BCE. Only a few sherds were dated to the Early Hellenistic period (3rd century–last third of the 2nd century BCE). The potsherds collected in the walls and the foundation trench of the building were mostly dated to the Persian period, some also continuing into the Hellenistic period. Thus, the building was founded during the Persian period and functioned at least until the 3rd century BCE. Pottery from the Iron Age II, the Persian and the Hellenistic periods was found here in previous studies.

Oil lamps (Fig. 3.18:10–13). The lamps have a typically elongated mouth. An Attic import with a shiny black slip is depicted in Fig. 3.18:10. The lamp in Fig. 3.18:11 is not slipped, and could be a local imitation of black-slipped lamps. A red-slipped lamp is depicted in Fig. 3.18:12. These lamps are dated to the 5th–2nd centuries BCE.

34

Amit Shadman: Chapter 3, The Sites

◄ Fig.3.18 No.

Vessel

Locus

Basket

Description

Parallels

1.

Skyphos

7259

70508/3, bathroom

Pinkish well levigated clay.

Yievin and Edelstein 1970: drawing 9:23

2.

Goblet

7193

70344, collapse above wall

Pinkish well levigated clay, very few white inclusions.

Kapitaikin 2006: Fig 13:32

3.

Bowl

7198

70547/2, collapse north of W7105

Pinkish well levigated clay, very few white inclusions.

Kapitaikin 2006: Fig 12:30

4.

Skyphos

7128

70265, the eastern wall of a large courtyard

Well levigated grayish clay.

Yievin and Edelstein 1970: drawing 9:23

5.

Jug

7263

70516/3, bathroom

brown clay, few black and white inclusions.

Baruch 2006: 16:1

6.

Jug

7259

70490/3, bathroom

brown clay, few black and white inclusions.

Baruch 2006: 16:1

7.

Juglet

7228

70450/1, collapse north of W7212

Orange clay, black and white inclusions.

Tal 1999: Fig 4.38:6

8.

Juglet

7187

70316, collapse on a floor

Pinkish-orange clay, many white inclusions.

Segal, Kletter and Ziffer 2006: Fig. 11:13

9.

Juglet

7091

70231, collapse in northern room

Brown-orange clay, many black and white inclusions of various size.

Smithline 2009: Fig 4.5:11

10.

Lamp

7000

70000/2, surface

Orange clay, white inclusions of various size.

Rosenthal-Heginbottom and Sivan 1978: 10 Type 11

11.

Attic lamp 7014

70016/10, surface

Orange clay, white, black and brown inclusions of various size.

Segal, Kletter and Ziffer 2006: Fig. 12:18

12.

Attic lamp 7186

70336/3, collapse on a floor

Well levigated well and fired pink clay.

Segal, Kletter and Ziffer 2006: Fig. 12:18

13.

Summary

carefully blocked, with no sign of fire or destruction. Thus, the abandonment of the building was done in an orderly fashion, and the owners apparently intended to return to it.

The building is clearly divided into two equal parts, with the southern wall of the central wing separating between the northern and the southern wings. These wings were probably used as storerooms, while the central wing served as a residential area, which included the stone-paved reception room (Room X) and the built sitting bath (Room VII). The building probably functioned as an agricultural–administrative center, perhaps owned by a wealthy landlord. It may have also served as an administrative center controlling several agricultural farms in the area. The sitting bath is not common in buildings from the Persian period in our region and may be a later addition made in the Hellenistic period. The bathroom attests to the high living standard of its residents (see also Torge 2012b: 21, Fig. 19). The site was well preserved but yielded only a few objects. In addition, most of the openings were

 The pottery finds included mostly jar fragments from the Persian and Hellenistic periods. No intact pottery and glass vessels were found, neither any animal bones. No water cistern was found within the building, but one cistern was discovered to its west. The numismatic finds, especially from the time of Ptolemy I, indicate that the building was abandoned at the beginning of the 3rd century BCE when it probably collapsed. The single Roman coin is probably only a coincidental find. The numerous installations near the site—the winepresses, field towers, and the enclosed plots— cannot be assigned to either the Persian or Hellenistic periods, and it is possible that these were used in later times. 35

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods

Numismatic finds: Ptolemy I, Alexandria, 294–283 BCE (IAA No. G70391, L7198, IAA 143563). Obv.  Laureate head of Zeus r. Rev. ΒΑΣΙΛΕΩΣ–[ΠΤΟΛ]ΕΜA[ΙΟΥ] Eagle with spread wings standing l. on thunderbolt; in l. field, A above I. Æ, , 18.76 gm, 29 mm. SNG Ptol.: Pl. Pl. III:96; Svoronos 1904: 46, No. 275; Cf. Lorber forthcoming, No. B91.

Site No. 3 H. Te’ena Kh. Umm el-Thine; Kh. Umm et-Tina Map reference: 198600/665100 A Byzantine monastery located on top of a gentle hill, c. 2.5 km east of Migdal Afeq. Excavations revealed a complex comprising a church, living quarters, an underground cavity, an oil press, a stable, and water cisterns (Figs. 3.19, 20). The entrance, on the west side of the church, led to an atrium (L10003; 12 × 15 m) with a partially preserved white mosaic with red dots. A plaza (L9998; 5.5 × 22.0 m), south of the atrium, led to the residential quarters, the oil press and the stable. 

Agrippa I (42–41 BCE), a bronze coin minted in Jerusalem (IAA No. 143562). Ottoman coin minted in Damascus (1618–1622; IAA No. 143564).

0

The edge of a mosaic frame, consisting of two rows of tesserae (white, gray, red and black) flanking a white stripe, was preserved at the southern end of the atrium. Based on the clear margins of the mosaic and a stone threshold, the atrium and the plaza were separated, perhaps, by a wooden partition that did not survive. It is possible that the earlier partition was a stone wall. Three stone bases and a crushed, fourth one were exposed in the southern part of the atrium, probably serving to hold the wooden posts that supported a rectangular shed (c. 4.30 × 8.00 m) within the open atrium.

1

Fig. 3.18.2. Coin from Hellenistic period (Ptolemy I)

Previous studies: Finkelstein 1978a: 11–12, Pl. 16; 1981: 332, 338–339, drawing 6; Kochavi and Beit Arieh 1994: Site 145; Shadman 2014; Tendler and Shadman 2015: 193–207; Hadad et al. 2015: 50–68.

Fig. 3.19. H. Te’ena. Aerial view (Sky View, A. Shadman).

36

Amit Shadman: Chapter 3, The Sites

A high bedrock outcrop with a rock-cut cistern was located in the northwestern corner of the atrium, near its western wall’s end (W10018). The atrium led eastward, to the narthex (L9984; 2.65 × 9.00 m); the doorways in the western walls of the atrium (W10018), the narthex (W9984) and the nave of the church were all aligned on a straight axis. The narthex was paved with a partly preserved colorful mosaic, where the southwestern corner of a rectangular frame and flowers or crosses made of red and black tesserae, were observed. A collapsed deep underground cavity (L10016), at the northern end of the narthex, was possibly a water cistern. This cavity was accessed

through a corridor covered with large-sized stone slabs (L10006) and located to the northwest and adjacent to the northern wall of the atrium (W10017). The nave was entered from the narthex through a wide doorway (0.55 × 2.00 m), whose monumental threshold was preserved. The door sockets were clearly visible on the threshold, indicating the presence of two doors opening into the hall. The southern aisle (L9990) was entered from the narthex through a separate door, whose threshold (0.46 × 0.95 m), engraved with a cross, was preserved. Here, too, two doors opened inward. This is likely a separate entrance that led into the northern isle (L10001), but which was not preserved. 

Fig. 3.20.1. H. Te’ena. Plan.

37

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods

Fig. 3.20.2. H. Te’ena. Inscription (A. Peretz).

The church (12.50 × 18.85 m) was a basilical structure, with two rows of columns dividing it into a nave and two aisles. All the architectural elements and the building stones had been robbed in the past, probably during the Ottoman period, when a limekiln operated east of the church. This was a common phenomenon in most ancient sites of the area.

“Under Theodosius the priest this place was built. Lord, peace, peace may be your coming in and your going out, amen.”1 At the eastern end of the nave were the stage (L9982) and the apse. The stage was elevated with a fill made of quarrying debris, and the shape of the apse was discerned from the negatives of the stones. This is an inner apse, built within the general frame of the church, and approached through a narrow passage (width 0.70 m). A row of columns, of which a small section of its stylobate (W9986) was preserved, separated the nave from the southern aisle (3.00 × 12.60 m). The mosaic floor of the southern aisle was better preserved, and its designs, simpler than those in the nave, included squares connected by intersecting diagonal lines. The northern aisle (3.00 × 12.60 m) was more severely damaged than the rest of the church. Only small sections of the mosaic were preserved, making them difficult to identify, but they were seemingly similar or identical to those in the southern aisle. The northern stylobate (W10026) was at a lower level than the mosaic floor and had an exterior face of large roughly hewn stones and an inner face of small stones (a typical technique of the period). It may also belong to an early phase within the church.

 The nave (L9989; 4.60 × 12.60 m) was paved with a mosaic with colorful geometric designs. A Greek inscription and four geometric patterns were found at the entrance to the nave, near the stone threshold (Fig. 3.20/1). The inscription, found in a typical tabula ansata (0.80 × 1.44 m) and consisting of five rows with letters averaging 7 cm in height, read: EPIQEODOÇIOU . . . . BUTE ROUEK. . ÇQHOT≥O≥ - - U—≥EI R≥H≥NHEIRHN. . . .ÇO - - ca. 5 l. - - UKEHEX≥ . . OÇ - - ca. 6 l. - - OITOAMHN ˇ L. 2: remains of horizontal line on ypsilon? ΔEpi; Qeodosivou ªpresbutev-º rou ejkªtivºsqh oJ tovªpo”. º eijrhvnh eijrhvnªh hJ eijªso-º ªdov" soºu ke; hJ e[xªodºov" ªsou gevnºoito, ajmhvn..

1 

38

Read by L. De Segni.

Amit Shadman: Chapter 3, The Sites

The northern ashlar wall of the church (W10007) was well preserved and could be seen from a distance. Some 10 m north of the eastern end of this wall was a hewn staircase that led to a cistern.

into a poorly preserved room (L10011), which had an opening (width 1.18 m) leading northward to a stonepaved hallway (L10000), and then eastward into an elongated hall (L9995; 3.20 × 9.30 m) adjacent to the church on the south. In the southern doorpost of the entrance to this hall (width 0.96 m) was a bar socket, revealing the locking mechanism. The entire hall was paved with white mosaic with a circular groove in the center of its eastern part. The well-laid margins of the groove indicate that this may be a remnant of a stone stand perhaps displaying a valuable item in the church. A blocked and plastered opening, clearly visible in the northern wall of the hall, led to the back room of the church (L9985). This blockage may point to a change in the function of the side hall. Elongated halls adjacent to one of the church’s aisles are known in Byzantine churches in the Land of Israel, and it seems that they served in ceremonies held there. A wide opening with a threshold (length 2 m) was exposed in the southern wall (W10005) of the plaza (L9998), leading southward into an unexcavated room. East of this room was another room (L10027; 2.60 × 5.10 m), paved with white mosaic, which could be entered through doors on each of its walls.

The walls of the church, like the other walls of the monastery, were built of an exterior face of large ashlars and an inner face of small fieldstones on which white plaster was preserved. A fragment of a limestone impost with a cross relief was found within the collapse of the southern aisle. The column base of a green marble screen, bearing the Greek letter π, was uncovered in the stage area. Many ceramic roof tiles were found on the floor of the church and in the other rooms of the monastery, showing the roofing technique. The plaza (L9998; 5.5 × 22.0 m), south of the atrium, was paved with a white mosaic. Two entrances leading into the southern rooms of the monastery were located in its southern wall (W10005). A staircase with three preserved built steps was uncovered in the northeastern corner of the plaza, apparently leading to a second floor to the south of the church. This floor was also evidenced by the numerous pieces of coarse white tesserae attached to a light lime bedding found in the collapse above the southern aisle and the oil press (below).

Remains of a built installation were exposed in the northeastern corner of the room, and an opening that led into a large room (L10012; 4.90 × 5.15 m) was paved with a white mosaic with red and blue tesserae. Adjacent to the eastern and western walls of Room

A wide opening with a threshold (2.14 m wide) was preserved at the southeastern end of the plaza. There were apparently two doors opening eastward

Fig. 3.20.3. H. Te’ena. The stable (A. Peretz).

39

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods

10012 were rectangular depressions built within the mosaic, probably the remains of stone piers that supported the arch of the ceiling. The southern wall of the room (W9954) was the outer wall of the entire complex and was built of large ashlars, some covered with fringed margins with a prominent boss. The wall was built over the opening of an undated rock-cut cave (L9965) that probably served for burial purposes (see below). It is possible that the arch in Room 10012 was built to disperse the construction weight on the ceiling of the cave. The northern wall of L10027 was not fully preserved, but it seems to have an opening which led into another room also paved with a white mosaic (L10015; 2.55 × 5.15 m). In the southeastern corner of this room, an entrance led into a large room (L10008; 5.15 × 5.85 m) to its south, identified as a stable. There was another entrance to this stable from L10012, through their shared W10024. Four rectangular stone basins aligned east–west (0.69 × 0.97 × 0.38 m, depth 0.25 m) were exposed at the center of the stable (Fig. 3.20/2), apparently used as feeding troughs. An oval stone trough (0.59 × 1.03 × 0.49 m, depth 0.30 m) with a drainage hole at its base was found next to the northern side of the western trough and perpendicular to it. The bedrock floor was leveled with small fieldstones and earth.

Fig. 3.21.1. H. Te’ena. The olive press (A. Shadman).

The stable seems to have accommodated donkeys that operated the adjacent oil press (see below) and used also for other needs in the monastery. The donkeys entered the stable through a wide opening in the southern wall, while people used three narrower openings in the western, northern, and eastern walls. The donkey operating the oil press would have walked out of the stable through a wide opening at the southern end of the stable’s eastern wall (W10009) leading to corridor L10011. There, it would have turned eastward to a wide opening (1.65 m) in the western wall of the oil press (W10010). A square stone basin (0.80 × 0.80 m), probably used also for feeding or giving water to the donkey, was exposed next to the opening of the oil press. A partially excavated circular installation (L10013; diam. 2 m) was exposed east of the corridor.

The crushing basin. Only the lower stone of the installation (recently damaged; diam. 1.50 m; height 0.50 m) survived. This stone may have rested, at first, where the screw weight was placed. The stone stood at the center of a circular bedrock surface (diam. c. 4 m) where a donkey walked. The uneven southern part of the natural surface was leveled with stone slabs, creating a flat walkway.

The oil press (Fig. 3.21). 

On the upper part of the piers were circular depressions aimed to insert a rod to stabilize the beam. A groove was hewn in each of the piers to facilitate the insertion and removal of the rod. A pressing basin (diam. 1.2 m), founded on a beaten earth bedding (L9959), was located c. 2.20 m west of the piers, and included a perimetric hewn groove (8 cm deep) aimed at flowing the liquid into the collecting vat. A small rock-cut cup was found north of the pressing basin, perhaps used for lighting. A rectangular collecting vat (0.60 × 1.26 m, depth 0.50 m), still bearing traces of pale gray plaster, was found north

The oil-pressing and extracting device. To the north of the crushing installation was the pressing area, which included pillars for anchoring the beam, a pressing basin, a collecting vat, grooves for stoppers, and the place for the weight and the screw. The two anchoring piers (height 1.80 m, thick c. 0.40 m) were found on the eastern part of the installation, one standing and the second lying next to it.

An oil press, partly hewn in the bedrock, was exposed south of the church, in what seems to be a pre-planned space for housing this installation. The oil press was operated by the ‘beam and a screw’ technique, but these parts were not preserved. Three main activity phases were identified: Phase I. The olive-press main activity phase, when it included crushing, pressing and collecting facilities. 40

Amit Shadman: Chapter 3, The Sites

Fig. 3.21.2. The underground cavity (A. Peretz)

Underground cavity. This cavity was found on the southern slope of the hill. It had a grand entrance, indicating it may have been used as a burial cave, and a convex, passage-shaped opening (width 1.20 m). This entrance included a hewn front room with a wall built of dry ashlars along its eastern side, founded on bedrock and preserved to a height of two courses (length 2.7 m, width 0.30 m, height c. 60 m). On this wall was the square impression of a building stone which had detached from it and was exposed in the western part of the entrance. A wall (Fig. 3.21.2; W9954) built of high-quality ashlars and founded on bedrock was discovered above the entrance to the cave. South of this wall, the edge of a room was found paved with a white mosaic (2 × 2 cm) consisting of a frame of two parallel rows of tesserae laid alongside the wall, and stones arranged in diagonal lines inside the room. The modern objects found in the cave point to the site being used as a shepherds’ shelter.

of the pressing basin. There was a shallow depression at the bottom of the vat for collecting waste. On two square stones, west of the collection vat, two grooves for stoppers kept the beam balanced. The grooves were distanced 0.50 m from each other, perhaps illustrating the thickness of the beam. An oval hewn pit (diam. 1.60 m, depth 0.60 m), found 7 m west of the piers, served to accommodate the weights. Phase II. In this phase, the press went out of use in its initial form, and if it continued to produce oil, it was by the ‘direct pressure’ method. At some point, the earlier weight was removed, and a double pavement was laid on top of the local soil and stone fragments that filled and sealed the hewn pit (L9967). The latest potsherds discovered in this fill were dated to the end of the 6th century CE, probably representing the time when the pit was blocked. Phase III. In this phase, the oil press was found covered with mosaic blocks that had apparently collapsed from the second floor. Second-floor galleries in churches, some paved with mosaics, are known from other sites, such as H. Hesheq in Western Galilee (Aviam 1994: 121). Hence, this phase actually represents the collapse of a second story above the press, after which the installation, and perhaps even the entire monastery, ceased to operate.

Water cisterns. A water cistern (diam. 1 m), with light gray plaster on its walls, was exposed on the southern slopes of the hill. Another cistern was discerned north of the hill, with three hewn steps (length 1.40 m) leading into it. Tepper (2013) suggested that these steps point to some connection to a nearby road. A third cistern was discovered on the northwestern slopes of the hill.

41

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods

The Agricultural Plot

where the edges of the road were built of fieldstones and partly dressed stones. No datable finds were found.

H. Te’ena dominates an area of about 1890 dunams spreading out in all directions. To the east, south and north of the site are rocky areas with agricultural terraces, whereas to the west, there are plateaus and alluvial valleys. Thus, most of the agricultural activity of this site took place on the west. The boundaries of the agricultural plot can be observed in several places. The eastern boundary is marked by a border stone, and the large stone that was discerned in the west probably marked the western boundary; to the south of the site, there was an enclosure wall, oriented east–west, which seems to have marked the boundary between this site and H. Yeqavim. The northern border is not clear, but it was probably bounded by the large areas west of Kh. Umm al-Hammam.

Road 3b. This road was located a few dozen meters south of Road 3a. It began c. 500 m southeast of the site, on the northern bank of a dammed gully, a tributary of Naḥal Shiloh, and ended 1 km west of the site, in a wide plateau with large enclosed plots. The road did not directly reach H. Te’ena, but in the absence of another site nearby, and the relative location of the fields to the site, it would seem it did indeed serve the farmers of H. Te’ena.  Enclosed agricultural plots (Fig. 3.22; Table 3.1). The built-up agricultural plots of this site were located to its west, in a large plateau suitable for cultivation. Twenty-four plots were identified, totaling 225 dunams. The largest plot (No. 1, one of the largest in the study area) was at the site’s foothill. This trapezoid-shaped 46-dunams plot was well demarcated by dividing walls. It was delimited on its eastern side by a double wall founded on bedrock, and by single walls on the other sides, also founded on bedrock and built of mediumsized fieldstones. The western wall of the plot merged with a dammed gully (see below). North of this plot, and adjacent to it, were five plots (Nos. 2–6), each extending over an 8–13 dunams area. Another cluster of plots was found northwest of the site.

Agricultural roads. The site was served by two parallel agricultural roads in a general east–west orientation (see Chapter 6, Fig. 6.4). These roads led to the cultivated fields on the east and the west sides of the site.  Road 3a. This c. 600-m long and c. 3-m wide road began southeast of the site (map ref. 198800/664940) and ran westward toward a large plot at the foot of the site. A 30-m long section was excavated southwest of the site

Fig. 3.22. H. Te’ena. Land distribution.

42

Amit Shadman: Chapter 3, The Sites

Table 3.1. H. Te’ena. Land distribution. Plot No.

Area in dunams

1

45.6

2

8.4

3

13.1

4

8.2

5

13.2

6

8.0

7

3.7

8

4.1

9

13.1

10

7.3

11

5.3

12

5.2

13

16.1

14

4.3

15

5.6

16

9.0

17

4.2

18

8.9

19

5.5

20

9.1

21

5.0

22

7.1

23

2.8

24

12.8

Total

Enclosure wall. The hill on which the site is located was enclosed by a wall built of two rows of stones founded on bedrock and preserved to a height of one course (width c. 1.0–1.5 m). North and east of the site, the wall is irregular and merges with the natural outline of the area. Its western part continued in a straight line on a north-south axis. Winepresses. Three winepresses were recorded at the site. The first winepress was located at the top of a gentle hill, at the center of the enclosed plots; it included a treading floor (2.15 × 2.75 m, depth 0.30 m) and a rectangular collecting vat (1.10 × 1.45 m, depth 1.25 m). The eastern part of this treading floor was damaged, leaving no traces of the ditch between it and the collecting vat. A hewn step on the southern wall of the collecting vat and the remains of hydraulic plaster (thickness 1 cm) were visible on all walls. A circular plastered depression (diam. c. 0.25 m), perhaps a settling pit, was found at the bottom of the collecting vat. The second winepress (No. 461; Fig. 3.24) was discovered southwest of the site. It was partly built and partly hewn into the bedrock and comprised a treading floor, a settling pit, and a collecting vat. The treading floor (L8018; 3.34 × 3.92 m) was hewn into the bedrock on its southern, western and eastern sides, while a wall (W8017), built of medium- and large-sized fieldstones and ashlars, was on its northern edge. The floor was coated with light plaster which survived only in the northwestern corner. The settling pit (L8048; 0.95 × 1.06 m, depth 0.29 m) was north of the treading floor, and the must was conducted into it through a hewn channel under W8017. A pipe made of gray bonding material coated with hydraulic plaster (L8050) was installed inside the channel. A poorly built channel connected the settling pit to the collecting vat (1.36 × 1.38 m, depth 1.30 m) to its east, which in turn had a small pit (L8051) in its southeastern corner. The settling pit and the collecting vat were coated with well-preserved light-colored plaster. The potsherds revealed in the excavation pointed out to the press having been operated in the Byzantine period, or that it at least ceased to function then.

225.53

The location of the agricultural crops. Most of the agricultural plot of H. Te’ena consisted of deep alluvial lands suitable for all local crops. On a hillside east of the site, were wide farming terraces with deep soil. To the west, a plateau drained into a wide damned gully, rich in high-quality soil. To the south and north of the site, there were rocky areas where farming terraces were built (Fig. 3.23). Given this location, the cultivated areas of H. Te’ena can be divided as follows: Grain was grown in the dammed wadis, and their wide outlets were covered with deep soil. The nearby threshing floor may support this suggestion. Vineyards and olive groves were grown in the rocky areas, where some winepresses were uncovered. The oil press, however, was found within a building belonging to the site’s constructed complex.

The third winepress was exposed on the northern slopes of the site. It comprised a square treading floor (3.40 × 3.70 m), a settling pit (0.70 × 0.73 m, depth 0.74 m) and a collecting vat (1.55 × 1.90 m, depth 1.30 m) coated with red plaster typical of the Byzantine period. A channel was hewn between the treading floor and the settling pit. It would seem that, at some point, the press was converted into a quarry.

43

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods

Fig. 3.23. H. Te’ena. The agricultural systems north of the site.

44

Amit Shadman: Chapter 3, The Sites

Fig. 3.24. H. Te’ena. Winepress No. 461 (50). 

Water cisterns. Guérin mentioned 10 water cisterns, presumably within the boundaries of the settlement. Other pits were identified within the agricultural plot, to the east and west of the site. The ceiling of one cistern had collapsed, but the gray plaster on its walls survived.

was from the north through an opening, whose fallen, large stone posts were found inside. The tower was preserved to a height of two–four courses, and the meager ceramic finds did not allow for accurate dating. A row of towers was found a few hundred meters east of the site. One tower, well preserved, was built of a double outer wall overlaid with small stones and earth. The tower’s interior space is small and is still roofed, today, with stone slabs. The entrance to the tower was from the northwest.

Stone clearance heaps. Many stone heaps were discovered near the site, most in a wide wadi to its west. The construction technique of the heaps varied, some being surrounded by a one- or two-courses high wall. Sometimes, the heaps leaned on the dams in the wadi.

Agricultural terraces. Terraces were found to the north, northwest, and southwest of the site. These areas are characterized by gentle slopes and natural steps on which the site’s agricultural terraces were constructed. 

Field Towers. Ten to fifteen field towers were discovered in the cultivated areas of the site, a relatively large number. Most towers were located on the rocky hills that surround the agricultural areas to the northeast and east of the site. Three circular towers and a rectangular one, located above a dammed wadi, were also recorded northeast of the site. One of these circular towers leaned on an agricultural terrace, and another large tower (outer diam. c. 8 m and inner diam. c. 2 m) was located near a rural road (2). The excavation revealed that this tower was built of two and sometimes three walls of large-sized fieldstones (0.50 × 0.70 m), some of which collapsed onto its interior. The entrance

Dammed wadis. The agricultural roads of the site led to two dammed wadis. Two hundred meters south of the site was a narrow wadi (width of c. 50–100 m) where about 60 dams were recorded along a c. 1.5-km long line (one dam/25 m). The wadi began west of H. Yeqavim, an area that was seemingly used by that site. However, in the lower run of the wadi to the west, the number of dams significantly decreased. Fifteen additional dams were recorded in a second wadi (length 750 m, width 250 m), 500 m west of H. Te’ena. 45

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods

Threshing floor. Exposed on the northern fringes of the site, this floor (diam. 17.5 m; Fig. 3.25) consisted of a circle of large-sized stones laid on bedrock. The interior of the circle was filled with local soil and small- to medium-sized fieldstones. In some places, this fill was c. 0.5 m deep. The location was selected in accordance with the prevailing winds.

The Pottery — Yehudah Rapuano Introduction. The sample of ceramic finds recovered from the excavations at H. Te’ena and illustrated here is intended to give the reader a general picture of the pottery discovered at the site. The assemblage includes several examples of imported Late Roman Red Wares and a single example of a Fine Byzantine bowl. Otherwise, the pottery is, for the most part, made of plain, coarse fabric. The pottery is considerably homogeneous in date. Consequently, it was decided by the excavator that it should be presented as a single assemblage arranged typologically.

Quarries. Two rather small quarries were discovered near H. Te’ena. One (2.30 × 5.00 m) was located south of the site, and the other to its southwest. In this latter quarry, whose quarrying was not yet complete, several items were discerned, among them a large pillar or lintel.

Bowls (Fig. 3.26:1-8). A number of Late Roman Red Ware bowls were recovered from the site. Four examples belong to the Late Roman C Form 3: Fig. 3.26:1 is a Hayes’ Late Roman C Form 3E bowl, decorated with three registers of rouletting on the face of the rim, and dated to c. the late 5th–6th centuries CE. Fig. 3.26:2 is a Hayes’ Late Roman C Form 3H bowl dated within the 6th century. In Fig. 3.26:3-4 are depicted pseudo ring bases of Late-Roman C Form 3 bowls. Fig. 3.26:4 shows a stamped cross decoration with a circle beneath its arm. Similarly, Fig. 3.26:5 is also a fragment from the base of a Late Roman C bowl with an impression of a cross

Caves. A natural cave (diam. 6 m, depth 5 m) was excavated southwest of the site. Two shafts, one square and the other rounded, were hewn in the ceiling and descended to its center. Unlike the surrounding areas, the cave was cut in a particularly soft rock which caused its total collapse. After the ceiling was removed, potsherds from various periods were found, especially from the Iron Age II and the Persian and Roman periods. It was not used as a burial cave, and since most of the sherds belonged to storage jars, it may have served for storing agricultural products.

Fig. 3.25. H. Te’ena. A threshing floor, a winepress, and an enclosed wall (Sky View, A Shadman).

46

Amit Shadman: Chapter 3, The Sites

with circles beneath its arms. Hayes dated this stamped motif from the late 5th to the early 6th centuries CE. Fig. 3.26:6 depicts a Cypriot Red Slip bowl, Hayes’ Form 9, dated between c. 580/600 and the end of the 7th century CE. Fig. 3.26:7 is the sole example of an Egyptian Red Slip Ware bowl found in the excavation, evidently Hayes’ Form J, dating from the late 6th to the early 7th centuries CE. A single Fine Byzantine Ware bowl, Fig. 3.26:8, with its distinctive out-turned rim, is a Magness’ FBW Form 2C bowl dating from the 7th to the mid-8th centuries CE.

Roof tile. Fig. 3.27:6 is a Byzantine period roof tile. The tegula was marked by a reed on its upright edge, evidently intended to assist the roofer in aligning the tiles properly.

Mortarium. Fig. 3.26:9 is an imported, Syrian Mortarium made of coarse dark brown fabric, with a heavy, externally folded, knoblike square rim. It is dated to the 5th and 6th centuries CE.

Conclusions. The ceramic finds from H. Te‘ena are for the most part very typical of the Late Byzantine pottery of the region. All the examples can be dated to the 6th and the 7th centuries CE. There is no clear indication from the pottery that the site continued into the Umayyad period. The Late Roman bowls, Late Roman C Form 3 and Cypriot Red Slip Form 9, are forms one would expect to occur in Judea at the end of the Byzantine period. The Egyptian Red Slip bowl is slightly rarer. The rest of the assemblage presented consists entirely of utilitarian vessels; the local storage jars and the Majcherek Form 4 Gaza jar are all very common. The Large candlestick lamp is the most common type found during the period. The one exception is the cylindrical object Fig. 3.27:5 pierced with holes and decorated with a herring-bone pattern. The definition or use of this object remain unclear. The impressed decoration is somewhat reminiscent of the ornamentation found on vessels at Khirbet el-Jiljil near Ramat Beit Shemesh (Vincenz de 2005:132-137).   

Oil lamp. The large-form candlestick lamp in Fig. 3.27:7 corresponds to Magness’ Form 3, dated by her from c. the mid-6th century to the late 7th/8th century CE. Sussman (2017: 96-101, Fig.76:2) classified it as Lamp B14.2:1, dating it from after the mid-4th century to the 7th century CE.

Basins. Two basins are presented: Fig. 3.26:10 is an Arched-Rim Basin, Magness’ Form 2B, with a band of wave combing on top of its rim and an impressed thumb decoration on the rim’s outer edge. It is dated from the 6th to the late 7th centuries CE, though Magness noted the existence of a few pre-6th century CE. examples. Fig. 3.26:11 is a basin fragment with a heavy triangular rim, with combing on its rim and a band of straight combing on its body below the rim. It is an Arched-Rim Basin Magness’ Form 3 or a variant, dating from the 6th to the late 7th/early 8th centuries CE. Storage Jars. Fig. 3.27:1 is an example of a Magness’ Form 6A storage jar with an upright rim and a cylindrical neck, and a collar ridge at the base of its neck, dated by her from the late 6th/7th to the 8th centuries CE. Closely related, Fig. 3.27:2 is a Magness’ Form 6B storage jar. It has an upright rim, a conical neck and a collar ridge at the base of the neck. Like the previous example, it is dated to the 6th/7th and 8th centuries CE. Fig. 3.27:3 is a coastal bag-shaped storage jar made of reddish-brown ware. It has an everted rim and a swollen neck. It lacks a collar-ridge at the base of its neck and bares sharp ridging on its shoulder. AdanBayewitz, following Riley, classified this type of jar as an Amphora IB at Caesarea, dated from about the 5th to the 6th centuries CE. Its description fits Magness Form 7 as well, dated by her from the late-7th to the 9th/10th centuries CE. The features of our jar indicate that it dates within the earlier part of this range. Magness noted that the 7th- and early 8th-centuries form had a lower neck, like our example. Fig. 3.27:4 is a Gaza storage jar, Majcherek Form 4, with a cupped-rim and accretions on the rim and shoulder. It is dated to the late 6th and the 7th centuries CE.

Summary The cenobitic monastery at H. Te’ena was probably founded during the 5th century CE and continued at least until the end of the 7th century, and possibly even later. Several changes were carried out during this period, such as removing a ‘screw weight’ in the oil press, blocking a room in the church, and adding an enclosed room east of the apse. The site was approached by the main rural road (3), pointing to its high status. This road connected H. Te’ena with Deir Balut, these two settlements being seemingly closely related in the Byzantine period. An extensive agricultural area, west of the site, was characterized by alluvial soil deposited in the wadis behind the dams. Farming terraces were built in the rocky areas, thus exploiting the area to the maximum extent. The oil press, the winepresses, and the large threshing floor illustrate the main agricultural crops in the region, at least during the Byzantine period.

Miscellaneous. Fig. 3.27:5 is evidently a cylindrical object pierced by rows of holes with a decorative line of herring-bone impressions in between. The purpose of the holes is unclear. They may have served for draining liquid or to expel hot air or gasses. 

Guérin: described a larger number of cisterns and ruins: “A totally destroyed village on a low hill. There were about twenty small houses, of which only the walls’ outlines can be discerned. About ten water cisterns were hewn there, some of which were later 47

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods

1 2

3

4

5

6 7

8

9

10

11 0

10

Fig. 3.26. H. Te’ena. Pottery.

3

2 1

4 5

6 0

10

7 2

0

Fig. 3.27. H. Te’ena. Pottery.

48

Amit Shadman: Chapter 3, The Sites

‹Table for Figs. 3.26, 3.27 Fig.

Form

Locus

Basket

Description

Parallels

3.26:1

Bowl

9998

96141

Red (10R 5/6) ware; no core; few, tiny, white inclusions

Hayes 1972: 329–338; Late Roman C Form 3E (suggested date: late 5th–6th century).

3.26:2

Bowl

9999

96153

Light red (2.5YR 6/6) ware; no core; no visible inclusions

Hayes 1972: 329–338; Late Roman C Form 3H (suggested date: 6th century).

3.26:3

Bowl

1012

96173

Red (10R 5/6) ware; no core; virtually no inclusions

Hayes 1972: 329–338; Late Roman C Form 3 (suggested date: 6th–mid-7th century).

3.26:4

Bowl

Stamped decoration: Crossmonogram with two pedants beneath its arms

Hayes 1972: 329–338; Fig.67:19, Late Roman C Form 3F (early 6th–mid-7th century. Stamp no.16 363-364, Late Roman C ware, stamp no.67k Group III dated late 5th–early 6th century.

3.26:5

Bowl

Stamped decoration: Cross with two pendants beneath its arms

Hayes 1972: 364, Late Roman C ware, stamp no.67j Group III dated late 5th– early 6th century.

3.26:6

Bowl

10001

96070

Red (2.5YR 5/6) ware; no core; virtually no visible inclusions

Hayes 1972: 379–382 Cypriot Red Slip Form 9 B bowl. Suggested date: c.580/600 to end of 7th century.

3.26:8

Bowl

9985

96033

Pink (5YR 7/4) ware; no core; red (2.5YR 5/6) int. burnished slip; few, tiny reddish-brown and white inclusions

Hayes 1972: 387–391 Egyptian Red Slip Ware Form J (late 6th– –early 7th century).

3.26:9

Bowl

9998

96154

Reddish brown (5YR 4/4) ware; no core; few, large to tiny, dark reddish-brown and white inclusions

Riley 1975, 36–37, nos. 41 and 42; Blakely, Brinkmann and Vitaliano 1992: Hayes J.W. 1967: 338, Pl.3:6; Ras al Bassit, Syria: Mills and Reynolds 2014: 134, Fig.7:28, 30 dated 5th–6th centuries.

3.26:10

Bowl

3.26:11

Basin

9998

96095

Light reddish brown (5YR 6/3) ware; no core; few, medium to tiny, dark grey and white inclusions

Jerusalem- Magness 1993: 204–208. Arched-Rim Basin Form 2B (suggested date: 6th–late 7th centuries (with a few pre-6th century examples).

3.26:12

Basin

9998

96085

Light reddish brown (5YR 6/4) ware; brown (7.5YR 5/4) core; few, medium to tiny, white and reddish-brown inclusions

Jerusalem- Magness 1993: 204–208. Arched-Rim Basin Form 3 (suggested date: 6th–late 7th/early 8th centuries).

3.27:1

Storage jar

10006

96156

Light red (2.5YR 6/6) ware; no core; few, small to tiny grey and white inclusions

Jerusalem- Magness 1993: 227–230 Storage jar Form 6A (suggested date: late 6th/7th–8th centuries).

3.27:2

Storage jar

9996

96029

Heavy, light reddish brown (5YR 6/4) ware; faint, dark gray (5YR 4/1) core; few, small to tiny, white, red and dark grey inclusions

Jerusalem- Magness 1993: 227–230 Storage jar Form 6B (suggested date: late 6th/7th–8th centuries).

3.27:3

Storage jar

10008

96133

Red (2.5YR 4/6) ware; no core; few, medium to tiny, pink, reddish-brown and dark grey inclusions

Jerusalem- Magness 1993: 230–231 Storage jar Form 7 (Suggested date: late 7th–9th/10th centuries).

Jerusalem- Magness 1993: 200 Fine Byzantine Ware bowl Form 2C.

49

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods ‹‹Table for Figs. 3.26, 3.27 Fig.

Form

Locus

Basket

Description

Parallels

3.27:4

Gaza Storage jar

10012

96175

Yellowish red (5YR 5/6) ware; no core; few, tiny, dark grey and white inclusions

Majcherek 1995:169 Form 4 (dated late 6th–7th centuries).

3.27:5

Cylindrical object pierced with rows of holes

3.27:6

Roof Tile

9989

96004

Light brown (7.5YR 6/4) ware; no core; few, tiny, brown and white inclusions

Ramat Rahel - Aharoni 1962: Fig.2:1 Strat. II (Byzantine).

3.27:7

Lamp

10021

96188

Brown (7.5YR 5/3) ware; no core; few, tiny reddishbrown inclusions

Magness 1993: 251–255 Oil Lamp Form 3, (suggested date mid-6th–late 7th/ early 8th centuries); Sussman 2017: 96–101, Fig.76:2. Lamp B14.2:1 dated after mid-4th–7th centuries.

The northern wing

enlarged and now serve as an underground shelter for the shepherds.” (Guérin 1984: 92).

This wing (11 × 19 m) included 11 rooms oriented east– west arranged in two rows. The southern row comprised five rooms (I–V). The entrance to Room I (2 × 4 m) was through the eastern wall, and Room II (2 × 3 m) was entered through three steps in the partition wall shared with Room I. Room III (3 × 4 m) was a central space entered through an opening in the middle of its southern wall. Room IV (2 × 3 m) was adjacent on the west to Room III, and further west was Room V (2 × 3 m). The northern row, seemingly a later addition, comprised six rooms (VI–XI) separated from the southern rooms and the courtyard by a massive wall (W142). The entrance to the northern row was through two openings in W160. The western entrance led into Rooms VI (2 × 5 m) and VII (2 × 5 m), which were isolated from the other rooms. The eastern opening led into Room VIII (3 × 5 m), where a floor bedding of smallsized fieldstones was revealed. Two small rooms or cells (c. 2 × 2 m each) were found east of this room. In the eastern wall of Room IX, was an opening into Room XI (2 × 4 m).

The PEF surveyors cited Guérin, but added: “Guérin gives this name to a small ruin lying due east of Mejdal Yaba. He found ten cisterns cut in the rock, and the foundation of houses. There is a nameless ruin on the map east of Mejdal Yaba, which is possibly the place he means.” (Conder and Kitchener 1882, II: 358). Numismatic finds: A Byzantine bronze coin dated to Justinian I (518–527 CE), minted in Constantinople; an Umayyad bronze coin from the beginning of the 8th century CE. Previous studies: Conder and Kitchener 1882, II: 358, Dar 1982a; Guérin 1984: 92; Kochavi and Beit Arieh 1994: Site 153; Safrai 1995b; 1995c; 1997a; Haiman 2012a; Tepper and Tepper 2013. Site No. 4 Naḥal Shiloh Map reference: 198850/665250

The Southern Wing

A farmhouse (28 × 36 m, Figs. 3.28.1, 3.29) founded at the end of the 8th century BCE, operating also during the Persian and the beginning of the Hellenistic periods. The building comprised two wings with a courtyard (10 × 21 m) in between. The southern wing had a single row of rooms, while the northern wing had two rows. Both the outer and inner walls were wide (1.00–1.20 m) and mostly built of two rows of roughly hewn stones. Some walls were preserved to a height of seven courses. The entrance to the courtyard was through two doorways: one in the eastern wall (W158) and the other in the northern wall (W143). The two wings were accessed from the courtyard, but the northern wing had also a direct entrance through the eastern wall of the complex (W182). Three installations were exposed in the courtyard: a tabun, in its northeastern corner; a rockhewn simple press, in its eastern part; and a circular silo, in its western part (diam. 1.5 m, 1.5 m deep).

This wing (8 × 24 m) was partially destroyed by a later Ottoman limekiln built in its center. Three rooms (XIV–XVI), well preserved up to the lintel level (c. 2 m), were exposed in the eastern part of the wing. A silver Tetra-Drachma minted in Athens, and a Sidonian silver coin, both dated to the 4th century BCE, were found in these rooms. The rather meager remains of five additional rooms were exposed: XII–XIII on the east and XVII–XIX on the west. As with other farms, here too, the rooms were intentionally filled. Six more rooms (XXI–XXVI) were excavated south of the southern wing, separated from it by a massive wall (probably the Iron Age II original wall of the complex). The construction of these rooms alongside a part of this wall indicates they were built at a later stage, probably in the Persian period. Numerous potsherds, mainly of jars from the Persian period and of a few 50

Amit Shadman: Chapter 3, The Sites

Fig. 3.28.1. Naḥal Shiloh. Aerial view of the farm, looking northwest (Sky View, A. Shadman).

Hellenistic amphorae, suggested this was a storage area. 

east and west sides. The agricultural plot extended over a c. 400-dunams area, most of it rocky in nature. Several field towers, probably of a later date, were discovered in this area. Excavation results showed that these towers were mainly in use during the Byzantine and Early Islamic periods. Four simple winepresses were discovered near the farm, one to its west and three to the north and east. Only a single water cistern was discovered within the boundaries of the agricultural plot.

Three rooms were found in the eastern part of the complex: Room XXI (4.30 × 7.00 m), accessed through an opening in its southern wall, and Room XXVI (2 × 4 m), entered from the west and from Room XXII, where a poorly preserved tabun was uncovered in its southeastern corner. An underground water reservoir, using a narrow natural karstic cavity at least 6 m deep and showing remains of gray coarse plaster on its walls, was found west of Rooms XXIII–XXV. Its upper part was covered with large stone slabs, and a paved staircase leading to Room XXIV was built in its eastern part. The reservoir was apparently intentionally filled, at some point, with local fieldstones. 

Ceramic finds: Iron Age II; the Persian, Hellenistic, Mamluk and Ottoman periods. The large kiln was apparently operated during the last two periods. Numismatic finds: A 4th century BCE silver coin minted in Sidon, a 4th century BCE Tetra Drachma minted in Athens, a bronze coin dated to Ptolemy I (305–283 BCE), a 4th century CE bronze coin from the reign of Constantine II.

Several installations were discovered near the farm. A rock-cut cistern filled with alluvium was located next to the western wall of the building. Three simple presses were on a bedrock surface west of the building, revealing, perhaps, the role played by olive oil in the local economy. The Agricultural Plot This site controlled extensive agricultural areas, and most crops were presumably processed in a rocky ridge north of the site, delimited by two dammed wadis on its

0

0.5

Fig. 3.28.2. A 4th century BCE silver coin minted in Sidon.

51

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods

Fig. 3.29. Naḥal Shiloh. Plan.

Room 2 was through a doorway placed on the same axis with the main entrance to the building. Room 2 (6.40 × 14 m) was a broad room and the largest in the building, from which broad Room 3 (2.20 × 14 m) was accessed.

Previous Studies: Finkelstein 1978a: 12, Table 16; 1981: 338–339 and drawing 6 (lower part), 346; Kochavi and Beit Arieh 1994: Site no. 151. Site No. 5 Naḥal Shiloh (West) Map reference: 197940/665300

The entrance to Room 4 (2.60 × 4.80 m) was through a doorway in its west wall. This entrance comprised two doorposts built of elongated fieldstones. Room 4 was connected to Room 5 (2.00 × 4.20 m) to its west through a doorway, also built of two doorposts made of elongated fieldstones and aligned with its own entrance. Room 6 (2.20 × 4.40 m) was exposed west of Room 5 and its entrance was aligned, again, with the doorways of Rooms 4 and 5. A corridor (1 × 9 m), beginning in the southwestern corner of Room 6, led to Room 10 and the southern end of the building. Two steps ascending to an internal complex (Rooms 7, 8—1 × 2 m each; Room 9—2.00 × 2.80 m) of sealed cells were revealed in the middle of the corridor. 

An agricultural farm on the northern gentle hillside, north of a tributary of Naḥal Shiloh and c. 600 m northwest of H. Te’ena. The main building was damaged as a result of contemporary works (Finkelstein 1978a, Kochavi and Beit Arieh 1994). The farm (26 × 40 m) comprised a central wing (26 × 26 m) with 14 rooms, and a western wing (10 × 18 m) with five rooms (Fig. 3.30). Another room (3.60 × 6.00 m), probably added to the building at a later phase, was traced in its northeastern corner, bringing the number of rooms to 20 in total. The entrance to the building was in the middle of its northern wall through an indirect gate approached, from the west, through a walkway parallel to the wall and subsequently turning south into Room 1 through a wide opening (1.80 m). A floor bedding built of small-sized fieldstones, laid on bedrock, was revealed in Room 1 (3.40 × 5.40 m). This room allowed passage to three rooms: Room 2 on the south, Room 4 on the west and Room 13 on the east. The entrance to



The walls in Rooms 1–9, preserved higher (c. 1.50 m) than any other walls in the building, were built of especially large-sized fieldstones, and the rooms were apparently used for storage. This part of the building seems to have had a second story or an attic. The bedrock in Room 9 was filled to level the surface, and a floor bedding of small stones may have served as the floor. Room 10 (2 × 5 m), at the southern end of 52

Amit Shadman: Chapter 3, The Sites

Fig. 3.30. Naḥal Shiloh (West). Aerial view of the farm (Sky View, A. Shadman).

the long corridor, allowed passage to Rooms 11 and 12 (width 3.20 m and 4 m respectively) to its east.

single row of particularly large-sized fieldstones and differed from the outer walls of the main building. The two wings were not connected, and the entrance to the western wing was via a doorway (width 1 m) on its west side. It is therefore clear that the western wing was a later addition to the main building. 

Only the doorway’s stone threshold, founded on bedrock, survived between these rooms. A row of smooth fieldstones laid on the bedrock, perhaps the remains of pavement, was exposed in the southern part of Room 12. Neither the southern wall of the building nor the southern end of this row of rooms was preserved. In the northern part of the building, Room 1 led to Room 13 (4.80 × 5.50 m) to its east through a doorway at the northern end of their shared partition wall. Room 14 (2.00 × 4.80 m), east of Room 13, had been damaged by the construction of an Ottoman limekiln. Room 15 (c. 3.60 × 6.00 m), adjacent to the building’s outer northeastern corner, was probably a later addition. The outer walls of the building were solid (width 1.40 m), built of two rows of large-sized fieldstones with roughly hewn stones and medium/ small-sized fieldstones in between. The interior walls were narrower (0.80 m), except for the northern wall of Room 2, as solid as the outer walls and perhaps supporting the rooms to its southwest built on the slope.

Two steps in the entrance to the western wing ascended to Room 16 (9 × 11 m). A row of six north–south oriented column bases, which probably supported the ceiling, was discovered across this room. Since the room was built on a slope, the floor level slanted downward from south to north, in accordance to the bedrock level. In the northern part of the room, the bedrock was hewn and leveled to serve as a floor. Two square installations of an unclear function were exposed on the floor of the room, one (1.60 × 1.70 m) north of the doorway and adjacent to the western wall of the room, and the other (1.55 × 1.55 m) in the northeastern corner of the room. Rooms 17 (2.60 × 4.00 m), 18 (2.60 × 3.00 m), and 19 (4.60 × 5.00 m), adjacent to each other, were exposed east of Room 16. Their respective doorways were not located. North of Room 16, Room 20 (5.00 × 5.80 m) accommodated a circular installation (diam. 3.40 m, depth c. 0.80 m) consisting of an outer circle of mediumsized fieldstones and an inner circle of small-sized fieldstones. Burned limestone and lumps of raw lime

Another wing (10 × 18 m), abutting the western wall of the complex’s central wing, was exposed west of the building. The outer walls of this wing were built of a 53

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods

were exposed above its bedrock floor. This installation seems to have served as a silo for grain storage. North of Room 19 was a rectangular installation (1.60 × 3.00 m) built next to the wall that separated the west wing from the central one. The installation, built of medium-sized fieldstones, had a pavement bedding made of smallsized fieldstones.

1994: Site 149), may have been used, in the Byzantine period, also by the farmers of H. Te’ena. This winepress was hewn on a leveled rock surface (Fig. 3.32) and comprised a treading floor (2.15 × 2.75 m, depth c. 0.30 m) and a rectangular collecting vat (1.1 × 1.45 m, depth 1.25 m, capacity c. 2000 liters). The treading floor, damaged or naturally corroded, caused the destruction of the channel or drainage pipe running between the floor and the vat. A hewn step, or a projection, on the southern wall of the collecting vat, enabled descending into it. Also, traces of hydraulic plaster (thickness 1 cm) were visible on all the walls. At the bottom of the collecting vat, a circular partially plastered settling pit (diam. c. 0.25 m) penetrated the northern wall through the bedrock and cut down to the level of natural soil below bedrock.

The high-level bedrock within this complex was sometimes incorporated in both floors and walls. For example, in Room 16, the bedrock was leveled into a flat floor. In the other rooms, floors were formed by leveling the bedrock and inserting small fieldstones and beaten earth in the gaps. Some rooms contained rock-cut cupmarks of an unclear function, although one of them probably served for grinding. Potsherds of various vessel types found in all the rooms suggest the farm was established in the Iron Age II (8th century BCE) and lasted until the Hellenistic period (3rd century BCE). A 4th-century BCE silver Tetra Drachma (Fig. 3.31) was found on the floor of Room 5, next to its northern doorpost. 

The second press was located a few hundred meters northwest of the site. It comprised a rectangular treading floor (3.00 × 3.80 m, depth 0.23 m), a rectangular settling pit (0.80 × 0.90 m, depth 0.70 m) and a rounded collecting vat. The treading floor sloped northward, and a cupmark (diam. 0.22 m) was found in its southwestern corner. Next to the northwestern corner of the treading floor was a settling pit with a depression (diam. c. 0.20 m) at its bottom and a connecting rock-cut channel (width c. 0.17 m, depth c. 0.27 m). No traces of plaster were observed on the pit’s walls. Another pit was hewn in the lower part of the settling-vat western wall but was later sealed. A second drainage opening (diam. c. 0.18 m), connecting the settling pit to the collecting vat, was c. 0.40 m above the bottom level of the settling pit. A bell-shaped collecting vat (diam. 1.1 m, bottom-level diam. 2 m) with remains of hydraulic plaster (thickness c. 0.10 m) was found west of the settling pit. A shallow depression (diam. c. 0.80 m) was found at the bottom of the vat, next to the eastern wall, and two rock-cut steps were exposed on the southern wall of the vat. Eight cupmarks and depressions, of various diameters, were discovered adjacent to the winepress.

A limekiln (diam. 4.4 m, depth 2 m), probably from the Ottoman period, was built over the eastern wall of the building. The fire opening was apparently cut in its southeastern part and the cone-shaped waste piled up around the kiln and covered a wide area. Limekilns were a common phenomenon in the area since the building stones of these isolated sites were an excellent source of available raw material for the lime industry. 

0

1

Water cisterns. A cistern (depth c. 5.5 m, Fig. 3.33) was found c. 30 m northwest of the farm, with its capstone preserved in situ. Parts of a thick plaster floor, upon which potsherds from the Ottoman period were found, were exposed around it. However, since the cistern was presumably initially hewn when the farm was first constructed, in the Iron Age II, and continued to function in later periods, the plaster remains may be attributed to the time of the limekiln discussed above. 

Fig. 3.31. Naḥal Shiloh (West). A Hellenistic silver coin. Left: the head of Hercules; right: Zeus.

The Agricultural Plot Fairly suited cultivation alluvial areas were located north and northwest of the site’s foothill. The boundaries of the farm are unclear, as two other farms were situated north and south of it (Sites No. 2 and 22 respectively). Based on the general distribution of these farms’ areas, the agricultural plot of this site had probably around 400 dunams.

Agricultural terraces. Four terraces, aligned in a general north–south orientation, were recorded c. 20 m northwest of the farmhouse. The terraces were built of medium-sized fieldstones and were preserved to a height of c. 1 m. They held a c. 1-m deep layer of grumusol-like alluvium and covered a total area of c. 10 dunams (for an OSL dating, see Chapter 6).

Winepresses. Two winepresses were excavated. The first, northeast of the site (Kochavi and Beit Arieh 54

Amit Shadman: Chapter 3, The Sites

Fig. 3.32. Naḥal Shiloh (West). A winepress. (Sky View, A. Shadman).

Fig. 3.33. Naḥal Shiloh (West). A water cistern northwest of the farm.

55

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods

Dammed wadis. A dammed wadi (width 100 m, length c. 250 m) was found west of the agricultural terraces. Six dams were built across the wadi and were apparently used to grow most of the agricultural crops of the site. West of the wadi was a rocky area where most of the stone clearance heaps were piled up.

Numismatic finds: A silver Tetra Drachma dated to 325 BCE. Previous Studies: Finkelstein 1978a: 11, Table 17; 1981: 332, 342–343, drawing 8 (upper part); Kochavi and Beit Arieh 1994: Site no. 142; Tendler and Shadman 2015: 193–207; Shadman, Tendler and Marcus 2015: 84–88.

Summary

Site No. 6 Wadi al-Bureid (East)2 Map reference: 199620/665580

The excavation results of this farmhouse showed that it was founded in the Iron Age II and continued to function continuously until the Hellenistic period (end of the 3rd century BCE; Fig. 3.34). No major changes seem to have taken place at the site during this long period. The latest addition was the construction of the western wing, clearly attached to the original building. The large silo and the winepresses attest to the agricultural produce in the area. However, during the Byzantine period, most of the agricultural systems, including the terraces, dams and field towers, were associated with the sites of H. Te’ena and Migdal Afeq, both active in this period.

An agricultural farm, c. 200 m northwest of Kh. Umm alHammam. The remains of four buildings were discerned on the southern hillside. The western building (7 × 10 m; Fig. 3.35), built on a northeast–southwest orientation, was preserved to a height of three–five courses. It was constructed of large- and medium-sized fieldstones and ashlars, some bearing a bonding material. The doorway, in the western wall of the building, had two large stone jambs, including the bare sockets. The stone lintel was discovered next to the building. West of this doorway was some sort of plaza (2 × 7 m). An additional doorway was found in the southern wall of the building: one of its doorjambs had drafted margins, and the other was

Ceramic finds from previous studies: Iron Age II, the Persian and Hellenistic periods.

2

1

3

4

5

6

7

9

10

8

11

13

12

14

15

16 0

10

Fig. 3.34. Naḥal Shiloh (West). Ceramic finds. Iron Age II: hemispheric bowls (1–3), cooking pots (4–5); Persian period: mortaria (6–7), cooking pots (8–10), jars (11–13), jug (14); Hellenistic period: jars (15–16).

2  The description of the agricultural plot of this site is included in the discussion of Kh. Umm al-Hammam (site No. 11).

56

Amit Shadman: Chapter 3, The Sites

Ceramic finds: Byzantine period (potsherds collected in the agricultural terraces west of the site).

made from a stone in secondary use, perhaps of an oil press. A room (3 × 4 m), probably entered from within the building, was discerned in its northeastern corner. A bell-shaped cistern with a capstone was discovered c. 2 m southwest of the building, and another cistern was found c. 70 m north of it. A large limekiln, from a later phase than the building, was found c. 20 m to its north. 

Previous Studies: Kochavi and Beit Arieh 1994: Site 155; Safrai 1995b; 1995c; 1997a. Site No. 7 Kh. Umm al-Bureid Map reference: 198800/666500

The eastern building (5 × 7 m) was located c. 100 m southeast of the western one and was accessed from the southeast. It was built of slightly worked fieldstones and was preserved to a height of two–three courses.

The site was located east of the junction of Wadi alBureid and Wadi al-Susi. A complex of rooms (35 × 40 m; Fig. 3.36) was discerned on its southern and eastern sides, perhaps a Byzantine monastery, although it could not be clearly identified. Presumably, it included a residential wing (No. 5), and a courtyard (4 × 10 × 15 m) to its north where the large opening (2 × 2 m) of a cistern was found. The width of the cistern’s opening and its shape indicate that it had been probably previously used as the collecting vat of a winepress. Two projections in the northwestern corner of the cistern’s entrance possibly led down into it.

Two additional structures were found c. 80 m north of the first two buildings. Only one course of the easternmost building (7 × 9 m), entered from its eastern side, survived. A fragment of a lower crushing stone (diam. 1.60 m) belonging, perhaps, to an active oil press, was discerned nearby. The western, smaller building (3 × 4 m) may have been a field tower or a storehouse. An agricultural road approached the site from Kh. Umm al-Hammam (Fig. 3.35). A northwest–southeast oriented field wall was found south of the northern building.

Fig. 3.35. Wadi al-Bureid (East). The western building, looking southwest (A. Shadman).

57

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods

A large reservoir (6 × 12 m, depth 4 m, capacity 288 m3; Fig. 3.37) was located on the southwestern side of the complex. Its lower part was hewn, while its upper part (width 1.4 m) was built of small-sized fieldstones with some bonding material and coated with reddish plaster. The reservoir was also used in later periods, undergoing some changes. Inside the reservoir, a rounded stone was found, probably a cistern’s capstone. A modern animal pen was located south of the reservoir, and an isolated hewn wall was found north of it. Further to the north was an underground cavity coated with a reddish plaster, similar to the one found in the reservoir. A staircase led from the northwest into this underground cavity and possibly also to the reservoir.

five rooms—three rooms on the west and two rooms on the east (1.0–4.5 × 5 m, 2.0–4.5 × 5 m)—located north of the courtyard. A wall, partly built of dressed stones, surrounded these rooms and formed a square enclosure. Numerous tesserae, many colorful, as well as fragments of roof tiles, were revealed near the site. A limekiln, probably from the Ottoman period, cut into the site’s northwestern side.

An installation (1.20 × 1.70 m), built of smallsized fieldstones and coated with reddish plaster containing potsherds, was discovered nearby. Although the nature of this installation could not be ascertained, it was seemingly associated with the reservoir and the underground cavity. An oil-press lower crushing stone was found in a room forming part of a cluster of

Summary

The agricultural plot was quite small and so was the number of the agricultural installations found at the site. The absence of winepresses and field towers is noteworthy, while the presence of an oil press may indicate that olive groves and oil production were the site’s main agricultural crop. 

The settlement at Kh. Umm al-Bureid was small, as evidenced by its size and small agricultural plot. Based on the numerous tesserae and many fragments of roof tiles, the building may be interpreted as a church or a monastery, and not necessarily as a domestic-

reservior

Fig. 3.36. Kh. Umm al-Bureid.

58

Amit Shadman: Chapter 3, The Sites

Fig. 3.37. Kh. Umm al-Bureid. The reservoir, areal view (A. Shadman).

agricultural settlement. The large reservoir may have supplied water to the site, perhaps also serving for irrigating the fields or for some religious purposes. 

down. Now its stones are piled, among which I noticed many white tesserae that once formed the floor of the church. Three broken pillars indicate that the nave and the aisles were separated by several columns that were probably removed elsewhere, or burned to lime in a nearby kiln. In the vicinity of the site ... around this building, walls of about twenty small square houses are visible. Water cisterns and a pool, partially hewn in bedrock and partially built of ancient blocks, and coated with thick plaster.” (Guérin 1984: 94)

The small agricultural plot and the low number of agricultural installations found at the site support this interpretation. The road that connected this site with Kh. Kesfa (Site No. 13) indicates that these sites were interconnected. The lack of a wall dividing between the areas of these two communities suggests that they shared the cultivated areas located west of Kh. Kesfa.

Ceramic finds from the present study: a few fragments from the Persian and Hellenistic periods; bowls and lamps from the Byzantine period (5th–7th centuries CE); bowls from the Early Islamic period (Fig. 3.38).

Guérin:  “... It spread on a hill surrounded by a wall built of more or less large poorly hewn stones, and I noticed some fuzzy walls from an ancient west–east oriented building, which was apparently a Christian church. I could not accurately measure it because it was torn

Previous studies: Guerin 1984: 94; Kochavi and Beit Arieh 1994: Site 108; Safrai 1995c: 102–103; Bagatti 2002: 158.

1

2

3

4

6 5 0

10

Fig. 3.38. Kh. Umm al-Bureid. Ceramic finds from Kochavi and Beit Arieh 1994: Site 108. Byzantine period: bowls (1–2), a jug (3), a juglet (4); Early Islamic Period: a bowl (5), a jug (6).

59

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods

Site No. 8 H. Dayyar  Kh. ed-Duweir Map reference: 198450/667280. 

walls (width of 1.20 m) on its north and east sides. The main, central room (8 × 15 m; Fig. 3.40) comprised a water cistern. The entrance to this room, and possibly to the whole complex, was from the north, via a 1.80m wide opening. A small room (2 × 2 m) was located northwest of the room. Another water cistern (now covered with a metal grille) and a stone clearance heap placed next to it were discovered between the eastern and western complexes. The southern complex (c. 15 × 20 m) was poorly preserved. Scattered large- and medium-sized stones and a hewn stone basin (diam. 0.50 m) were found in its area.

The many architectural remains found on the top of a hill, above the southern bank of Naḥal Rabah, with numerous tesserae visible on the surface, attest to the presence of a monastery and/or a church. Three meticulously built and well-preserved complexes (eastern, western and southern; Fig. 3.39) were found, including some thresholds and doorjambs. The eastern complex consisted of nine rooms with two towers— one in its northeastern corner and the other in its northwestern corner. Parts of an oil press and a lintel (length 1.40 m) were discovered in the southwestern part of the complex. Remains of plaster, which seems contemporary, were also discerned. A rock-cut cistern (perhaps a small storage pool, 1.6 × 1.7 m, depth 1.1 m) was found east of the complex, with a hewn projection on its southern wall that served to support a roof. Remains of reddish plaster are still visible on its walls. A small channel ran from the west to the cistern, suggesting the presence, at some stage, of a winepress.

The Agricultural Plot H. Dayyar was bordered on its north, east and south by three gullies. The western boundary of the settlement generally followed to the 198 longitude line, an area currently underneath the eastern neighborhoods of the town of Rosh Ha‘ayin. Thus, the study of this area was based mainly on previous studies and aerial photographs. Winepresses. Two simple winepresses were discovered south of the site, one with a rectangular treading floor (1.2 × 1.9 m) and a collecting vat (0.40 × 0.60 m), and the other, c. 200 m to the south, with a treading floor and an irregular collecting vat.

The western complex had 11 rooms built of well-preserved, high-quality, large-sized stones (max. length 1.30 m). The complex was bordered by double

Fig. 3.39. H. Dayyar. Plan.

60

Amit Shadman: Chapter 3, The Sites

Fig. 3.40. H. Dayyar. The central room in the western complex, looking west (A. Shadman).

Field Towers. Four towers were discovered on the slope of a spur, south of the site, two rounded and two square (Kochavi and Beit Arieh 1994: Sites 75, 104, 106).

Guérin: “A completely destroyed village on a hill. It was no more than a small group of residential buildings built as square towers of medium-sized stones. A fragment of a pillar and many small white stones, remnants of an ancient mosaic floor, attest to the fact that there was a Christian church that was totally destroyed. The entire site is now covered with tobacco plants. A large rockhewn pool extends nearby. It was initially a quarry for ashlars, and later it was converted into a circular pool, approached by hewn steps.” (Guérin 1984: 94).

Quarries. A large quarry (18 × 32 m, depth 4.5 m) was found southwest of the site’s summit (map ref. 198360/667180). South of this quarry (map ref. 198350/667050), remains of a building were found, built of medium- and large-sized fieldstones and preserved to a maximum height of three courses (Kochavi and Beit Arieh 1994: Site 75). Summary

PEF:

H. Dayyar is located on a hilltop overlooking its surroundings, at the junction of Naḥal Rabah and Wadi al-Susi. Its location marks, in fact, the northern boundary of the study area. The site was seemingly occupied by a self-sufficient monastery since no evidence was found of a possible connection with another settlement. The agricultural sectors of the site were mainly on its southern and western sides, in farming-terraces areas. Since most of the site underwent modern development, and past studies dealt mainly with the building itself, it was almost impossible to study its agricultural plot.

“The third ruin (K p) called Khurbat ed Duwer is near Serta on the east. Traces of ruins only remain on the west, by a goat-fold, is an ancient watch-tower like those at Kurawa Ibn Hasan.” (Conder and Kitchener 1881-1883, II: 333). Ceramic finds: Byzantine, Mamluk, and Ottoman periods (18th–19th centuries).

61

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods

Numismatic finds: A bronze coin from the late Byzantine period (around 640 CE).

below). It should be noted that a few potsherds from earlier periods were also found, but these were only ‘field sherds’ which do not attest to an actual settlement. The third phase, possibly from the Early Islamic period, revealed a building founded on an earlier structure perhaps also dated to the same period (see Safrai 1995c: 112).

Previous studies: Conder and Kitchener 1882, II: 333; Guérin 1984: 94; Kochavi and Beit Arieh 1994: Site 75; Safrai 1995b; 1995c: 102; ‘Ad 2000; Bagatti 2002: 153–154. Site No. 9 H. Yeqavim  Kh. Umm el-Ikba Map reference: 199550/66410



A ‘rural town’ (100 × 150 m) in an alluvial valley, where remains of many buildings and rooms, most well preserved, were discerned, some rising to a height of five–six courses. Their outlines, however, were not always discernible (Fig. 3.41). The site was described at length by Kochavi and Beit Arieh (1994: Site 205) and by Safrai (1995b, 1995c) after their surveys. Three main construction phases were observed at the site, but only a full-scale archaeological excavation can corroborate this.

The residential quarter was divided into three sections, with a wide, open area (possibly a large courtyard) in between which probably served the daily needs of the residents. The northern section contained 12 rooms, most of them small, and an oil press (one of four found at the site) in its northwestern part. One of the numerous vaulted buildings, built of small stones and plaster (c. 20 × 20 m), seems to date to later than the Byzantine period. An r-shaped building (c. 8 × 13 m), adjacent to the vaulted building on the east, with red plaster on its northern wall, was entered via a builtin passage in its eastern wall. South of the northern section was an east–west oriented road leading to the cultivated areas located to the west.

Based on the evidence presented in Kochavi and Beit Arieh (1994), and on the results of the present study, the site seems to have been first settled during the Byzantine period (its first two phases; Fig. 3.45, and

The eastern section, southeast of the northern one, comprised 15 rooms, and oil presses. In a large building (18 × 18 m), preserved to a height of four courses, several water cisterns were discovered.

Fig. 3.41. H. Yeqavim. Aerial view (Sky View, A. Shadman).

62

Amit Shadman: Chapter 3, The Sites

The western section, c. 30 m west of the eastern one, comprised 25 rooms, some 20 of these belonging to a single large building (c. 18 × 30 m; Fig. 3.42) located in the southwestern part of the site. Two rows of rooms, accessed from the north, were found in the southern part of the building. North of these rooms, in the northeastern corner of the eastern courtyard, were two courtyards and a cell of an unclear nature. The largest space in the building was discerned north of the two courtyards, and an oil press and two water cisterns were recorded on its northeastern side. The outer southern wall of the building was well preserved and delimited a large plot (c. 40 dunams) which probably formed part of the entire complex.

Oil Press

North of the western complex was a built road (width c. 2 m) which led to a partially plastered underground cavity, probably a miqveh, descended by stairs. A road (perhaps a street) led westward toward a burial ground (see below) and then to the agricultural areas. The church. About 200 m north of the residential quarter was an elevated compound entered through a doorway located on the southwest, of which two well-hewn stone doorjambs and a threshold survived. A rectangular building (8 × 10 m), built of high-quality ashlars, probably a church, was discerned in the northern part of the compound. Tesserae were found scattered in the immediate vicinity of the building, and two simple rock-cut installations (one redplastered) and a square shaft were recorded next to its northeastern corner.

Fig. 3.42. H. Yeqavim. Plan of a large building on the southwest part of the site.

there. About 100 meters after exiting the plot, the road intersected with rural Road 1. Road 9b. This road connected the center of the site with the church to its north. The road left the site northeastward and 100 meters later intersected rural Road 3. The road continued a few dozen meters to the north and then turned toward an enclosed plot southeast of the church. A limekiln was built here at a later stage.

The Agricultural Plot The site appears to have owned large agricultural areas, as demonstrated by the agricultural roads leading to them (see below). South and east of the site were enclosed plots with well-built walls, and a large terraced area was to its west. A large boundary stone, which probably marked the site’s western border, was discovered c. 500 m west of the settlement. The agricultural area of the site totaled c. 2863 dunams.

Enclosed agricultural plots (Fig. 3.43). The plots are located in the eastern part of the site, bordering, at some point, the cultivated areas of Deir Balut. Thirtyfive plots were discovered, totaling 443 dunams. Generally, the plots on the eastern side of the site were larger than those on the west. Plot 2, the largest plot, extended on an area of over 80 dunams, and it was possibly subdivided.

Agricultural roads. Two agricultural roads (9a and 9b, see map in Chapter 6; Fig. 6.4) led to the agricultural areas. These roads intersected with two rural roads, Nos. 1 and 3 (see Chapter 6 below), which approached the area from the west and the north.

Plot 16, south of the large building in the southwest area of the site (see above), was possibly owned by a landlord who dwelled in the building. Plots 3–9, northeast of the site, shared some similar characteristics and were carefully measured, arriving at a total area extending over 16 to 20 dunams. Plots 21–24, west of the site, were relatively small (3.5–5.5

Road 9a. This road ran northwest–southwest and connected rural Roads 1 and 3. The road began at the junction with rural Road 3 and ran southwest into the site, going out on its eastern side. At map reference 199564/663994, the road reached a large plot that seems to have belonged to a large building located 63

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods

dunams each). It seems, therefore, that the distribution of the land among the inhabitants of the site was thoughtfully done, reflecting a social–economic hierarchy among the landlords. Installations, such as field towers and winepresses, were found in the various plots, but no consistent pattern was discerned.

church, where vineyards were probably grown. Grain was presumably grown in a large area characterized by fertile alluvial soil accumulated in farming terraces, west of and adjacent to the site. The winepresses and the field towers were discovered alongside rural Road 4, attesting to the presence of vineyards.

Location of agricultural crops. The agricultural plots were generally located west and southwest of the site, while to the east and south were rocky areas with shallow soil. The best-quality area, adjacent to the rural Road 1, which ran west of the site, comprised particularly wide terraces and deeply accumulated soil. Further west, along another rural Road (No. 4), was a wide terraced area with shallow soil.

Winepresses. Surprisingly, only four winepresses were discerned at the site, although others may not have been identified in the various studies. One winepress was found in the southern part of the site, but most of it was covered with alluvium, making it difficult to draw its plan. A second winepress with a rectangular collecting vat (0.80 × 1.10 m, depth 0.30 m) was discovered west of the settlement and north of rural Road 4 (Kochavi and Beit Arieh 1994: Site 190). A third, simple winepress was found a few hundred meters northwest of the site (Kochavi and Beit Arieh 1994: Site 197). West of the church, on a bedrock terrace, was the fourth winepress, with a treading floor, a settling pit, and a collecting vat. In a later phase, a cistern was hewn in the collecting vat.

The agricultural crops can thus be distributed as follows: olive groves were probably cultivated in the areas east and south of the site—olive groves belonging to Deir Balut are grown east of the site even at present. Winepresses and field towers were hardly identified in this area. A large winepress and some field towers were found north of the settlement, in the vicinity of the

Fig. 3.43. H. Yeqavim. Land distribution.

64

Amit Shadman: Chapter 3, The Sites

Table 3.2. Plot No.

Area in dunams

1

23.6

2

82.3

3

16.2

4

20.2

5

18.1

6

17.3

7

18.1

8

17.5

9

17.3

10

6.1

11

12.7

12

13.0

13

14.3

14

4.8

15

15.3

16

38.2

17

11.4

18

6.2

19

6.6

20

4.7

21

5.4

22

4.5

23

4.2

24

5.7

25

3.6

the site and north and south of rural Road 4. In certain cases, the terraces were covered by later dividing walls. Dammed wadis. Two dammed wadis were recorded near the site. One was east of the site, where Road 7 passed through connecting H. Yeqavim and Kh. Umm al-Hammam. The wadi was 800 m long and 100 m wide and comprised 40 dams, i.e., one dam every 20 m. Another 1-km long wadi, northwest of the site and originating to its west, included quite long dams that shortened down the gully on the west. Tombs. Two grave clusters were discovered, including a mausoleum. The northern, smaller cluster, was located west of the site and northwest of the church, while the other, larger one, was on a rocky hill c. 150 m west of the site. The northern cluster included several hewn tombs, some were arcosolia caves clearly looted in the past. One tomb had an entrance with a hewn vault facing east to the church. Half of another grave collapsed eastward. The second cluster included four cist tombs which seemed to form part of a mausoleum. Three of them were parallel to each other, while the fourth, the smallest of the four, was perpendicular to them (0.65 × 2.10 m; 0.57 × 2 m; 0.80 × 1.80; 0.50 × 1.30 m). Fragments of a large sarcophagus lid (length 2.5 m; Fig. 3.44), bearing a cross in relief, were discovered near these tombs to the north. East of the mausoleum and next to it were a circular hewn installation (possibly a small reservoir), a large lintel (0.70 × 1.90 m), and a stone threshold (0.70 × 1.20 m) with two projections. Additional tombs were discovered west of the mausoleum. Quarries. Two building-stones quarries were associated with the site. One, rather small, was located in the site’s northeastern part. The second, larger one, was discovered west of the mausoleum (see above).

Field Towers. Fifteen field towers were discovered in the site’s surroundings, most of them square-shaped. Three towers were discovered north of the site alongside rural Road 3, while the rest were found along the course of rural Road 4. A wall, leaning on one of the towers, created a space serving perhaps for storage or as an animal pen. A building (5 × 10 m), with two levels and an entrance in its southern wall, was either a tower or a small farmhouse. It survived to a height of five courses and was later used as the platform of a stone clearance heap.

Summary Horvat Yeqavim was occupied mainly during the Byzantine period, serving as an important center within the region’s rural system. It comprised about 20 buildings, each averaging 10 × 15 m in size. Of particular interest was the large building with numerous rooms found in the southwest part of the site, apparently occupied by a wealthy person, perhaps the head of the village.

Water cisterns. Fourteen cisterns were discerned among the many presumably present at the site. Most cisterns were incorporated in the buildings, but several were discovered some distance afar.

Three important rural roads (Nos. 1, 3, 4; see Chapter 6) crossed the area of the site and connected it with other sites in its vicinity; particularly significant was the road that connected the site with Migdal Afeq, the largest site in the area. Horvat Yeqavim was surrounded by many agricultural areas, and it was clear that a pre-planned allocation of the land was

Agricultural terraces. A large, c. 1-km long and c. 500-m wide terraced area, with accumulated alluvium eroded from the hills to its north, was located west of 65

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods

Fig. 3.44. H. Yeqavim. Fragments of a sarcophagus lid, looking north (A. Shadman).

practiced at the site. Many plots were approached by agricultural roads often connecting to rural ones. The main agricultural crops were olive and vine.

entrances observed in some - one with lintel. South wall (1 m wide) preserved to height of 2 m and more. (2) Structure (8 × 13.5 m) adjoined by two small rooms. (3) Long structure (5 ×14 m) on north-south axis, divided into three units. In northern (smallest) unit—pit. Wall 14 m long extends from southwest of structure, enclosing, with two walls (0.6 m wide), two rooms (4 × 5.5, 4.5 × 6.3 m) 5.5 m away from structure. (4) Remains of structure divided into three units (4 × 6.5, 3.8 × 5, 5 × 2.5 m; walls 0.7 m wide). Southwest unit survives; jambs (1.2 m high) flank east facing entrance. (5) Square structure; two plastered walls (2 m high), set diagonally to eastern side, with comer niches. To south—pit (roofed?) and remains of oil-press: crushing floor and stone. (6) Rectangular structure (1.5 × 2.8 m): walls are particularly thick with comers of north wall thickened further. Ashlar masonry in southwest comer; one stone has marginal drafting and square-cut hole. (7) Rock-hewn installation (roofed?) containing square pit with L-shaped opening, connected to large pit (cave?). To west—additional pit, covered by stone. To its north–kokh, hewn on two levels. (8) Remains of installation (oil-press?) on hill. Some 130 m to west of the site—mausoleum: four tombs, sharing hewn facade, adjoining rockhewn pit. Nearby—three column bases and rolling stone fashioned as sarcophagus lid. Some 10 m to west—pair of tombs and hewn entrance to pit. On east side of hill—two long rooms.” (Kochavi and Beit Arieh 1994: site 205).

Four oil presses were discerned at the site, all of them incorporated in the site’s buildings. The winepresses, in contrast, were outside the boundaries of the site, suggesting that the wine industry was less significant than oil production. The church, built high above the settlement, points to the identity of the site’s population that was associated with the church. Guérin (1984: 96) reported a building which included a rectangular chapel and the meager remains of an apse built on the hill. It is not clear whether he referred to the structure under discussion, but since the site was built in the plain and the church on a hill, it would seem Guérin described the same building. The grave clusters and the cross relief also support the existence of a church. The underground cavity with its staircase could have been a typical water reservoir, but it also could have served as a miqveh—a rarely occurring facility in the study area. In the Ottoman period, the site’s building stones were used as an available material for producing lime in the kiln. The vaulted structure may also belong to this period. Kochavi and Beit Arieh: “Ruin (approx. 100 × 100 m) including remains of structures and installations—most well-preserved—in small valley northeast of Nahal Shiloh. Principal remains: (1) Large structure (35 m long) containing several rooms; 66

Amit Shadman: Chapter 3, The Sites

2 1

3

4

5 0

10

Fig. 3.45. H. Yeqavim. Ceramic finds from the church area. Byzantine period: bowls (1–4); Early Islamic period: a jar (5).

Guérin:

apparently to some cistern beneath. Near to this is well and a Matruf roller… There is also a rock-cut birkeh, 7 feet by 5 feet deep; there are also caves, and a number of Metamir or caves sunk in ground for storing corn. This point to the place being held sacred.” (Conder and Kitchener 1881-1883, II: 357).

“... The remains of a fence partially built of neatly aligned ashlars and partially of bossed stones. The length of the eastern front was 34 steps, the western one was 34 steps, the northern front was 28 steps and the southern one was 45. It is trapezoidal, in accordance with the topography of the hill. In several places, the lower courses are laid on the leveled bedrock. There are two rock-hewn pits, the walls of a number of buildings that were completely destroyed, including a rectangular chapel measuring 20 × 9 steps, oriented from west to east. It ends in an apse whose blurred outlines can be still seen; it was paved with white mosaic, most of which looted. To the south of this wall, I saw another smaller one but arranged in great harmony. A few lower courses survived. A small cell is at some distance to the west. The other site described here is named after this small half-destroyed shrine, built of fine ashlars that had been taken from an earlier building. South of it I saw four water cisterns and a pool, twelve steps wide and six in width, partially carved and partially built in the rock.” (Guérin 1984: 96). 

Ceramic finds from the settlement area: Roman period (1st–2nd centuries CE)—a Terra-Sigillata bowl base; Byzantine period (5th–7th centuries CE)—kraters; Early Islamic, Crusader, Mamluk and Ottoman periods. Ceramic finds from the previous survey: “From the Byzantine period (including an imported LRC bowl) and from the Early Arab period.” (Kochavi and Beit Arieh 1994: Site 205). Previous studies: Conder and Kitchener 1882, II: 357; Guerin 1984: 96; Kochavi and Beit Arieh 1994: Site 205; Safrai 1995b: 189–208; 1995c: 111–112. Site No. 103 Wadi al-Bureid (south) Map reference: 199400/665000

PEF: 

A farmhouse (c. 30 × 30 m; Fig. 3.46) southwest of a gully descending towards Wadi al-Bureid. Some parts of the building were constructed with large, well-hewn stones (some 1.2 m long, or more). The building comprises eight rooms and two courtyards, arranged on an upper southern level and a lower northern level. The southern rooms (Nos. 1–7) were built of fairly high-quality fieldstones and ashlars and were probably used for residential purposes. Rooms Nos. 1–3 were especially large (7 × 10 m, 10 × 10 m, 10 × 12 m, respectively) and were seemingly entered from the courtyard, although

“Seems to be a small ruined village. There are foundations of building about 40 feet square, of stones some 2 feet in length. On the south-west, a door with a lintel stone, 6 feet long. An enclosure of late date is built on to the building on this side, the wall having a straight joint. Ruined walls and houses of fair-sized masonry exist lower down the hill on the west. In the building above noticed there are cornerstones with boss, 4 feet by 2 feet by 2 feet. A small ruined Kubbeh stands apart lower down the hill. It has a rubble roof of groined arches, supporting a little dome; the walls are of small ashlar , with rubble on the inside, cemented over. An earthenware drain pipe from the roof descends

3  The agricultural plot is discussed with Kh. Umm al-Hammam (Site No. 11).

67

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods

Summary

no doorways were discerned. Two smaller rooms (Nos. 6 and 7), were discovered west of these rooms.

The location of this building within the agricultural plot of Kh. Umm al-Hammam (Site No. 11) suggests that the two sites maintained close relations. The building, characterized by high-quality construction and a strategic location on a hill overlooking Kh. Umm al-Hammam, was likely operated in conjunction with the large neighboring settlement. Accordingly, it is plausible that a wealthy person dwelled here, perhaps even the village master of Kh. Umm al-Hammam. Similar buildings were discovered in the study area, some identified as churches or monasteries, but in the absence of typical finds such as roof tiles or crosses, the building at the present site did seemingly not serve religious purposes.

In the southwestern corner of the building, in the western wall of an entrance room (Room No. 5, 5 × 20 m), a doorway with two dressed stone doorjambs was recorded. Room No. 8 (5 × 7 m), in the lower northern level, accommodated an oil press. A lower crushing stone (diam. 2 m; Figs. 3.47, 3.48) of this oil press, and two piers (height 1.2 m), which supported the wooden beam, were found. South of the oil press, a corridor ran along the western wall of the building and ended in Room No. 5 (see above). A wall found in this area, built of small-sized fieldstones and white plaster, seems to be a later addition. Between the residential rooms and the oil press was the northern courtyard, used for daily life, which included a cistern. To its east, the eastern courtyard accommodated a hewn basin (0.50 × 1.1 m, depth 0.30 m) presumably used for animals. A limekiln, of a later period, was found adjacent and east of this courtyard. South of the building, a hewn cist grave was revealed. An open cavity (possibly a damaged cistern) was traced on a stepped bedrock area north of the farm, and further north was a rock-cut surface, perhaps a winepress treading floor.

Ceramic finds: Byzantine and Early Islamic periods. Previous Studies: Kochavi and Beit Arieh 1994: Site 159; Safrai 1995b: 193; 1995c; 1997a: 220–221.

Fig. 3.46. Wadi al-Bureid (south). Areal view (M. Haiman and A. Shadman).

68

Amit Shadman: Chapter 3, The Sites

Fig. 3.47. Wadi al-Bureid (south), looking southeast (A. Shadman).

Fig. 3.48. Wadi al-Bureid (south). An oil press lower crushing stone (A. Shadman).

Site No. 11 Kh. Umm al-Hammam Map reference: 199720/665300

The buildings were arranged around three courtyards: a large courtyard (c. 30 × 40 m) with a cistern in the southwest; a small courtyard (c. 20 × 20 m) in the northeast; and a courtyard in the southeast (15 × 20 m), probably used by the adjacent oil press (see below). Five entrances (1–5) led into the settlement. The two western ones led to the large courtyard: Entrance 1 was a corridor (length 20 m, width 2 m) that led to the large courtyard, while Entrance 2 was a wide passage (2.40 m) between two buildings, with a massive wall (1.80 m wide) to its south. Entrances 3 and 4 were located in the southern part of the settlement: Entrance 3 also led to the large courtyard, while Entrance 4 led directly to a large oil press in the southeastern part of the settlement.

A ‘rural town’ (100 × 100 m) on a gentle slope, c. 200 m south of Wadi al-Bureid. This was a large settlement with many buildings and dozens of rooms (Fig. 3.49). The buildings, preserved to a minimum height of four courses, were built of ashlars and fieldstones, some of them plastered. Safrai (1997a: 217) claimed that this was a two-strata site composed of about 19 buildings and about 60–90 rooms. The buildings had collapsed and the stone debris allowed for only a general view during the present survey. 69

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods

Entrance 5, in the northeastern part of the settlement, led into a narrow corridor (perhaps a room) that extended southward toward the small courtyard. A water cistern was discerned in this courtyard, and an underground cavity, west of this cistern, was probably another cistern whose ceiling had collapsed. Remains of an oil press were discovered in the western part of this courtyard. Three complexes were built around this small courtyard. Complex A (c. 20 × 40 m), comprising 14 rooms, was discerned to its northwest; its largest room (A1; 12 × 15 m) was in its northern part, and a cistern was east of this room. Complex B (20 × 30 m), east of the small courtyard, consisted of six rooms (5 × 7 m). This complex was entered from the north via Entrance 5, where two water cisterns were traced. 

one of the main entrances to the entire settlement, and the southern building in Complex D may have been associated with it. Complex E (c. 25 × 35 m) was located southwest of the large courtyard and comprised 11 rooms. Five rooms (5 × 5 m), aligned east–west, were discerned in its southern part, and one of them comprised a hewn cistern. This complex is similar to the structure at H. Yeqavim (Site No. 9), located only 1.5 km away. Several oil presses were found at the site, one of them in a courtyard located in its southeastern part (see above). The courtyard was surrounded by eight rooms which possibly served the oil press. Two large piers for anchoring a beam and a lower crushing stone (diam. 1.80 m, depth 0.30 m) survived. Two or three additional oil presses were discovered in other parts of the site, with piers for anchoring the beams next to them (height c. 2 m; Safrai 1997a: 220; Buildings 14, 18). Two arcosolia tombs were discovered in the western part of the site, adjacent to a plastered cavity with steps descending into it.

South of the small courtyard was complex C, consisting of six rooms arranged around a square courtyard (10 × 10 m) accessed from the northwest. Complex D was located northwest of the large courtyard and comprised 12 rooms, and a cistern in the complex’s center. To its south, Entrance 1 was probably

Fig. 3.49. Kh. Umm al-Hammam. Plan.

70

Amit Shadman: Chapter 3, The Sites

Location of the agricultural crops. Kh. Umm alHammam owned only limited alluvial lands. To increase the agricultural areas of the site, the local residents dammed several wadis in its vicinity and prepared other parts for cultivation. Two wide dammed wadis, south of the site, had fertile accumulated alluvium. North of the site was a gentle slope with wide farming terraces. The slope west of rural Road 7 was mild and rocky and was apparently used for vineyards and olive groves. This area, comprising an oil press and a winepress, was probably owned by the estate (Site No. 10). The rest of the site’s agricultural plot was rocky, with shallow soil. Thus, most of the agricultural products of the site were seemingly grains and olives.

  A few isolated structures were discerned in the immediate vicinity of Kh. Umm al-Hammam, probably associated with the settlement. One of them (6 × 6 m) was situated on a higher spot southeast of the site. The building, possibly a church, survived to a height of two courses and was built of high-quality ashlars. Guérin (1984, see below) referred to a small church above the site, perhaps relating to this building. The entrance to the building was from the west through steps extending between two well-dressed stones, where numerous tesserae and a cistern were visible. East of this building was a square courtyard which seems to be a later addition. A late plastered vaulted structure was discerned between this building and the site, and another late structure was on a hill, c. 100 m above the ancient Byzantine settlement, perhaps a road station from the Mamluk or the Ottoman periods (Fig. 3.50).

Wheat was grown in the dammed wadis and the gentle slopes, as evidenced by the several threshing floors discovered in the eastern part of the site. In addition, an extensive area of agricultural terraces, east of the site, included almost no winepresses and field towers. This may indicate that the area was designated mainly for olive groves. The winepress discovered next to the church indicate that vines were also grown, although apparently in small scale. North of the settlement, in the vicinity of the farmhouse (Site No. 6), was a rocky area that was not suitable for agriculture. The presence of an oil press in this area attests, perhaps, to the main crop there. Further north, was a particularly steep slope descending to Wadi al-Bureid, where no agricultural features were identified, perhaps used for grazing. The number of threshing floors around Kh. Umm al-Hammam, large in relation to the areas suitable for cultivating grain, may indicate that the residents of the site provided threshing services to the surrounding communities.

The Agricultural Plot The northern boundary of Kh. Umm al-Hammam was in Wadi al-Buraid, bounded on the north by the plots of Kh. Kesfa (Site No. 13). East of the site was the road that led to Deir Balut (see Chapter 6, Road No. 9), and Kh. Umm al-Hammam and its agricultural plots did seemingly not extend beyond this road. On the south, the site was probably bounded by the plots of H. Yeqavim (central map reference 199800/664700). The main rural road in the area (Road 1) ran west of the site, and it seems to have served as the boundary between Kh. Umm al-Hammam and H. Te’ena. A prominent large border stone was found by the road. Unlike other sites, the agricultural areas of Kh. Umm al-Hammam were not characterized by enclosed plots. Most of the plots were located near the site, in a relatively large and topographically convenient area, and were characterized by dammed wadis and terraced slopes. The total cultivated area of Kh. Umm al-Hammam was c. 2062 dunams.

Winepresses. Only two winepresses were discovered in the agricultural plot. One was discerned north of the church and included a treading surface (3 × 4 m; an unclear floor) and a collecting vat. The two circular columns lying on their sides south of and above the treading surface may have been used for secondary extracting of the grapes’ skin. North of the treading surface was the collecting vat (1.80 × 2.30 m, depth 1.70; capacity of 7000 liters). The vat, with hewn steps descending into it in its northwestern corner, was coated with reddish plaster. The second winepress was discovered c. 300 m southeast of the site (Kochavi and Beit Arieh 1994: Site 195).

Agricultural roads. Three agricultural roads were used by the farmers of this site (see Chapter 6, Fig. 6.4): Road 11a. This road connected the site with a farm to its northwest (Site No. 6). It is only 200 m long and, in some places, its margins were used as retaining walls.  Road 11b. This road began east of the site, at a junction with rural Road 6. It turned southwest, to an agricultural terraced area descending eastward. The road ends about 400 m south of the site and perhaps connected with rural Road 7.

Field towers. A square field tower was discovered northeast of the church, in an extensive agricultural area. Two other towers were located west of the site, above the western bank of a gully, but these seem to be later structures.

Road 11c. This road apparently led to the church, southeast of the site. Near the church, the road turned eastward towards a large area of agricultural terraces and enclosed plots. It was about 150 m long and ended in Road 11b. 

Water cisterns. In addition to the numerous cisterns discerned inside the settlement, several cisterns were 71

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods

found in buildings such as a church, a farmhouse, and an estate, all located outside the settlement. The cisterns in these buildings were described in Sites Nos. 6 and 10.

The second dammed wadi (width c. 100 m, lengthc. 500 m), comprising a dam every 16 m, was on a gentle spur located south of the settlement which started northeast of the church. This wadi also ended in the same tributary of Wadi al-Buried mentioned above.

Reservoirs. An irregularly shaped reservoir was found north of the church, with a wall (probably of a later period, built by local shepherds) abutting it on the southeast. A smaller irregular reservoir (3 × 3 m, depth c. 1.5 m) was discovered north of the site, between two farm buildings of Site No. 6 which formed, in fact, an integral part of the settlement.

About 250 m south of the church, was the third dammed wadi (100 m wide; 400 m long), which began where a rural road leading from Deir Balut to Kh. Umm al-Hammam passed. The fourth wadi was south of the settlement, at the boundary between Kh. Umm al-Hammam and H. Yeqavim. The fifth wadi, located southwest of the settlement, began in rural Road 1, on a flat area that descends eastward into a gully c. 100–200 m wide and c. 500 m long. Later limekilns were discerned on both ends of this wadi.

Agricultural terraces. Most of the agricultural activity of the site took place in terraced areas. These terraces were built almost in every agricultural area. Most terraces were discerned east and southeast of the site and were characterized by gentle ridges descending eastward.

Tombs. Two burial areas were discovered, both related to this settlement: One in the western part of the settlement, and the other adjacent to the site of Wadi al-Bureid (South). Most tombs were arcosolia caves.

Dammed wadis. Five dammed wadis were associated with this settlement, and most of the site’s agricultural activity was probably carried out in these. The northernmost wadi, between the site of Wadi alBureid (East) and the central complex of Kh. Umm alHammam, was a wide gully (c. 250 m) that descended westward and ended in a tributary of Wadi al-Bureid. An agricultural road was found in the lower section of the wadi’s bank.

Quarries. Remains of a quarry were discovered west of the church, where natural bedrock terraces were observed. This is a small quarry that does not correspond to the size of the settlement and the vast number of buildings there.

Fig. 3.50. Kh. Umm al-Hammam. A Mamluk/Ottoman structure (M. Haiman).

72

Amit Shadman: Chapter 3, The Sites

Fig. 3.51. Kh. Umm al-Hammam. Threshing floors and sheep pens (Sky View, A. Shadman).

Threshing floors. Five stone circles were found east and south of the settlement. The circles were meticulously built of large-sized fieldstones, and some were founded on a bedrock terrace. Two of the measured circles were uniform in diameter (12 m) and in the manner of their construction, i.e., with a wall of one or two courses built of medium- to large-sized stones. The location of the circles took into account the prevailing winds, and their construction indicates these served as threshing floors (Fig. 3.51).

pre-planned, and the passages that led to the builtup sections were not planned streets. At least four oil presses were incorporated in the settlement, occupying relatively large areas within the complexes. In most cases, the remains of these presses were lower crushing stones, and pillars (PIERS?) for anchoring a beam. Three additional oil presses were discovered in buildings on the margins of the settlement (Sites Nos. 6 and 10). Based on this data, the oil industry seems to have been an important and perhaps central sector of the settlement’s economy, while only two winepresses were recorded both within the site and its cultivated areas, as well as a meager number of field towers. It is possible that the preference for oil products resulted from the local topography and the quality of the soil. Most of the cultivated area in the eastern part of the settlement was rocky and of poor quality. Farming terraces were seemingly constructed in this area during the Byzantine period to improve the agricultural potential of the area. In addition, dams were built in every available wadi and valley, increasing the cultivated area to its maximum extent and enabling growing grains that required deeper fertile soil. The presence of a high number of threshing floors, compared to other neighboring communities, attests to the importance of the grains. The various sites around Kh. Umm al-Hammam, such as a farm (Site No. 6) where an oil press was recorded, and an estate (Site No.

Sheep pens. Two square pens were discovered in the eastern part of the settlement (Fig. 3.51). One was a rectangular enclosure located at the foot of a natural bedrock terrace. It was built of small- and mediumsized stones and survived to a height of c. 1 m. The second pen was adjacent to the walls of a later building (undated) southeast of the settlement. The pen survived to a height of c. 0.5 m and it was constructed similarly to the first pen. Presumably, these pens were of a later date than the settlement. Summary The large Byzantine-period rural settlement of Kh. Umm al-Hammam controlled large agricultural areas in its environs (c. 2062 dunams). The layout of the buildings within the settlement was not uniform or 73

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods

Previous studies: Conder and Kitchener 1882, II: 356–357; Guérin 1984: 93; Kochavi and Beit Arieh 1994: Site 158; Safrai 1995b; 1995c: 107–109; 1997a: 217–222; 1997b; 1998; Tepper 1997: 229–274; Magen 2008b: 279; Bagatti 2002: 154–155.

10) with high-quality building and oil press, attest to the centrality of the settlement in Kh. Umm al-Hammam. Guérin: “... a town whose site is now disrupted by the dense vegetation of mastic trees (Pistacia lentiscus), among which lower courses of many buildings are seen or buried. Most of the buildings were usually built of relatively large undressed stones, which had collapsed. Here and there, the mouths of water cisterns are seen among these ruins, beneath the feet of a wandering person. A vaulted building, Arab in nature, is partially preserved outside the site. The name of the site, Kh. al-Hammam comes from the name of this building (alHammam – the bath), while the original name of this site was completely forgotten. On a nearby hill, located higher than the town and overlooking it, I saw the remains of a small church, oriented west–east; twenty steps long and sixteen steps wide. Despite its modest dimensions, it seems to have three halls and several stone columns lying on the ground, which are probably the remains of the partitions between the nave and the aisles. People would not enter this religious building from the west but from the north, through three rectangular gates with monolithic lintels mounted on their posts. The houses adjacent to the church, which had been completely destroyed, probably belonged to an ancient monastery where several hewn water cisterns were preserved both on top of the hill and its slope. The hill was surrounded by a wall, mostly collapsed.” (Guérin 1984: 93).

Site No. 12 Naḥal Shiloh (Southwest) Map reference: 198700/664400  A small square-shaped fortress (22 × 22 m, external walls 1.40 m wide) on a high hill located between the tributaries of Naḥal Shiloh (Fig. 3.52). The site overlooked the area, and especially the rural road that connected Migdal Afeq to H. Yeqavim. The fortress comprised a square courtyard with rooms on its southern and eastern sides (Fig. 3.53). The entrance was via a built opening (1.40 m wide) in the eastern wall. An external room, adjacent to the south of the entrance, was possibly a later addition. Another room, covered with a large stone (width 0.80 m, height 1 m), was discerned north of the entrance. There may have been another entrance to the fortress, blocked at a later stage. In the southwestern corner of the fortress, on a higher level, was a room (6 × 9 m) built of partially hewn large-sized stones (1 m or more) and preserved to a height of five courses (c. 1.5 m), with retaining walls built against the western and southern steep slopes. Based on the vast quantity of stone debris, this room was seemingly a tower overlooking the entire area. East and adjacent to the tower were several rooms covered with stone collapse. These rooms (possibly a residential area) were built of medium- to large-sized, partially hewn, fieldstones. Another east-west wall, abutting the middle of the western wall of the fortress, was probably built to support the tower. A second building, probably a modern sheep pen, was discerned southwest of the fortress. A rock-cut square, possibly the opening of a cistern, was recorded between these two structures. A winepress collecting vat (1.0 × 1.1 m) was discerned southeast and near the fortress. The fortress was surrounded on the north and west by a poorly preserved stone fence, and north of the latter was a massive agricultural terrace that provided a suitable surface for cultivation. On the southern slopes of the hill, a winepress with an irregular treading floor and a square-shaped collecting vat were located. East of the winepress was a water cistern, with its capstone, and a stone trough beside it. An agricultural road was discovered next to the cistern, apparently extending from the rural road leading from Migdal Afeq to H. Yeqavim. Another sheep pen, crescent-shaped and leaning on a natural rock step, was found on the northern slopes of the hill.

PEF: “A ruin with a Hummam or birkeh; between it and Kefr Kasim is a tomb- a rock-sunk loculus only. The birkeh measures about 50 feet by 75 feet. It is of masonry, of good size; one of the corner-stones is 4 feet 9 inches in height; with several others it has two bosses; the marginal draft is about I inch broad, the bosses rough, projecting from I inch to 6 inches. On the south interior there is rubble and good white mortar. Close by lies a fragment of a pillar, 2 feet diameter of limestone, with a double fillet. On the east is a second birkeh, 20 feet by 12 feet. It ia of rock, and lined with good cement, but irregularly cut; there are places at the side to receive the haunch stones of an arch; it is about 7 feet deep. A millstone, 6 feet diameter, lies near and there is ruined tower of small masonry, apparently a modern gardenhouse.” (Conder and Kitchener 1882, II: 356–357). Ceramic finds: From the area of the church— Byzantine (mainly), Early Islamic and Ottoman periods. From the settlement—Byzantine period (mostly the 6th–7th centuries CE; cooking pots and a lamp with an inscription); Early Islamic (8th century CE) and late Ottoman periods. 74

Amit Shadman: Chapter 3, The Sites

Fig. 3.52. Naḥal Shiloh (southwest). Aerial view (Sky View, A. Shadman).

Retaining wall

Rooms

Entrance

Tower(?)

Fig. 3.53. Naḥal Shiloh (southwest). Plan (M. Haiman and A. Shadman).

Summary

location. The fortress was on a small steep hill which limited its expansion and had dammed wadis at its foothill. The agricultural installations indicate that the local inhabitants engaged in wine production and grazing. An environmental computerized study

This fortress was unique within the regional settlement pattern. No other structure or settlement in the study area were like it, be it in nature or topographical 75

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods

Building 2. A square building (c. 4 × 5 m), c. 10 m south of Building 1. A water cistern was discerned near its southern end, and a wall stretched westward toward a large winepress (see below).

showed that this summit overlooked the neighboring communities, and particularly the roads that crossed the area. Hence, it is plausible it was occupied by a regional garrison detachment during the Roman period (see Chapter 7, the Roman period). Ceramic finds: Iron Age II; the Persian, Roman, Byzantine, and Early Islamic periods.

Building 3. Meager remains of a large building (c. 12 × 15 m) forming part of a complex of buildings in the southern part of the settlement.

Previous Studies: Finkelstein 1978a: 30, Table 13; Kochavi and Beit Arieh 1994: Site 189.

Building 4. A small building (5.2 × 6.0 m) that probably served the residents of buildings 3 and 5 (see below).

Site No. 13 Kh. Kesfa Map reference: 199990/666700

Building 5. Located north of Building 4, and later than it, this building had entrances in its northeastern and southeastern corners. Its northern side was shared with Building 6 (see below).

A ‘rural town’ (c. 100 × 150 m) on the northern end of a plain located southeast of a gully descending to Naḥal Rabah. Twenty buildings and many rooms built of ashlars (some with drafted margins) and fieldstones were found, as well as installations attesting to extensive oil and wine production. A cross was discerned in one of the rooms (Safrai and Fixsler 1993: Building 30), pointing to a Byzantine Christian population. The settlement plan (Safrai and Fixsler 1993: 204; Fixsler and Silman 2000: 203–211) indicates that it was not pre-planned; no streets, built-in passages, or public buildings were discovered.

Building 6. Similar to Building 5, it apparently comprised two rooms and an entrance in its western wall. Building 7. A rectangular elongated structure (5.0 × 15.6 m) with an additional room abutting it. The entrance was through an open courtyard to its east. Buildings 8–10. A pair of large buildings and a small additional structure on the eastern side of the large central courtyard (c. 25 × 30 m) of the settlement. Building 8 was a rectangular room (5 × 14 m) whose entrance was probably in its northern wall. To its north was Building 9 (14 × 12 m). The smaller Building 10 (4.1 × 8.3 m) was west of Buildings 8 and 9; although Building 10 shared its walls with Buildings 8 and 9, no passage was identified between them.

A large courtyard (see below), which may be associated with a public building, was identified at the center of the settlement. Three phases were observed at the site:4 [1] the early phase—a Byzantine (Fig. 3.59) settlement that included residential buildings, together with various installations such as enclosed plots, roads and tombs (see below); [2] The intermediate phase—a Byzantine settlement with buildings and installations (including oil presses, see below) founded on the early phase; [3] The late phase—a settlement founded on the Byzantine remains, damaging them, probably dated to the late Arab period or later. An ashlar building was preserved from this phase.

Buildings 11–12. Two buildings south of the central courtyard. Building 11 (7.2 × 8.0 m) was entered from the east. Building 12 (11.7 × 13.0 m) was east of the courtyard, but its entrance was not located. Buildings 13–15. An r-shaped complex. Building 13 was a large structure (12.6 × 28 m) found in the eastern part of the complex. A narrow corridor and Building 14 were to its west. The southern wall of these two buildings is actually the northern wall of the central courtyard. The passage between them was, perhaps, in the southwestern corner of the wall. Building 15 (8 × 18 m) was located west of Buildings 13 and 14. A cistern was found west of Building 15.

The buildings found in the site are the following (Figs. 3.54, 3.55): Building 1. On the western side of the settlement, this building had two rooms: the eastern one (3.5 × 4.8 m), probably used for dwelling, with an entrance in its northwestern corner; and the long western room to its west (3 × 9 m). No passage was discerned between them. Two courtyards (possibly for animals) were found adjacent to the building, one west of the long western room, and the other south of the complex.

Buildings 16–18. A row of buildings on the hillside northeast of the site, high above a gully that descended towards Naḥal Rabah. A pair of rooms in Building 16 was apparently entered from the south. Building 17 (12 × 40 m) comprised several rooms and a courtyard (?). Further east, Building 18 had three rooms. Four water cisterns were discovered southeast of these structures.

4 

These phases were defined by Safrai and Fixsler (1993). The oil presses were excavated by Magen (2008b).

76

Amit Shadman: Chapter 3, The Sites

Fig. 3.54. Kh. Kesfa. Plan (Safrai and Fixsler 1993).

Building 19. The easternmost structure (10 × 10 m) of the site, this structure was apparently not part of the residential area. No internal partition walls were discerned, but these may have not survived. A water cistern was located to its west.

pit (1.0 × 1.3 m), west of the treading floor, was coated with reddish plaster containing potsherds. To the west and south of the settling pit was an r-shaped collecting vat (c. 2 × 4 m). At a later stage, the site was converted into a quarry.

Building 20. An elongated rectangular building (3.0 × 10.5 m) comprising two rooms and located east of the central courtyard.

A stable dated to the Byzantine period or later (Tepper 1997: 261–262) was recorded within the limits of the settlement. A double field wall (width 1.5 m), oriented north–south, extended south of the settlement’s central complex, and the space between its two sides seems to have been used as a passage between plots. A built rural road was discovered north of the cultivated areas (see Chapter 6, Roads). South of this road was a square building (5 × 5 m) built of small-sized fieldstones with white plaster. Four hewn burial caves, two of these apparently arcosolia caves, were identified south and southeast of Building 21. A particularly large stone, probably a boundary stone, was found west of the caves.

Buildings 21–25 (oil presses). Five oil presses were discovered at the site, four of these in its southern and northeastern parts, situated perhaps in accordance with the prevailing winds. Some presses were excavated by Magen (2008b: 275–278). The southern oil press (6.3 × 8.0 m) comprised at least two phases: an arcosolia burial cave was found underneath a pressing surface, pointing to an oil press dating to a later phase. A winepress, consisting of a rock-cut treading floor and a collecting vat to its east, operated in the building at a certain stage (perhaps even in the earlier phase). Remains of reddish plaster were visible on the walls of the collecting vat. The bottom of the vat was destroyed and was perhaps later converted into a cistern.

The Agricultural Plot The agricultural, cultivated areas of this settlement were divided into two sections. The eastern section began at the center of the settlement and consisted of a flat plateau suitable for agricultural crops and enclosed plots (see below). The western section included a rocky area with enclosed plots, field towers, and large burial complexes (see below). The northern, western and southern boundaries of the agricultural

A complex winepress. A large rock-cut winepress was discerned west of the residential area. The treading floor (7 × 7 m) was hewn and paved with mosaic, and in its center was a square depression (0.40 × 0.40 m) designed for a device for secondary extracting of the grapes’ skins (see also Frankel 1997: 193–207). A settling 77

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods

Fig. 3.55. Kh. Kesfa. General view, looking south (A. Shadman).

Enclosed plots (Fig. 3.56). Sixty-five plots, totaling c. 555 dunams, were recorded. These plots, situated south and east of the settlement, were particularly small, ranging from 0.5 to 11 dunams. There were two main plot clusters within this area: the northern cluster consisted of smaller plots measuring 1–2 dunams (Plots 30–33, 37–39, 40, 49–54 and 63), and they seem to have belonged to poorer residents. The southern cluster included larger plots measuring 3–10 dunams (Plots 2–22, 27–28, 41–48).

plot were quite clear, being bordered by three wadis: A tributary of Naḥal Rabah on the north (where a modern large quarry is located); Wadi al-Susi and Kh. al-Bureid on the west, and the steep Wadi al-Bureid on the south. The eastern boundary was not clear, but it was presumably around longitude 201, since further east were the agricultural areas of the villages of Rafat and Kh. Deir Qassis. The agricultural plot totaled c. 2562 dunams. Agricultural roads. Two roads reached out from this site: One to the farming areas located south and the other to the west, where quarries were located.

There were also several large plots west of the eastern cluster, but their borders were unclear; Safrai and Fixsler (1993) noted that these plots may have been owned by wealthy people. Another large cluster of plots was discovered west of the settlement, east of the Kh. Umm al-Bureid monastery. These plots extended over ridges and rocky slopes and extended over tens of dunams (Plot 58 was the largest, with over 45 dunams).

Road 13a. This north–south oriented road was in the southern part of the settlement. It passed through the middle of a flat plateau, characterized by alluvium and enclosed plots. The road may have enabled convenient passage to the fields and was also connected to the rural road in Wadi al-Bureid (Road 8).

Location of the agricultural crops. Kh. Kesfa dominated vast and varied areas. To the south and southeast of the site was a flat plateau, characterized by fertile alluvium, although not very deep. To the east and south of this area were gentle and rugged slopes. The area to the west, between Kh. Kesfa and Kh. alBureid (Site No. 7) consisted of a harsh, rocky landscape. The archaeological evidence from both, excavations

Road 13b. This road extended from the western part of the settlement westward, into a harsh rocky area. It was c. 500 m long and of variable width. Several hewn tombs (see below) were found next to it, as well as a large border stone and a later limekiln. The road ended high above the quarries west of the settlement and was probably used for transporting stones to Kh. Kesfa. 78

Amit Shadman: Chapter 3, The Sites

Fig. 3.56. Kh. Kesfa. Land distribution.

and surveys, indicated that the oil and wine industries were undoubtedly a central sector in the economy of Kh. Kesfa. Thus, taking in consideration the importance of grain, the cultivated areas can be divided as the following: The plain areas to the south of the site were apparently allocated mainly to olive groves and cereals (field towers were hardly discovered in this area). In the rocky areas to the west, vineyards were grown, as evidenced by the field towers and winepresses. As stated, five large oil presses were discovered within the settlement itself, attesting to the status of this industry. Other olive groves were probably in the extensive and rugged areas south and east of the plain, and seemingly also on the slopes above the northern bank of Wadi alBureid.

part of its cultivated areas and perhaps related to the winepresses scattered in these. Nine field towers were spread in between enclosed plots, where bedrock was exposed, and where vineyards were probably the main crop. Three additional towers were found south of the road mentioned above, and south of the settlement. Water cisterns. Most cisterns were identified within the confines of the buildings. A cistern, with its capstone, was recorded outside the built-up area, east of a large reservoir (see below). Reservoirs. An irregularly shaped reservoir (8 × 12 m, depth c. 3 m) was discerned in the southeastern part of the settlement. It was hewn in the bedrock and had a staircase leading into it on the northern side. A second reservoir (3 × 6 m, depth 2 m) was recorded west of the settlement, and a third irregularly shaped one was west of the burial complex situated west of the settlement (see below).

Winepresses. Besides the two winepresses within the settlement (see above), three winepresses were located in the agricultural plot: Two simple winepresses to the west of the site, and a third to its southwest, close to a rural road (Fig. 3.57).

Burial complexes. Three burial complexes were found. One was found on the south, near Buildings 21–25 (see above). The other two complexes were located one on

Field towers. Twelve field towers associated with the settlement were discerned, most of them in the western 79

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods Table 3.3. Kh. Kesfa. Land distribution. Plot No.

Area in dunams

Plot No.

Area in dunams

1

23.0

34

3.0

2

7.1

35

2.3

3

3.5

36

4.3

4

12.6

37

0.6

5

21.4

38

0.9

6

37.4

39

0.6

7

10.3

40

1.1

8

38.3

41

3.6

9

29.3

42

5.7

10

27.5

43

1.7

11

19.2

44

1.6

12

5.3

45

0.9

13

3.8

46

0.5

14

3.0

47

1.6

15

4.1

48

0.7

16

5.3

49

1.4

17

7.0

50

1.9

18

6.5

51

2.0

19

5.8

52

1.5

20

10.6

53

1.8

21

4.7

54

1.2

22

3.5

55

4.9

23

8.4

56

38.1

24

3.9

57

32.2

25

3.4

58

45.5

26

7.9

59

18.8

27

3.0

60

10.6

28

3.6

61

0.1

29

8.5

62

0.1

30

1.0

63

3.1

31

1.0

64

14.0

32

2.2

65

15.6

33

1.3

Total

555.56

settlement, and a large one, northwest of it (map reference 199900/666830) extending over a large area and probably serving as the main source for the construction of the settlement. The large quarry was scattered on a steep slope descending to Naḥal Rabah and was the largest in the study area. Although the area was badly disturbed in recent years, Kochavi and Beit Arieh (1994: Site 80) estimated the dimensions of the northwestern quarry at 200 m.

the northwest and c. 300 m west of the settlement. The northwestern complex comprised four rock-cut tombs whose openings were measured by Kochavi and Beit Arieh (1994: Site 115). Three of them were arcosolia caves. The western complex included five rockcut tombs: two rectangular arcosolia caves, a small rectangular tomb, and tombs without arcosolia. Quarries. Two building-stones quarries were discovered: A small one (Fig. 3.58), east of the 80

Amit Shadman: Chapter 3, The Sites

Fig. 3.57. Kh. Kesfa. A rock-cut winepress near Rafat–Migdal Afeq’s rural road (A. Shadman).

Fig. 3.58. Kh. Kesfa. A small quarry east of the site (A. Shadman).

81

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods

Sheep pens. A well-preserved round sheep pen was discovered a few hundred meters west of the settlement. The pen was built of small- and mediumsized fieldstones. Its date of construction is unclear and may have been built in a later period.

field towers and winepresses were found, pointing to vineyards and olive groves being grown there. Findings from various studies conducted at the site, such as a church and crosses (Guérin 1984; Safrai and Fixsler 1993: 203), attested to a Christian population. The building at Kh. Umm al-Bureid (Site No. 7), built of ashlars and containing colored tesserae and roof tiles, was perhaps a monastery, thus supporting the idea that a Christian population inhabited the site. Although this structure may have been the mansion of the village master or someone on his behalf, the overall picture at the Samaria foothill indicates that monasteries were a common regional phenomenon (see Chapter 7 below). The presence of monasteries next to large Byzantine settlements is well-attested for in archaeological research (e.g., Hirschfeld 1997b: 120).

Summary The settlement at Kh. Kesfa was larger compared to other settlements in the region and spread over vast lands in its vicinity. About 25 structures were recorded at the site, comprising about 80 rooms (Safrai and Fixsler 1993). The five oil presses were mostly built in the eastern part of the settlement, possibly due to the prevailing winds. Several oil presses were excavated (Magen 2008b), most of them operated by the ‘beam and screw’ technique. At some point, some of them were abandoned and their components were reused. A vast, yet not uniform, agricultural plot, was operated by the inhabitants of the site. Small plots were discovered in the eastern part of the settlement, whereas large plots were found to the south and west. Although this difference may reflect a socio-economic hierarchy, with wealthy people dominating larger plots and poor people owning small plots, most of the large plots to the west of the settlement were located on rocky soil, apparently inferior in quality to the eastern plots.

Guérin: “A large village, whose almost all ruined houses were built of relatively large stones, located on a hill covered today with vegetation and overlooking the deep valleys of Wadi Kesfa to the north and east. A building, on a west–east axis, which had been divided into three quarters, was apparently an ancient church. Two columns were still lying on the ground inside the site, while the others were burned into lime in a nearby kiln. At some distance away, I found a fine damaged stone on which it was possible to discern five concentric circles with a cross of equal arms in each of them; this stone probably belonged to the church. This stone was perhaps a door lintel. Then my attention was drawn to several rectangularly carved graves, two pools, and a dozen rock-hewn cisterns.” (Guérin 1984: 93).

Estimating the population of the settlement led Safrai and Fixsler (1993) to suggest a correlation between the number of buildings and the number of plots. However, this approach may be misleading for two main reasons. First, as noted above, the plots were not equal in size, and it is also possible the plots were cultivated by people who did not live there. Second, most plots within the agricultural plot were cultivated. In other words, besides the enclosed plots, extensive farming was also carried out in the dammed wadis, the hillsides, and wherever possible.

PEF:  “Modern ruins, apparently of a village” (Conder and Kitchener 1882, II: 337). This note was followed by Guérin’s full citation (see above).

Other the oil industry, a wine industry was also set in place, as evidenced by the winepresses discovered both in and around the settlement, one of these being the largest in the study area. The location of the crops within the agricultural plot revealed that in the area south and east of the settlement, characterized by fertile alluvium, as well as in the terraced wadis, wheat was grown, for it requires a deeper soil. West of the settlement was a rocky area where most of the

Previous studies: Conder and Kitchener 1881-1883, II: 337; Safrai 1983a; Guérin 1984: 93; Frankel 1986; IAA Archives; Safrai and Fixsler 1993; Kochavi and Beit Arieh 1994: Sites 115; 120; 121; Safrai 1995b; 1995c: 103– 104; Tepper 1997; Fixsler and Seelman 2000; Bagatti 2002: 158; Magen 2008b: 275–278; Frankel 2009; Magen and Kagan 2012: 136.

1

2 0

10

Fig. 3.59. Kh. Kesfa. Ceramic finds: A Byzantine LRC bowl and a jar.

82

Amit Shadman: Chapter 3, The Sites

Site No. 14  H. Migdal Afeq; Migdal Tzedek Mejdal Yaba Mirabel  Map reference: 196000/665300 

surrounding communities. The Mamluk Baybars conquered it in 1263, but the nearby village continued to exist, and its name was changed to Majdal Yaba, i.e., the Jaffa Tower. The site was captured by the Ottomans in the 16th century and was included in the territory of Damascus. According to an account from that period, only a few tens of people lived there.

A fortress, overlooking the entire region. The origin of the Hebrew name ‘Afeq’ is usually related to a water channel, and it can be attributed to the location of the site on the banks of Naḥal Shiloh, or to the Yarkon River to its northwest. The site was occupied from Iron Age I through late periods, but most of the visible remains are dated to the Crusader and Ottoman periods. The site, which controlled the Afeq Pass (the narrow passage between the sources of the Yarkon on the west and the Samaria Hills on the east), where the Via Maris ran, was a strategic point throughout history.

Sheikh Sadeq al-Jam‘ouni, an unreliable person, dwelled here in the 19th century, dominating several dozen villages in the area and flour mills along the Yarkon River. The site was thus named after him. During the period of the Jam‘ouni family, the fortress underwent several changes and rooms were added. The Ottoman authorities, which suspected the tax collection carried out by Sadeq al-Jam‘ouni, expelled him and nominated his son, Musa al-Jam‘ouni. A traveler who visited the site in the 19th century, the Dutchman van de Velda (1851), mentioned a monastery there, probably based on the Greek inscription at the entrance to the inner courtyard of the fortress. The French traveler Guérin was even more assertive and attested to an ancient church at the site, relying on the same Greek inscription stating that the church had been dedicated to St. Cyricus. However, except for this inscription and other stones in the fortress, no real building remains from the Byzantine period were discovered. 

Josephus Flavius mentioned the site twice. The first time, he referred to Alexander Jannaeus who built fortresses along the ‘Yannai Line’ in 86 BCE, between the coast and the mountains, meant to stop Antiochus VI’s army: “he cut a deep trench between Antipatris, which was near the mountains, and the shores of Joppa; he also erected a high wall before the trench, and built wooden towers, in order to hinder any sudden approaches. But still he was not able to exclude Antiochus, for he burnt the towers, and filled up the trenches, and marched on with his army.” (Wars I, 4:7). The second time, he mentioned the site as a Jewish settlement during the Great Revolt: “And now Gallus, seeing nothing more that looked towards an innovation in Galilee, returned with his army to Caesarea: but Cestius removed with his whole army, and marched to Antipatris; and when he was informed that there was a great body of Jewish forces gotten together in a certain tower called Aphek, he sent a party before to fight them; but this party dispersed the Jews by affrighting them before it came to a battle: so they came, and finding their camp deserted, they burnt it, as well as the villages that lay about it. But when Cestius had marched from Antipatris to Lod, he found the city empty of its men, for the whole multitude 28 were gone up to Jerusalem to the feast of tabernacles; except fifty people.” (Wars II, 19:1).

In the 1920s, Alt identified Aphek-Antipatris in Tel Aphek (Ras al-Ein), claiming that Migdal Tzedek should be sought in the hills to the east of this tell (Frankel and Kochavi 2000: 14). At the beginning of the 20th century, the site and its surroundings served as a center for lime production (Sasson 2014: 159–173). During the War of Independence, the village was abandoned, and the area was deserted until the fortress became a national park in 2005. Extensive conservation work was carried out at the site in recent years. Although this is an impressive and important site, no extensive archaeological excavations have been conducted there. Most excavations in the area, carried out by the Israel Antiquities Authority, followed development works (Shadman 2014; Tendler 2014). The existing citadel, identified mainly with Sheikh Sadeq, was founded on Crusader and Byzantine remains. Built of large ashlars, the Crusader northwestern tower was preserved to a considerable height (walls c. 2 m wide, inner dimensions 8 × 8 m). A few architectural items, such as a column fragment and a decorated stone, were found in secondary use (Tsuk, Bordowicz and Taxel 2016: 37–88). This fortress seems to have been of the ‘square’ type. Today, the remains of the Arab village spread south and southwest of the citadel, but according to a photograph from 1948, the village seems to have extended also to its north.

In the 12th century (probably around 1122 CE), the Crusader fortress of Mirabel was built here, replacing the site of Tel Aphek which had by then lost its importance. The Crusader fortress was destroyed in 1191 by the army of Saladin. A year later, after the defeat of the Muslims in the Battle of Arsuf in 1192, Saladin settled in the area of the citadel but abandoned it upon the arrival of Richard the Lionheart. In the 13th century, Mirabel received the status of an independent seigneurie and even served as a civic center for the

83

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods

The Agricultural Plot

(see below) were found, some already operated by Iron Age II–Hellenistic farms, and others in the Roman and Byzantine periods, and possibly a little later. Therefore, the wine industry was evidently prosperous in this area. Schumacher’s early 20th-century map (No. S.O.59 Rafat) showed the presence of agricultural crops (trees) on the ridge where the Ottoman fort was built. Presumably, these were olive groves. The function of this area in earlier periods remains unclear.

Migdal Afeq was the largest settlement in the region. It dominated large areas to its east and perhaps even in the plain to its west. The boundaries of the settlement followed its geographical location, the surrounding sites, and the main road—the Via Maris—which passed at its foothill. Naḥal Shiloh probably marked the boundary between the site and H. Zikhrin, located to the south. On the north was the site of Qurnat al-Haramiyya and on the east was a large fertile agricultural area, where most of the crops of Migdal Afeq grew. The western boundary of the settlement is not clear enough, and it is uncertain whether the inhabitants cultivated areas west of the Via Maris. It should be noted that no agricultural roads extending westward were observed in aerial photographs. In addition, vast areas west of Migdal Afeq were actually marshy during most periods. Hence, it is plausible that there was no daily traffic between the settlement and the area to its west.5

Winepresses. Eight rock-cut winepresses were discovered in conjunction with Migdal Afeq, most of these to its east. The winepresses were built on bedrock surfaces situated on the margins of the alluvial areas. These were mostly simple installations that included a treading floor and a collecting vat. Although some winepresses were excavated, it was difficult to date them since the site was occupied over a long period. Oil presses. Usually, these installations were not found outside the built-up area of Byzantine and early Islamic settlements. Nevertheless, a single lower crushing stone (diam. 2.10 m) of an oil press was found on the bedrock southeast of Migdal Afeq, perhaps in secondary use.

The agricultural-plot area of Migdal Afeq totaled almost 4000 dunams. However, most of this area, especially the eastern part, was heavily damaged by the large quarries operating there since the beginning of the 20th century. The area was later turned into a military zone. Nevertheless, the study of the agricultural plot was made possible by reviewing aerial photographs and descriptions made by 19thcentury travelers.

Water cisterns. About 20 water cisterns were discovered within the agricultural plot, most of them by Kochavi and Beit-Arieh (1994). Most cisterns were located on a rocky slope descending westward, while the others were found scattered over the cultivated areas east of the site. Following are their descriptions of several cisterns: “map reference:  195950/665330— a cistern with a square opening. Parts of its ceiling survived; map reference: 195800/665320—(diam. 7 m, depth 4 m), whose ceiling had collapsed; map reference: 195830/665260—A bell-shaped water cistern with a rounded capstone. Nearby, another cistern (currently in use) and a hewn trough, as well as the opening of a third cistern.” (Kochavi and Beit Arieh 1994: 70).

Enclosed plots (Fig. 3.60). Most of the area east of the site was characterized by vast areas and enclosed plots, clearly visible in the aerial photographs west of H. Te’ena. Fourteen plots, extending over a total area of c. 160 dunams, were identified, but it is obvious that many others did not survive. Location of the agricultural crops. Most of the areas east of Migdal Afeq were characterized by a fertile deep alluvial soil (min. depth 1 m at several spots). These areas were crowned by flat hills and low ridges characterized by a bare bedrock on which numerous and varied installations were hewn through history. Based on late sources (e.g., for the Crusader period— Amar 2000; see also Chapter VII, the Byzantine period), the cultivated areas of Migdal Afeq may have turned this settlement into the largest producer of wheat in the region. It was difficult to trace other crops since the settlement continued until the modern era, a long period during which major changes occurred in both the landscape and the needs of the inhabitants (e.g., a significant decline in wine production is seen during certain stages of the Islamic period). However, in almost all the rocky areas around the Migdal Afeq, winepresses

Reservoirs. An irregular-shaped reservoir (3 × 6 m, depth 3 m) was discovered on a gentle slope west of the fortress. A larger reservoir (4.10 × 6.20 m, depth 3.50 m), found to its north, was a hewn pool coated with gray plaster, with niches in its northern and western walls for supporting the roof. A hewn lower surface, perhaps representing a working surface associated with the reservoir, was exposed to its south, with three cut steps cut descending to it. Agricultural terraces. Large terraces were identified northeast of the settlement and south of Qasr asSitt (Site No. 1). These were built of medium-sized fieldstones (more than 2 m wide and high) on a gentle slope that descended eastward and ended in a small valley with fertile alluvial soil. In certain places, the terrace was used as an agricultural road, enabling traffic between various agricultural areas. Excavation

5 

It is possible that the roads did not survive due to modern development.

84

Amit Shadman: Chapter 3, The Sites

Fig. 3.60. Migdal Afeq. Land distribution.

of the terraces showed that they have an over 1-m deep fertile grumusol soil layer.

Table 3.4. Migdal Afeq. Land distribution. Plot No.

Area in dunams

1

4.3

2

16.6

3

6.4

4

11.5

5

11.8

6

5.2

7

9.0

8

15.5

9

9.3

10

18.5

11

9.4

12

12.6

13

20.5

14

8.2

Total

Burial Complexes. Many burial complexes were found and excavated, mainly west and north of the fortress (Figs. 3.61, 3.62). A rectangular rock-cut cave (0.8 × 2 m) was discovered c. 150 m west of the fortress and another one (0.5 × 2.0 m) next to this cave. Tombs belonging to several periods were traced on a hill north of the fortress. A cemetery from the Second Temple period (Tsuk 1993: 67–76) was excavated on the eastern and southern slopes of the hill. Two burial caves were exposed (‘The Magnificent Cave’ and the ‘kokhim cave’) with fragments of ossuaries that attested to a Jewish population during this period. A cemetery from the Late Ottoman period was excavated further down the hill, beneath which earlier burial caves were traced but not excavated. On the eastern slopes of the hill was another Muslim cemetery. The tomb of Sheikh Baraz ed-Din was identified on the summit, while the tomb of Sadeq al-Jam‘ouni was apparently several hundred meters southeast of the fortress.

158.81

85

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods

Fig. 3.61. Migdal Afeq. The ‘kokhim cave’ from the end of the Second Temple period, looking southwest (A. Shadman).

Fig. 3.62. Migdal Afeq. Plans of ‘The Magnificent Cave’ (right) and the ‘kokhim cave’ (left; Tsuk 1993).

86

Amit Shadman: Chapter 3, The Sites

Quarries. Stone quarries were discovered northwest of the fortress, in an area severely damaged by modern quarrying.

the name Majdal Yaba was once Migdal Oz. Indeed, the name Majdal is a transliteration of the Hebrew word Migdal, which also means a fortress, a fortified place, which we find in the Bible in construct state: MigdalEl, Migdal-Gad, which were two different cities, one in Naphtali and the other in Judea. However, there were many more fortified cities that are not mentioned in the Bible, as evidenced by the present one, and whose names consisted of the word tower, Arabic Majdal, and another word (here, for example, Yaba) to distinguish between them.” (Guérin 1984: 90–91). 

Summary The strategic location of Migdal Afeq made it an important site since the earliest occupation of the region. The ancient settlement spread over two hills separated by a saddle. It is difficult to estimate the built-up area of the settlement, but it was clearly more extensive than the extant remains. Potsherds discovered in previous studies showed that the site was already settled in the Iron Age and possibly even earlier, continuing without interruption until modern times. However, the site seems to have reached its peak during the Hellenistic period, and especially in the Roman and Byzantine periods. The main pre-Hellenistic site in the nearby area was at Qurnat al-Haramiya (see below, Site No. 15).

PEF: “A large and important village, evidently an ancient site, having ancient tombs and remains of church. It stands on high ground above the plain, and contains a house or palace of large size for the Sheikh; it was the seat of famous family who ruled the neighborhood… The water supply is from wells and cistern.” (Conder and Kitchener 1882, II: 286).

The most important historical reference to Migdal Afeq during the Hellenistic-Byzantine periods, Josephus Flavius, described the site and its Jewish inhabitants during the Great Revolt. This Jewish presence was evidenced by the burial complexes and small finds such as stone vessels (Fig. 3.63). According to Guérin, a Byzantine church existed in the site. The agricultural plot of the site was in the vast areas to its east.

“The house of Sheikh at this place is so built that one wall is against the wall of a building, possibly a church; this is used as stable, and is of massive masonry, with a side door surmounted with a lintel bearing the inscription:… ‘The Memorial (Church) of Saint Cerycus’. Over the lintel is a semicircular arch with a keystone… The inscription is on a winged tablet; the letters are about 4 inches long. The doorway is slightly pointed. The lintel is probably not in situ, but must have belonged to church, probably of Byzantine period. A little further north are fragments of building, which appear to be Crusading. At Sheikh Baraz ed Din there are several rough tombs and caves, one cemented. There is also a tomb of the kind called ‘rock-sunk’ with a vault 6 feet deep, and two unusually broad side loculi; thus, once more, the ‘rock-sunk tomb’ appears in connection with Christian ruins.” (Conder and Kitchener 1882, II: 360‒361).

Guérin: “On the summit of the hill, you can see the remains of an ancient Christian church whose pointed window arches apparently attest to Crusade construction. It is mostly destroyed, and the remaining part is used now as a den where the large herds of the sheikh are caged in the evening. His residence looks like a citadel, the upper part of which was renovated about twenty years ago and built on the ruins of an earlier citadel, much larger than it, probably of the same period as the church. When I came to visit this sheikh, I passed through the corridor of his feudal-like mansion and saw that its gate was constructed of ancient beautiful stones, one of which served as a lintel on which I read the following Greek inscription: HAPTYPIOMTOY. This inscription, whose engraved letters are very beautiful, lies within some kind of rectangular frame with a triangle on either side surrounded by another larger rectangular frame. This stone is two meters and twenty centimeters long. As the inscription points out, it covered the grave of St. Cyricus, whose body may have been placed under the altar of the nearby church—a church that apparently was built over another, much earlier one. Presumably,

Previous studies: Conder and Kitchener 1882, II: 286, 360–361; Rey 1883; Clermont-Ganneau 1896: 340–341; Albright 1923; Loewerstamm 1950: 501–503; Beyer 1951; Press 1951–1955, 3: 527; Prawer 1963, Vol. I: 306, 380, 446, 470, 540, 550, Vol. II: 77; Lange 1965; Benvenisti 1970; Vilnai 1977a, 1977b; Hoade 1978; Finkelstein 1978a: 20, Pl. 19; Bagatti 1983; 2002: 155–156; Guérin 1984: 90–91; Benvenisti 1984; Hartman 1987; Ellenblum 1992; Khalidi 1992; Tsuk 1993: 67–76; Kennedy 1994; Kochavi and Beit Arieh 1994: Sites 134, 135, 136; Tsafrir, Di Segni and Green 1994: 64; Pringle 1994; 1997: 67; Safrai 1995c: 104–106; Eliaz 2002; Raviv and Ellison 2014; Shadman 2014; Tendler 2014; Sasson 2014; Dayan 2015; Tsuk, Bordowicz and Taxel 2016: 37–88; IAA Archives.

87

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods Iron Age

2

3

1

Persian and Early Hellenistic Periods

4

5

6

7

Early Roman Periods 10 8

9

15 11

13

12

14

Byzantine

Crusader

16

17

Stone Vessels

18

19

21

20

0

10

Fig. 3.63. Migdal Afeq. Ceramic and stone vessels (courtesy of A. Tendler). Iron Age: bowl (1), jars (2–3); Persian and Hellenistic periods: mortaria (4), jars (5–6), jugs (7); Early Roman period: cooking pots (8–10), jars (11–13), juglets (14–15); Byzantine period: LRC bowl (16); Crusader period: glazed bowls (17); Early Roman Period: stone vessels (18–21).

88

Amit Shadman: Chapter 3, The Sites

Site No. 15 Qurnat al-Haramiya Map reference: 196250/666500 

(Fig. 3.65), cupmarks, water cisterns, and a simple oil press were exposed (Hadad 2011). The ceramic finds of these excavations were dated to the Iron Age through the Hellenistic period (Fig. 3.66).

Remains of a settlement on a rocky slope, c. 1 km north of Migdal Afeq. A modern Sheikh’s tomb built on top of an earlier circular structure was recorded, as well as water cisterns, cupmarks, and other rock-cut installations. A building, built of large-sized stones, the foundations of another rectangular structure also built of large-sized stones, as well as isolated wall sections and four columns were noted. A well-built wall of a compound,6 some building remains, and a number of rock-cut winepresses were discerned c. 300 m southeast of the site. Salvage excavations revealed remains of a settlement dating from Iron Age I to the Persian– Hellenistic period (Avner-Levi and Torge 1997; Fig. 3.64). The site’s main period of occupation was apparently in the Iron Age II, from which a casemate wall was found (Torge 2007: 22). Several simple pressing installations were discovered as well, attesting to the production of olive oil. In another excavation conducted at the site

The Agricultural Plot The location of the site on a summit above the plain, and its strategic control over the Via Maris and the ‘Aphek Pass’ defined its boundaries. To the north of the settlement was Naḥal Rabah, and the agricultural fields of the settlement did apparently not transcend this geographical barrier. To the south and west were Migdal Afeq and the Via Maris respectively, and it is unlikely the lands extended beyond these borders. East of the site was the small site of Kh. Taha (map reference 196900/667200), but the interrelations between the latter and Qurnat al-Haramiya is unclear. The two may have been associated (a farm?), sharing cultivated areas. It seems, therefore, that the eastern border of these two sites did not transcend longitude 198.

Fig. 3.64. Qurnat al-Haramiya. Site plan after the 1995 excavations (Torge 2018). 6  Kochavi and Beit Arieh (1994: Site 93) suggested that the compound was Salah ed-Din’s base while he besieged Mirabel (Migdal Afeq). Following several excavations conducted at the site (Tendler 2013ab), it is clear that the compound wall was actually an agricultural terrace.

89

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods

Fig. 3.65. Qurnat al-Haramiya. The ‘compound wall,’ looking south (courtesy of A. Tendler).

1

2

3 4

5

7

6

8 9

10

11 0

10

Fig. 3.66. Qurnat al-Haramiya. Ceramic finds from the ‘compound’ (Hadad 2011). Iron Age: bowl (1), chalice (2), cooking pot (3), jar (4); Persian period: jars (5–7); Early Hellenistic period: bowls (8), cooking pots (9–10), jars (11).

90

Amit Shadman: Chapter 3, The Sites

The agricultural plot totaled about 3605 dunams, with an extensive agricultural system. However, recent development works heavily damaged the ancient remains, and the site was studied based on past research and recent excavations (Avner-Levy and Torge 1997; 2018). Excavations on the eastern part of the agricultural plot yielded mainly agricultural installations.

therefore, an important site since the beginning of the Iron Age. This site, like Migdal Afeq, was almost continuously occupied until the modern era. The salvage excavations that followed the development works of the city of Rosh Ha‘ayin, conducted only on selected parts of the site, revealed remains dating from the Iron Age I and the Hellenistic period. However, additional studies showed that it had also been occupied in the Roman, Byzantine and later periods. Despite these studies, the interaction between this site and Migdal Afeq, the largest settlement in the area, has not yet been clarified. Studies conducted in the farming areas indicated that Qurnat al-Haramiya was similar in nature to the surrounding settlements.

Location of the agricultural crops. Most agricultural areas of the site were heavily developed over the years, making it difficult to evaluate the location of the agricultural crops in antiquity. The Sharon coastal plain, characterized by fertile alluvial soil, was west of the site, but the Via Maris that crossed this area through the ‘Aphek Pass’ created a barrier that apparently prevented the local residents from farming this area. Thus, the eastern plots of Antipatris seem to have extended not far from the site. The rest of the agricultural plot of Qurnat al-Haramiya was characterized by quite low hills and rocky ridges suitable mainly for growing olive groves and vineyards. Although some archaeological evidence for these crops was found, especially for the wine industry, the study of these areas was quite limited.

The nearby site of Kh. Taha, occupied from the Iron Age II to the Hellenistic period, was possibly a satellite site of Qurnat al-Haramiya. Another c. 150 × 300 m site was discovered 500 m to the northeast by Kochavi and Beit Arieh (1994: Site 87). However, according to their descriptions, this site was apparently an agricultural area with enclosed plots, rather than a settlement. It is surprising that the 19th-century travelers did not discover this site, as it was described by Kochavi and Beit Arieh (1994) and Finkelstein (1978a) as particularly large. In summary, the Iron Age occupation at Qurnat al-Haramiya was a large settlement that dominated the farms in its vicinity. It lost its status only in the later stages of the Hellenistic period when Migdal Afeq became the main settlement.

Winepresses. Eight winepresses were discovered within the settlement, most of them in its southeastern rocky section. These were mostly simple installations, some irregular-shaped, comprising a treading floor and a collecting vat. Water cisterns were found near four of the winepresses. Oil presses. Installations possibly pointing to an oil industry were discovered in previous studies (Haddad 2011; Tendler 2013a). A crushing stone was discovered southeast of Kh. Taha (Kochavi and Beit Arieh 1994: Site 68).

Guérin: “A small square enclosure, seventeen steps long, with its walls made of huge almost unworked stones, placed on top of each other without any bonding material. This stronghold on top of a hill, within a thicket of mastic tree (Pistacia lentiscus) and oak.” (Guérin 1984: 94–95).

Field Towers. Two field towers were identified in the farming areas of the settlement, one of these was discovered in a large area of enclosed plots. Water cisterns. Six water cisterns were discovered in the agricultural plot, but many others certainly existed. Most of the cisterns were located east of the built-up area, and several rock-cut winepresses were discovered near some of them.

PEF: “Stones of good size in foundation; near it a small rubble-work tower, apparently not ancient.” (Conder and Kitchener 1882, II: 360).

Tombs. A decorated arcosolia cave and a water cistern were discerned in the northeastern part of the settlement, and a shaft tomb and the opening of a burial cave were identified in its northwest part. Three rockcut tombs were found further on the spur, one with steps and the other with a carved socket for a rolling stone located on the opening’s edge.

Ceramic finds from previous surveys: Iron Age I and II; Persian–Roman periods. Ceramic finds from the ‘compound’ area: Byzantine, Early Islamic, Crusader and Mamluk periods. Previous studies: Conder and Kitchener 1882, II: 360; ; Finkelstein 1978a: 20; Guérin 1984: 94–95; Kochavi and Beit Arieh 1994: Sites 89 and 93; Avner-Levy and Torge 1997; 2018; Torge 2007; ‘Ad 2007; Elisha 2010; Hadad 2011; Tendler 2013a; 2013b; Drezner 2014; IAA Archives.

Summary Qurnat al-Haramiya is situated on the Via Maris, which passed through the nearby Aphek Pass, making it, 91

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods

Site No. 16 Elevation point 136 Map reference: 198600/665700

Summary This small farm overlooked two dammed wadis located to its west and east. To its south were a small flat plateau with several enclosed plots, two cisterns, and two simple winepresses. The agricultural area of the farm was difficult to evaluate since it may have been connected to one of the surrounding settlements. Unlike most of the farms in the area, the building was founded in the Hellenistic period and was reused in the Byzantine period. Based on the installations found in its vicinity, the inhabitants were seemingly engaged mainly in wine production and sheep farming.

A farmhouse at the top of a hill, west of a dammed wadi descending towards Wadi al-Bureid. Three rooms and an adjacent courtyard to their east were revealed in the excavations (Fig. 3.67, 3.68). Two of the rooms (L2013, L2034; 4 × 16 m) were in the western part of the building, and the third room (L2042; 4 × 10 m) was perpendicularly built in the eastern part of the building. Their walls (width c. 1 m), preserved to a height of 2 m, were built of two rows of fieldstones (max. length 1.5 m) filled in between with stones of various sizes. All the rooms had beaten-earth floors with ash spots (thickness 0.2–0.5 m) that served to level the rocky surface. The floor’s filling contained only worn Hellenistic sherds, but it was overlaid by mixed Hellenistic and Byzantine sherds, probably from the secondary use of the building. Flat stone slabs (length c. 1.5 m), probably used for roofing, were found in the building’s debris. The rooms were entered from the south. The eastern room (L2042) was entered through its narrow wall, and to the two western rooms (L2013, L2013) were accessed through their wide walls.

Ceramic finds: Hellenistic and Byzantine periods Previous Studies: Kochavi and Beit Arieh 1994: Site 147.

Fig. 3.67. Elevation point 136. Plan.

92

Amit Shadman: Chapter 3, The Sites

Fig. 3.68. Elevation point 136. Areal view.

Site No. 17 Naḥal Shiloh (2) Map reference: 197500/664600

North room

A farmhouse (28 × 28 m) on a gentle slope, northeast of Naḥal Shiloh (Fig. 3.68). The farm was built of medium- and large-sized fieldstones and comprised an upper (southern) and a lower (northern) level. A later limekiln cut into the southeastern corner of the building, destroying most of it. As a result, only the exterior walls of the building were preserved, rising to a height of two–three courses, as well as numerous potsherds.

Pool South room

Three poorly preserved rooms were discerned in the building: A trapezoidal room (c. 10 × 12 m) in the southwestern corner, with an irregular subterranean cavity, probably a reservoir, to its north; a room (c. 4 × 4 m) of an unclear function, northeast of the reservoir, founded on top of an agricultural terrace oriented east– west; and a third room (5 × 8 m), located outside the northwestern corner of the building. 

Limekiln

Fig. 3.69. Naḥal Shiloh (2). Plan.

93

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods

Previous studies: Finkelstein 1978a: 13; Kochavi and Beit Arieh 1994: Site no. 180.

Kochavi and Beit Arieh:  “Remains of complex (farmstead?) on the southern slope of plateau, north of Nahal Shiloh: large two-level courtyard (28 × 28 m), bounded on three sides by walls of large fieldstones (1.0 × 1.2 m). Entrance apparently centered in southern wall. Rectangular room (5.5 × 6.5 m) adjoins courtyard from northwest. Later lime kiln (5 m diam.) in the southeastern corner.” (Kochavi and Beit Arieh 1994: site 180).

Site No. 18  H. Tukkim  Kh. Umm et Tawaqy Map reference: 198950/663100 A rural settlement (c. 50 × 60 m), c. 500 m north of the steep northern bank of Naḥal Shiloh. The walls of the buildings, mostly built of ashlars, were preserved to a height of four–five courses. The site was not densely occupied, and the buildings were easily discerned.

Summary This farmhouse was one of the similar structures discovered along ‘the farm strip,’ but the limekiln, dating from a later period, apparently damaged the ancient remains, and the poor state of preservation did not allow for a thorough investigation. The site overlooked a wide wadi to its north with fertile land suitable for agriculture. The northern part of this area was apparently occupied by the residents of the neighboring site of Khallat es-Sihrij (Site No. 22).

Four building complexes, comprising 23 rooms and other spaces, were identified (Fig. 3.70): Complex 1 (Rooms 1–5; c. 10 × 20 m). Located in the southwestern part of the settlement, its rooms were quite small (5 × 5 m) and were accessed through Room 4. Rooms 5 and 1 were connected through a passage in a shared wall. An open courtyard (6) was east of this complex.

Ceramic finds: Iron Age II; Persian and Hellenistic periods.

Fig. 3.70. Kh. Tukkim. Plan.

94

Amit Shadman: Chapter 3, The Sites

Fig. 3.71. Kh. Tukkim. A pressing bed of an oil press.

Complex 2 (Rooms 7–14; c. 15 × 25 m). This complex, in the northwestern corner of the settlement, was accessed through Room 10. Room 7 was the largest (6 × 10 m) and was probably used as a watchtower (see Safrai 1995c: 112–113, Building no. 2). The northern part of Courtyard 6, south of the complex, comprised two cisterns. West of Complexes 1 and 2, a circular water reservoir was found, with hewn projections for roofing beams on its walls. A large winepress was discovered south of this reservoir.

Summary Kh. Tukkim was probably a satellite village of either the nearby community at Kh. Ta‘amur (Site No. 19, below) or H. Yeqavim (Site No. 9, above). This idea is supported by the presence of Road 18a which led from Kh. Tukkim to the first site and of the Main Road 5 which provided access to the second (see Chapter 6, Fig. 6.4). Both roads indicate routine traffic between these three settlements, but the proximity of Kh. Tukkim to Kh. Ta‘amur may perhaps point to a closer connection between these two. Naḥal Shiloh, south of Kh. Tukkim, marked its southern boundary, and the northern and western borders apparently coincided with the boundaries of H. Yeqavim. The multi-room complexes at Kh. Tukkim point to about 30 residents. The oil press and the winepress to its west attested to the production of oil and wine at the site. The agricultural areas of the settlement extended in all directions: to the northeast was a dammed wadi with agricultural terraces; to the north were dividing walls and a square field tower; and an agricultural area with farming terraces was discerned west of the site. The burial area, west of the site, comprised three hewn tombs. Further to the west was a limekiln dated to a later period.

Complex 3 (Rooms 15–17; 12 × 25 m). Located southeast of Courtyard 6, this complex probably served as an oil press. East of the complex was an open space, some kind of courtyard. A pressing bed of an oil press, leaning on one of the outer walls of the complex, was identified (Fig. 3.71). Complex 4 (Rooms 18–22). This complex, located at the eastern end of the settlement, with Courtyard 23 to it west, had a higher construction quality than any other part of the settlement and it may have served as the residence of the site’s master. The complex was entered from Courtyard 23, through Room 18. There was also a doorway leading from the courtyard out of the settlement, toward an agricultural road that passed through the site. A doorway in the western wall of Courtyard 23 led into other complexes in the settlement. A water cistern was located east of the complex. Contemporary traces of plaster and stone fences were discerned.

Previous studies: Conder and Kitchener 1881–1883, II: 358; Kochavi and Beit Arieh 1994: Site 224; Safrai 1995c: 112–113; IAA Archives. 95

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods

Site No. 19  Kh. Ta‘amur Map reference: 199900/662700

Complex 1 (Rooms 1–5). This complex was located in the western part of the settlement. A large courtyard in its center was probably entered from the east. The complex was adjacent to the western gate on the north and was probably associated with it. Room 1 (3.9 × 5.9 m) was in the northeastern corner of the complex and its northern wall delimited the street (above). Room 2 (2.5 × 7 m), with an internal subdivision, was south of the courtyard and had an entrance in its northwestern corner. Room 3 (2.9 × 8.2 m) was to the west of the courtyard, and further west was Room 4. A smaller room, Room 5 (2 × 3 m), was in the northwestern corner of the complex; its function and entrance remain unknown. 

A ‘rural town’ (100 × 100 m) on a saddle, c. 600 m north of Naḥal Shiloh. A compound, and scattered potsherds, were found on a hill southwest of the site. H. Masmer is a few hundred meters to its east and above Naḥal Shiloh. Numerous domestic buildings and installations, built of ashlars and large-sized fieldstones (length 1.5 m or longer, Fig. 3.72), were found scattered throughout the site. According to Safrai (1997a: 227–232), there were 16 buildings and a small market. The entrance to the main complex was via a passage or gate (width 1.80 m) discerned in the eastern part of the site (Fig. 3.73) and approached by a rural road from the northeast. The gate, built of partially hewn stones, was preserved to a height of four courses. Another gate (width 2.80 m), built of partially hewn stones and equally preserved to a height of four courses, was recorded in the western part of the settlement. It seems that the road that reached the eastern gate crossed through the middle of the settlement and exited it through the western gate, creating a road that extended further westward from the latter (see Chapter 6, Roads). A third gate was identified in the northwestern part of the settlement.

Complex 2 (Rooms 6–19). This complex is composed of 12 rooms and two courtyards. Room 6 was on its northwestern corner, and its entrance was on the west, adjacent to a cistern. East of Room 6 was Courtyard 7 (3.6 × 5.4 m), with two entrances in its northern wall. Rooms 8 and 9, apparently used for residential purposes, were approached from this courtyard. Three small adjacent Rooms (10–12) were in the southwestern part of the complex, each entered from the north, with another doorway linking Rooms 10 with 11. Room 13, east of Room 12, was adjacent to a large r-shaped courtyard (14) and to another courtyard (15) to its east. Rooms 16–19, aligned north–south in a row, were in the eastern part of the complex. Room 16 was entered through Courtyard 14, and Room 17 from Courtyard 15.

The site was densely occupied, making it difficult to locate the buildings, but it is estimated that it comprised seven complexes with about 50 rooms:

Passage

N Courtyard

East Gate Courtyard

Street

Legend Dwell Oil press m

Fig. 3.72. Kh. Ta‘amur. Plan (after Safrai 1997a: 227).

96

West Gate

Amit Shadman: Chapter 3, The Sites

Fig. 3.73. Kh. Ta‘amur. The eastern gate, looking west (A. Shadman).

Complex 3 (Rooms 20–26). Seven trapezoid-shaped rooms covering an area totaling 345 sqm were found in the southeastern part of the settlement. The elongated Room 23 (23 × 2.2 m), in the northwestern corner of the complex, was accessed from the settlement’s central courtyard. To its south was Room 20 (5.2 × 5.8 m), whose entrance was not traced. Room 21 (7.8 × 11.4 m), east of Room 20, was the largest in the complex and gave access to Room 22. Room 26 (5.0 × 10.4 m), north of Room 21, was apparently the central courtyard of the complex. Rooms 23–25 were approached through this courtyard. In the northeastern corner of Room 24 was an additional entrance to the complex.

(2.4 × 4.6 m) was to its north. The rooms were accessed from the courtyard to their west. The southern room had one entrance while the northern room had two. Complex 6 (Rooms 36–43). Eight rooms surrounding the large Courtyard 43 (10.0 × 17.7 m) in the northeastern corner of the settlement, where a lower crushing stone of an oil press and a hewn installation were found. The courtyard opened onto several rooms. Room 36 (2.1 × 8.6 m) was located on the southern side of the complex and was subdivided, but no entrance was identified. Room 37 (3.0 × 4.2 m) was discerned in the southwestern corner of the complex. It was entered through an opening in its southeastern corner and consisted of two inner spaces connected by a passage. North of this room was Room 38 (2.9 × 6.2 m), whose outline was unclear and may have had additional spaces that did not survive. Further north was Room 39, entered from the east. North of Courtyard 43 were three rooms: two elongated rooms, 41 and 42, and one small room, 40, in the northwestern corner of the complex. No entrances to these rooms were identified.

Complex 4 (Rooms 27–34). An irregular-shaped cluster of rooms. Room 27 (6.0 × 8.1 m) was the largest room, with an entrance in its northeastern corner. The room was not subdivided and was possibly a courtyard that bordered the north side of the street that crossed the settlement. A passage led from Room 27 to the Open Space 28 to its north, where an underground cavity was discovered, possibly an entrance to a hiding complex (?). Four rooms (29, 30, 31, 34), north of Space 28, were entered through a passage in the southern wall of Room 29 (3.0 × 6.2 m) and a doorway in the eastern wall of Room 34. To the north were Rooms 32 and 33, accessed through a doorway in the eastern wall of Room 32. A cistern was found in the middle of an open space located east of the complex.

Complex 7 (Rooms 44–49). A large complex with six rooms, located north of the central courtyard of the settlement. A corridor, in the southern part of the complex, was flanked by rooms on both sides. Two of the rooms (44, 48) were identified by Safrai (1997a: 229) as a central market. Room 44 (3.1 × 6.4 m) was located in the southwestern corner of the complex and was entered from the corridor to its east. North of this room, Room 45 (3.1 × 6.2 m) included two internal spaces entered from the same corridor. Room 46 was

Complex 5 (Building 35). A two-room building north of the eastern gate of the settlement. The southern room (2.4 × 5.0 m) abutted the gate, and the northern room 97

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods

the northernmost room of the western row of rooms, and it was also entered from the corridor. East of the corridor were two rooms:  an elongated room, 47, with two spaces (3.1 × 7.9 m; 3.1 × 4.6 m) and entered from the central courtyard of the settlement to its east; and the small Room 48 (3.0 × 3.1 m), in the southeastern corner of the complex, also accessed from the central courtyard. In the northern wing of the complex was Room 49 (8 × 12 m), with a courtyard adjacent to it on the west. This room could be accessed from this courtyard and through another opening on the east leading to a narrow passage which might have been another entrance to the settlement.

that connected Kh. Ta‘amur with Deir Balut. After exiting the residential complex of the settlement, the road turned west, going through a dammed wadi and connecting to rural Road 1 100 m down the stream (map ref. 199700/62685). Road 19b. A road leading from Kh. Ta‘amur northward (see Chapter 6, Fig. 6.4), to an area characterized by gentle slopes and enclosed plots. The road began on the western bank of a dammed wadi (map ref.199763/662727), where it intersected with road 19a. About 800 m down the road, it intersected with a northeast–southwest rural road (5). From this junction, the road led northward, to the southern plots of H. Yeqavim.

Complex 8 (Rooms 50–52). A large complex (224 sqm) located north of the western gate of the settlement. The complex was accessed from the street that crossed the settlement through a narrow passage. Room 50 was a large space with piles of stones which impeded identification of the rooms. The trapezoid Room 51, in the northeastern corner of the complex, was entered through a doorway in its southern wall. South of this room was a large area, possibly a courtyard, with a water cistern. Room 52 (7.0 × 9.9 m) was in the southeastern corner of the complex, near the street that crossed the settlement and to its western gate, and may have served as a watchtower. Many thresholds and lintels scattered throughout the complex, as well as a cistern and a rectangularly hewn bedrock (SLAB?) (possibly a tomb) were found.

Enclosed plots (Fig. 3.74). Most of the enclosed plots of the settlement were to its north and were approached through the agricultural Road 19b. A total of 15 plots were discovered, covering an area of c. 207 dunams. The plots were larger than others in the vicinity, with an area ranging between 10 and 25 dunams. The plots of Kh. Ta‘amur did not extend beyond the rural road between Kh. Tukkim and Deir Balut, where the plots of H. Yeqavim were located. Location of the agricultural crops. Considering the agricultural plot of Kh. Ta‘amur, most of the crops seem to have been vineyards and olive groves, as evidenced also by the archaeological finds. There were almost no areas suitable for growing wheat, except a dammed wadi located west of the settlement. When comparing the oil and the wine industries, it became clear that olive groves were more extensive than any other agricultural crops.

The Agricultural Plot The boundaries of Kh. Ta‘amur were determined based on topography and its relative location to the surrounding communities. Naḥal Shiloh, with its steep banks, was clearly its southern border. No agricultural systems, including roads, were discerned south of it. To the east was a sharp and steep curve of Naḥal Shiloh which delineated the eastern boundary of the slot. A rural road between Kh. Ta‘amur and Deir Balut to the east ran along the western bank of the gully. The Kh. Tukkim–Deir Balut road crossed the area north of the site, and the cultivated areas of Kh. Ta‘amur did seemingly not extend beyond this line. The presence in this area of many plots belonging to the neighboring H. Yeqavim is worth noticing. The western boundary of the site broadly stretched alongside the north–south rural Road 1. The agricultural area of the settlement totaled c. 2568 dunams (including the agricultural plot of Kh. Tukkim).

Only one winepress was discovered in the cultivated plots, and hardly any field towers, while the four oil presses found within the boundaries of the settlement attest to olives as the main crop in the area. However, it was impossible to distinguished between the plots used for olive groves and for vineyards. The existence of a market within the settlement indicates that the local residents imported other agricultural products from the surrounding settlements. Winepresses. Only one winepress was found in the agricultural plot, east of the site. It comprised a treading floor (4 × 7 m) and a collecting vat (1.20 × 1.20 m). Oil presses. Four oil presses were discerned at the site, one of them outside the main building complex, near the eastern gate. The high number of oil presses, compared to winepresses, attests to the central place of the oil industry in the settlement.

Agricultural roads. Two roads were associated with this site: Road 19a. A road that connected Kh. Ta‘amur with the main rural Road 1 on the west (see Chapter 6, Fig. 6.4). This road was, in fact, an extension of the road

Water cisterns. Seven cisterns were discovered in the settlement. Five were incorporated within the 98

Amit Shadman: Chapter 3, The Sites

Fig. 3.74. Kh. Ta‘amur. Land distribution.

residential complex and two were c. 100 m to its southwest. The capstone of one cistern was found in situ.

Table 3.5. H. Ta‘amur. Land distribution.

Reservoirs. An irregular opening, which seems to be that of a reservoir, was found north of the main residential complex.

1

24.9

2

22.3

3

23.8

4

20.2

5

5.5

6

12.8

7

10.3

8

11.7

9

21.7

10

9.6

Plot No.

Dammed wadis. An 800-m long dammed wadi was located west of the settlement. About 40 dams were recorded along its length at various distances from one another, depending on the width of the gully and its incline. The dams on the upper part of the gully were several hundred meters long, while those downstream reached only a few meters in length. Summary Kh. Ta‘amur was a significant component of the regional settlement system during the Byzantine period. Several roads led into it, including two that connected it to two large settlements in the area—Deir Balut and H. Yeqavim. This road network suggests this site was either a satellite village of Deir Balut or had

11

7.7

12

10.3

13

10.7

14

3.5

15

12.4

Total

99

Area in dunams

207.40

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods

special connections with H. Yeqavim. It may also have been an important independent settlement. Based on Safrai (1997a), it is suggested that Kh. Ta‘amur was a market town that supplied the needs of the residents of the area. Safrai mentioned that such markets were common during this period, although they were usually larger than the one in Kh. Ta‘amur (Safrai 1997: 232). A similar market was found, for example, in the town of Umm Reihan (Dar, Tepper and Safrai 1986: 37).

animal pen. Remains of another room (L9744) were discovered in the southwestern corner of the courtyard. Two entrances led into the building; one on the north (in W9701), leading into Room L9777, and the other on the south (L9755), leading to a fairly well-preserved staircase built of medium-sized flat fieldstones. The staircase descended into the courtyard but allowed also passage eastward to room L9719 (2 × 5 m). A doorway in the northern wall of this room provided passage to the rooms east of the courtyard. Due to the poor preservation of the interior, it was difficult to trace the inner subdivisions and the doorways. 

Unlike typical Byzantine sites in our region, and although the buildings at the site had a rather dense layout, the site seems to have been pre-planned. The several complexes found at the site flanked a street crossing it from east to west. This street apparently supports the idea that the site was a marketplace. The fifty rooms or built-up spaces at the site could accommodate about 50 families (at 3–5 persons per family). Accordingly, the whole population of Kh. Ta‘amur was of approximately 150–200 inhabitants dwelling on c. 10 dunams. Although the site was not excavated, the surveys clearly demonstrated it was founded during the Byzantine period (6th–7th centuries CE). The scant Early and Late Roman potsherds collected in the survey do not necessarily point to a real settlement during these periods.

In the northern part of W9726, which separated Rooms L9721 and L9720, an opening was exposed, with a monolithic doorpost preserved to a height of c. 1 m, and an adjacent stone threshold. Rooms L9735 and L9750 were approached, perhaps, from Room L9720 to their north. Two openings were located between Rooms L9739 and L9738, and Rooms L9738 and L9736. The entrance to the inner courtyard was through a wide doorway (L9776) placed on its western side. This was, presumably, the main entrance into the courtyard, and it is likely that it was also intended for the passage of sheep. Since no installations were discovered in the courtyard, it apparently served as a sheep pen. 

Ceramic finds: Early and Late Roman periods (few sherds); Byzantine period.

Later on, perhaps in the Persian period, two rooms were added in the southeastern part of the building. W9723 was built against the southern side of W9783 of the earlier phase. Partition W9785 was added between Rooms L9738 and L9732, creating four small cells (2 × 2 m each). Another modification of unclear extent was made south of the eastern row of rooms, where W9715, built of medium-sized fieldstones and preserved to a height of five courses, was built in the western part of Room L9719. Based on the Byzantine potsherds collected in this area, it is assumed that this was a field tower. Farming terraces from this period, built of fieldstones of various sizes and forming a kind of courtyard, were exposed north of the building and extended northward and northeastward for a considerable distance. A cistern, with its capstone, found west of the farmhouse, was presumably used by the residents of the farm since its establishment in the Iron Age II.

Previous Studies: Kochavi and Beit Arieh 1994: Site 253; Safrai 1990b: 113, 1995b, 1995c: 113–114, 1997a: 227–232, 1997c. Site No. 20 Naḥal Shiloh (3) Map reference: 1997070/664870 A farm on a hillside spur, several hundred meters east of H. Te’ena (Fig. 3.75). The site was first discovered during our survey and was later excavated. The excavation revealed a farm that existed in the Iron Age II and in the Persian period (Fig. 3.76). At a later stage, probably in the Byzantine period, an agricultural system was built in the area that included partition walls, farming terraces, and a rural road. The building was trapezoidal (21 × 22 m) and was built on a stepped bedrock slope. The natural steps were filled and paved with stone slabs. Most walls were founded on bedrock and usually built of two rows of local and sometimes slightly dressed stones occasionally filled with small stones. The exterior walls were well preserved to a height of four courses, but the internal partition walls were poorly preserved.

The Agricultural Plot No agricultural installations were found in the farm’s agricultural plot, and the few field towers in its vicinity were probably dated to the Byzantine period. A fairly simple winepress was discovered southeast of the farm and a second one c. 200 m to its west. 

The building comprised 12 rooms (2 × 3–3 × 3 m), arranged along the eastern side of an inner courtyard (12 × 18 m) which probably served as an

Another farmhouse was discovered nearby, above a dammed tributary of Naḥal Shiloh. The building (13.0 × 18.5 m, preserved to a height of two courses) was built 100

Amit Shadman: Chapter 3, The Sites

Fig. 3.75. Naḥal Shiloh (3). Plan.

of medium- and large-sized fieldstones, and comprised three main elongated rooms. The largest room (13 × 14 m), in the western part of the building, incorporated an additional smaller room (5.0 × 5.5 m). Another room, to the east and adjacent to the large room, was accessed from the south through a stepped entrance. A floor, built of flat fieldstones forming a leveled surface with the bedrock, was exposed in some parts of the building.

The building was poorly preserved, probably due to its proximity to a later limekiln built to its west. The ceramic finds recovered from the building dated from the Iron Age II and the Persian period. Previous Studies: Tendler and Shedman 2015: 193–207.

101

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods

1

2

5

4

3

7

6

8

10

9

11

12

13

14 0

10

Fig. 3.76. Naḥal Shiloh (3). Ceramic finds. Iron Age: bowls (1–5), holemouth jars (9–10); Persian period: a krater (6), cooking pots (7–8), jars (11–13), a jug (14).

Site No. 21 Naḥal Shiloh (4) Map reference: 197900/664040 A square farmhouse (20 × 20 m; Fig. 3.77), on the western bank of a tributary of Naḥal Shiloh. It is situated above the junction of two wadis comprising a wellcultivated large area (see below). Kh. Tukkim is about one kilometer to the southeast, but based on the road network, these two sites did seemingly not maintain mutual relations. The farm was built of medium- and large-sized fieldstones (up to 1.30 m long), and its outer walls (1.30 m wide) were well preserved to a height of three courses (1.1 m). The entrance to the building was through an opening in the southeastern wall, approached through a natural rock terrace. A residential room (8 × 8 m), mostly not preserved, was discerned in the southwestern corner of the building. In other parts of the building, covered with stone collapse, the rooms were not preserved. Fig. 3.77. Naḥal Shiloh (4). Plan.

A massive wall running along the northwestern side of the building was discerned, possibly an earlier farming terrace, but its function remained unclear. A semicircular stage, built of small fieldstones and abutting a bedrock terrace, was found c. 12 m north of the building and was possibly a later addition. 

This farm was connected to the local road network through two agricultural roads. The first road, north of the site, on the western bank of a wadi, connected to the 102

Amit Shadman: Chapter 3, The Sites

rural Road 4 (H. Yeqavim–Migdal Afeq). The second was discerned east of the farm and also connected to rural Road 4 in the enclosed-plots area west of H. Yeqavim. Considering these two roads and the site’s location midway between H. Yeqavim to the east and Migdal Afeq to the west, it is difficult to assess the affinity of the farm to either of these large communities. 

surrounding a large courtyard. The two-faced outer walls of the building were built of medium- and largesized fieldstones (c. 1.5 m long), while the rooms’ walls were built of a single row of large-sized stones. Two rooms were discerned in the building: one in the southeastern corner (c. 4 × 6 m), preserved to a height of three courses, and a poorly preserved second room in the northeastern corner. West of this room was a raised platform, of an unclear function, and there may have been an additional room. A row of rooms was also most likely constructed in the western part of the building. The entrance to the building was through a wide opening in its western wall, where two walls running eastward and forming a kind of corridor flanked the opening on both sides.

The cultivated plots of the farm were in its immediate vicinity and included three dammed wadis. This area provided all the residents’ needs, and even more. No wine or oil presses were identified in the cultivated areas of the farm, possibly indicating that the inhabitants engaged in grain crops and sheep farming. No water cisterns or other storage facilities were discovered at the site.

An underground water reservoir, in the western part of the courtyard, was entered from the west through a hewn passage which may have comprised some steps, although none were preserved. The opening of the reservoir (0.90 m wide; Fig. 3.79.1–2) led into a large cavity of an unknown depth, whose walls were coated with gray plaster. A rock-hewn cistern (diameter of opening 1 m, min. depth 2.30 m) was discovered c. 11 m east of the reservoir’s opening and possibly connected to the reservoir. 

Ceramic finds: Byzantine period. Previous Studies: Kochavi and Beit Arieh 1994: Site 183. Site No. 22  Khallat es-Sihrij Map reference: 197600/664900  The remains of a settlement (50 × 60 m; Fig. 3.78) on the western slope of a gentle hill, c. 1.5 km east of the Migdal Afeq and c. 300 m north of Naḥal Shiloh 2 (Site No. 17), dated to the Iron Age II, the Persian and the Hellenistic periods. It may have been a satellite village of Migdal Afeq, or a farm composed of rooms

A limekiln, probably from the Ottoman period, which made use of the available stones, was found in the southwestern corner of the complex. This kiln, together with contemporary works carried out in the area, damaged the ancient settlement.

Fig. 3.78. Khallat es-Sihrij. Aerial View (photo credit: Gilad Itach – Israel Antiquities Authority).

103

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods

Fig 3.79.2. Khallat as-Sihrij. Water reservoir, looking outside (photograph: A. Peres – IAA).

Fig. 3.79.1. Khallat es-Sihrij. The entrance to the water reservoir (photo credit: Gilad Itach – IAA).

The Agricultural Plot

The size of the building, the remarkable water reservoir, and the farm’s location—dominating fertile alluvial lands—seem to indicate it supervised over other farms in the nearby area. The later plaster repairs applied to the water reservoir attest to its continued activity long after the residential building ceased to function. However, since the site was not excavated, it is difficult to know how the reservoir related to the building during the Iron Age II, the Persian and Hellenistic periods.

The boundaries of Khallat es-Sihrij were determined based on topography and on the site’s relative location to the surrounding communities. Most of its agricultural plots seem to have been in a fertile alluvial area that bordered the fields of another farm west of the settlement (Finkelstein 1978a: 11, Site 65/3; Kochavi and Beit Arieh 1994: Site 138). About 300 m south of this area were the fields of Naḥal Shiloh 2 (Site No. 17), through which the main rural road (Road 4) passed. North of the site was another farm (Site No. 5), with its own agricultural plot. To the east, there were harsh steep slopes that could be cultivated on a limited scale. Since only a single simple winepress and a few field towers were discovered in the cultivated area of Khallat es-Sihrij, the main agricultural activities seem to have included growing grain and sheep farming. In conclusion, the farm, in this site, was probably larger than the other farms scattered throughout the area.

Ceramic finds: Iron Age II; Persian, Hellenistic and Ottoman periods. Previous studies: Finkelstein 1978a: 1, Pls. 12–13, Tables 6, 17; 1981: 332–333, 342–343, drawings 3, 8; Kochavi and Beit Arieh 1994: Site No. 176; Brandl and Itach 2019: 211–226.

104

Amit Shadman: Chapter 3, The Sites

Table 3.6. List of sites in the study area. No.

Name

Type

Longitude

Latitude

No. in Rosh Ha‘ayin map

Period

Notes

1.

Qasr es-Sitt

Farm

197750

666000

102

IAII, Persian, Hellenistic, Byzantine

Excavated

2.

Wadi al-Bureid

Farm/ administrative building

198300

665800

145

Persian, Hellenistic

Excavated

3.

H. Te’ena

Monastery(?)

198600

665100

153

Byzantine, early Islamic

Excavated

4.

Naḥal Shiloh

farm

197940

665300

142

IAII, Persian, Hellenistic

Excavated

5.

Naḥal Shiloh (west)

farm

142

IAII, Persian, Hellenistic, Byzantine

Excavated

6.

Wadi al-Bureid (east)

Farm

199620

665580

155

Byzantine

Seattleite site of Kh. Umm alHammam

7.

Kh. Umm alBureid

Monastery(?)

198800

666500

108

Persian, Hellenistic (few sherds), Byzantine, early Islamic

Guerin mentioned a church; excavated

8.

Kh. Dayyar

Monastery/ church

198450

667280

75

Byzantine, Mamluk, Ottoman

9.

H. Yeqavim

Rural town

199550

664100

205

Roman (few sherds), Byzantine, early Islamic

10.

Wadi al-Bureid (south)

Estate

199400

665000

159

Byzantine, early Islamic, Mamluk

11.

Kh. Umm alHammam

Rural town

199720

665300

158

Byzantine, early Islamic, Ottoman

12.

Nahal Shiloh (south-west)

Fortress

198700

664400

189

IAII, Persian, Roman, Byzantine, early Islamic

13.

Kh. Kesfa

Rural town

199990

666700

120, 121

Byzantine, early Islamic

14.

Migdal Afek

Fortress and a town

196000

665300

134–136

IAII, Persian through early Excavated Islamic, Crusade

15.

Qurnat alHaramiya

Satellite village

196250

666500

89, 93

IAI-II, Persian through early Islamic, Crusade, Mamluk

Excavated

16.

Elevation point 136

198600

665700

147

Hellenistic, Byzantine

Excavated

197500

664600

180

IAII, Persian, Hellenistic

180

IAII, Persian, Hellenistic

Several oil presses were excavated

17.

Nahal Shiloh 2

Farm

18.

Kh. Tukkim

Farm

19.

Kh. Ta‘amur

Rural town

199900

662700

253

Roman (few sherds), Byzantine

Hidden complex (?)

20.

Nahal Shiloh 3

Farm

199070

664870

-

IAII, Persian, Hellenistic

Excavated

21.

Nahal Shiloh 4

Farm

197900

664040

183

Byzantine

22.

Khallat es-Sihrij Satellite village(?)

197600

664900

176

IAII, Persian, Hellenistic, Ottoman

105

106

CHAPTER 4

THE ARCHAEOLOGICAL SURVEY—METHODOLOGY

Introduction

It seems, therefore, that despite the high potential to cover large areas through high-resolution surveys, the surveyors must be aware of the limitations of the terrain, which is sometimes deceptive. Dagan (2011a: 7) published the results of the several surveys conducted prior to the construction of the Cross-Israel Highway. He mentioned the skills required of any surveyor: “An archaeological survey requires training and experience that enable the surveyor to deal with this unique way of collecting data: [1] To know the study area; [2] To examine the changes made in the area, e.g., construction works; [3] To pinpoint the survey sites using GPS; [4] To accurately document the area and the collected finds; [5] To accurately identify the indicative potsherds; [6] To know and use the various means at his disposal, in order to organize the data according to the GIS system.”

The present study is based on two main archaeological methods: surveys and excavations. An archaeological survey has many advantages over excavations, for it allows to cover large areas in a relatively short time without causing damage. Also, in contrast to excavations, surveys allow for collecting preliminary information (besides gathering potsherds) on a particular area/site, with minimal resources (Ben-Yosef 2007), as well as for reconstructing and discerning demographic changes that occurred in the various geographic areas and assess the agricultural and technological capabilities of the inhabitants (Banning 1996). This archaeological method (see also Banning 2002) enables to conduct an extensive occupational survey and create a horizontal-spatial stratigraphy of a given region on which period maps can be based (Dagan 2011: 6). However, the archaeological survey also has some limitations (Faust and Safrai 2015: 9–15). The quality of the survey is affected by the characteristics of each region, e.g., ancient remains in the mountainous region differ in nature from those in the coastal area. Naturally, it is easier to trace ancient remains in the bare rocky mountains than identify them in alluvial lands where they have been covered over the years. Moreover, the older the remains, the less likely they will be found in an archaeological survey that does not penetrate the ground. Another problem is the quality of the surveyors and their skills for identifying, both during fieldwork and after it, while analyzing ceramics or flint.

Site Definition The definition of the term ‘site’ varies from one surveyor to another. Some define ‘site’ as a locality where any human activity took place at a specific time (Rupp 1983: 26–27), and others distinguish between a site as an ancient locality, and man-made elements such as agricultural installations, cisterns, roads, and so on (Zertal 1996: 15). In the present study, a distinction is made between survey sites (features), which present elements of archaeological interest such as scattered pottery, installations and so forth, and archaeological sites—the settlements themselves—ranging from oneroom farmhouses to multi-building settlements. Types of Surveys

Collecting potsherds is one tool on which archaeological research relies, and thus, often, also its weakness. Incorrect reading of the pottery, collecting potsherds washed from a higher spot, or contemporary intervention—all these, and more, may affect the results of the survey and its interpretation. For example, in the vicinity of Shoham, south of the study area, a sevenstrata site was excavated (Dahari and ‘Ad 2008), but the surveys carried out prior to the excavation traced no remains (see Gophna and Beit Arieh 1997). It should be taken into consideration, however, that the contrary may also happen, i.e., large-scale sites are identified in surveys but yield almost nothing during excavations.

Surveys can be divided into the six following types: Map Surveys. These surveys served as the basis for the ‘Israel Survey’, which started in 1964 and ended in the 1980s, aimed at conducting a full archaeological survey of the country, and for which the Association for the Archaeological Survey of Israel was established. The country was divided—based on the Israel Mapping Center method—into 10×x10 km (100 sq km) ‘squares,’ each representing an independent unit comprising a wide range of topographic, demographic, climatic, and 107

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods

Emergency Surveys. Emergency surveys are carried out when a region is under risk, mainly resulting from political actions. For example, an emergency survey was conducted in the Negev region at the end of 1978, before Israel’s withdrawal from Sinai and the construction of new military bases (Haiman 2014: 20– 21; 205: 98).

other characteristics. As of today, 130 survey maps have been published, covering about half of Israel’s area. Three survey maps of the Samaria foothills were studied within the present study: the Map of Rosh Ha‘ayin (Kochavi and Beit Arieh 1994), covering our entire study area; the Map of Lod (Gophna and Beit Arieh 1997), including the region south of the study area; and the Map of Kfar Saba (Beit-Arieh and Ayalon 2012), including the area extending from Rosh Ha‘ayin northward.

Development Surveys in the Rosh Ha‘ayin Area As part of Rosh Ha‘ayin’s development, many surveys were conducted in the area, followed by extensive salvage excavations. The first surveys were carried out by Haiman and Barda (Haiman 1998; 1999) in the 1990s, east of Migdal Afeq and H. Anusha (about the latter, see Haiman and Barda 2005). Korenfeld (2014: 24–26) and Assis conducted surveys between 2010 and 2015 in areas designated by the Ministry of Construction and Housing (Fig. 4.1). Up to present, seven surveys have been conducted in the study area, covering 20 sq km and enabling a high-resolution and reliable study of the area.

Regional Surveys. These surveys are carried out on a broad regional basis and are usually aimed at specific geographical areas such as, for example, southern Samaria (Finkelstein and Lederman 1997), the western Jezreel Valley (Ben-Tor 1980: 30–44), and the Judean Shephelah (Dagan 2004). However, in this regard, Zertal’s work in the Manasseh Hill Country, which lasted over 35 years, is probably one of the most impressive and meticulous (Zertal 1992; 1996, 2005, 2012; Zertal and Mirkam 2000).

Reliability of Surveys vs. Excavations in Rosh Ha‘ayin

Theme Surveys. Surveys designed to examine a particular finding, or theme, studied by the researcher. For example, Finkelstein (1978) surveyed the same region of our study area, exploring the farmhouses in the Samaria foothills. He focused on studying a particular type of structures and specific periods, ignoring other findings in the region, regardless of their value. Obviously, he was aware of the other findings, but they were barely included in his research.

As noted above, each research method has its own advantages and disadvantages. In rocky mountainous areas, the survey makes it possible to define the general chronological framework of a site, but not always the function of its various sections, something that can be achieved through excavation. For example, during the survey of a farmhouse, potsherds dating from various periods are likely to be found. On the other hand, finding potsherds dating only from the last occupational phase, during excavations at the site, may lead to ignoring the other occupation periods. Frequently, the residents of the later occupation cleared the earlier remains, leading to a false picture. This apparent discrepancy is pronounced with farmhouses—where almost no architectural changes were made—as seen in the current study area. Another factor that may bias potential conclusions is the fact that many farms were abandoned or used as seasonal dwellings.

Development Surveys. Surveys carried out prior to new development projects. Any findings of archaeological interest are documented and mapped—from isolated potsherds to large sites. The surveys are conducted in areas designated by the developers. Development surveys for projects initiated by various governmental agencies are usually carried out by the IAA staff. In some surveys, mechanical equipment is used to locate ancient remains not visible on the surface, especially in the alluvial regions. One of the most extensive development surveys was carried out, in recent years, along Route 6—the Cross-Israel Highway (Dagan 2011a). This survey resulted in a wide variety of archaeological data.

The surveys conducted in the study area yielded archaeological data that proved most reliable. Finkelstein’s surveys (1978, 1981), for example, revealed that most farmhouses were occupied during the Iron Age II through the Hellenistic period. These dates were confirmed in the excavations conducted by us in these farmhouses (e.g., Sites 2, 4, 5). Also, Kochavi and Beit Arieh (1994) dated the large settlements in their survey to the Byzantine period, dates again proved reliable after excavation, such as in H. Te’ena, for example, dated in the survey to the Byzantine and Early Islamic

Surveys and Test Probes. In recent years, additional surveying methods have been used, mainly by combining a surface survey with excavated probes (Faust and Katz 2012: 158–185; Shai and Uziel 2014: 172–190). This method, seemingly more effective near multi-strata tells than in open or mountainous areas, can provide a higher resolution assessment, as it penetrates the archaeological strata.

108

Amit Shadman: Chapter 4, The Archaeological Survey—Methodology

Summary

periods (Kochavi and Beit Arieh 1994: Site 153), and confirmed in the excavations. However, Kochavi and Beit Arieh’s suggested dating for the various installations seems to be less reliable. It is impossible to deal with findings from sites (villages and farmhouses) and installations (winepresses, agricultural terraces or roads) in the same manner, since, compared with agricultural installations, sites contain much larger amounts of potsherds and coins, often found in or near the buildings.

The importance of archaeological surveys and their contribution to research is widely acknowledged. Various factors, including population growth in Israel which caused the expansion of many villages and towns, political changes, academically initiated studies, and so forth, led to implementing many archaeological surveys. All the surveys conducted since the 19th century are milestones in comprehensive archaeological studies, including excavations. The database collected through the surveys provides researchers with a preliminary understanding of the area to be explored/excavated. Among the surveys conducted within the study area, the seven high-resolution development surveys are of particular importance for the research of the region, showing the problems of each area. Wherever bedrock is exposed, there is a limited probability of finding an unknown architectural feature. However, it is always possible to discover underground cavities such as hiding tunnels. Potsherds found in any given site are still the primary dating tool, yet by introducing new methods originating in the exact sciences (see below), additional and valuable information is at hand.

It seems, therefore, that surveys in mountainous regions, such as the Samaria foothill, where bedrock is exposed, are quite reliable, as confirmed in excavations. Also, since the archaeological survey alone does not allow to fully understand a given site, excavations are required to achieve this. For example, the “three elongated structures” discerned by Kochavi and Beit Arieh (1994: Site 145) at the site of Wadi al-Bureid (see Chapter 2, Site No. 2) were found to be, in retrospect, a single structure. However, there are other cases where surveys successfully traced the sites’ plan prior to excavation.

Fig. 4.1. Rosh Ha‘ayin. Map of archaeological features found in development survey (courtesy of the Israel Antiquities Authority).

109

CHAPTER 5

LANDSCAPE ARCHAEOLOGY

Introduction

can apparently also help locate urban complexes.1 Extensive studies covering large areas in the vicinity of Jerusalem were later carried out by Ron (1977a, 1977b), while Dar studied Mt. Hermon and the Western Samaria Hills (1978, 1982a, respectively). Gibson’s extensive research on agricultural systems of various regions (Gibson 1995, 2001a, 2003) is also worth mentioning. Later studies analyzing large areas have shed additional light on open-area installations and have contributed to further clarify their nature (Portugali 1982; Ben-Tor 1980; Dar 1982a; Gibson and Edelstein 1985; Dar, Tepper and Safrai1986; MacDonald 1988; Kochavi 1989; Gibson, Ibbs and Kloner 1991; Dauphin and Gibson 1992–93; Greenhut 1998; Gibson 1995; 2001; 2003; 2015; Edelstein, Milevski and Aurant 1998; Gorzalzany 2007; Leibner 2004; 2009).

Landscape Archaeology studies vast regions within their geographic and archaeological landscapes. The archaeological and environmental components of a given settlement (sometimes of several settlements), both within and around it, are examined as a whole: public buildings, residential buildings, industrial facilities, roads, agricultural fields, and more. Open-area installations have not been the focus of archaeological research in the past, only in recent decades becoming a subject of interest among researchers. Presumably, the interest in landscape and environmental archaeology increased when archaeologists tried to understand the areas surrounding the settlements, and the interactions between the two. After common archaeological features in open areas (farming terraces, dams, field walls, roads, and more) seemed to be somewhat out of context, the scholars, having acknowledged the limitations of standard surveys and excavations, sought for an additional tool to understand these features within a more comprehensive framework.

The Question of Dating Different landscape types should be taken into account when dating agricultural systems. Most surviving agricultural systems in the Land of Israel are in mountainous regions, moderate hills, and wadis, where erosion is significant. Soil instability in these areas causes the transfer of potsherds, which are the primary dating archaeological tool at hand. Excavating an agricultural terrace, a rural road or any installation in these areas will probably yield ceramic finds from several periods, allowing for only a general definition of the chronological range.

The first studies of open areas, conducted in Western Europe during the 1930s, were done through aerial photographs (Gibson 2001a: 32) initially used by the military but of great value for archaeological research (Dar 2000: 155). Early, important research in landscape archeology was carried out in the 1950s by Crawford (1953) and Bradford (1957), and later in the 1970s by Fowler (1975). In his book, Bradford (1957: 62, 70) presented important aerial photographs of Iraq, depicting, among other things, an ancient road he associated with the Roman army and ancient cities. Similar studies were also conducted in the Land of Israel and its environs. Between the two world wars, Glueck (1960) carried out some interesting studies on the rural environment of the Jordan Valley and the Arava, while Reifenberg (1950) studied the rural environment of the Land of Israel, Syria, Transjordan, and Iraq. According to Reifenberg (1950: Photo 53), aerial photographs

Using old aerial photographs and updated maps has become the main method for solving the problem of dating agricultural systems. Almost every region in the Land of Israel has been photographed from the air, either by the foreign armies operating in the early 20th century (the German Air Force, and especially the Bavarian Squadron No. 304, the Australian troops, and the British RAF) or by the 1 

The ruins of Caesarea prior to the large-scale excavations are visible in one photograph. The aerial view enabled to locate previously unknown features.

111

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods

Fig. 5.1. H. Yeqavim and H. Te’ena-Deir Balut road.

Israel Mapping Center and the IDF. These aerial photographs provide information about open areas before these underwent intensive development, and also enable viewing certain regions where no human or mechanical activities left their clues in hundreds of years (see Roll 2000: 131–141, who studied ancient roads, tracing them from aerial photographs). After reviewing the photographs and getting acquainted with the terrain (also from the ground), the sites in the given region should be evaluated in conjunction with their surroundings. For example, the site of H. Yeqavim (Site No. 9) is approached by several roads and surrounded by enclosed agricultural fields. However, viewing an aerial photograph and a survey of the area revealed remains of an ancient road between H. Te’ena (Site No. 3) and H. Yeqavim, which continued to Deir Ballut (Road No. 3: H. Te’ena–Deir Balut).

Furthermore, the plots of H. Yeqavim, adjacent to that 5th-century road, are aligned with it and do not cut it, confirming that they were used together or else the latter would have ignored them. This is a good example of ‘horizontal stratigraphy’ (Fig. 5.1).

The presence of this ancient road raises two questions: Whom did it serve? When was it constructed? The answer to the first question was already at hand after observing the aerial photograph. The road began at H. Te’ena and continued toward Deir Balut, connecting the two sites. However, dating the road required an indepth analysis. Examining H. Te’ena showed that the site had been occupied from the Byzantine through the Early Islamic period. Deir Balut, on the other hand, had been occupied for a longer period. Therefore, since H. Te’ena was founded in the 5th century (possibly as a monastery), and there was no apparent reason for constructing a road leading to it before its foundation, the road, and the settlement, were both likely built in the same period. The agricultural plots of H. Yeqavim, located east and northeast of the settlement, were also probably built in the Byzantine period as this is the main occupational period of the settlement.

Determining the boundaries of the ‘agricultural plots’, namely the settlement’s agricultural hinterland, posed some difficulties. While the area of an urban settlement may clearly be defined from its surrounding fortification system (Gibson 2001a: 33), such is not the case with rural settlements and their adjacent agricultural plots. Analysis of the agricultural systems in the study area shows that the agricultural plots were located close to the communities which owned them. The farmers commuted to their fields for only a short distance and for a limited time of no more than half an hour.2 This reality was revealed after examining the lands of the traditional Arab villages, showing that the share of the

In conclusion, assuming the H. Te’ena-Deir Balut road was constructed during the 5th century CE—at the same time or after founding H. Te’ena—then the plots of H. Yeqavim must have been built at a later time yet still, perhaps, within this century. It should be emphasized that even if H. Yeqavim had been founded before the 5th century, the plots could still be dated to the above proposed period, or slightly later. Defining the boundaries of the Settlements in the area

2 

The distance between a settlement and its adjacent agricultural plots was not equally fixed. We know of rural regions in the Land of Israel where the distance exceeded 5 km and sometimes even reached 20 km (Robinson and Smith 1856, Vol. I: 212, II: 216–220; Ron 1977a: 37–30; Amiran 1990: 20–34).

112

(1993: 216), it is unlikely that a farmer would cross a steep wadi to reach his fields (such as Wadi al-Bureid, which separates Kh. Umm al-Hammam and Kh. Kesfa). This is supported by the agricultural-roads network which did not continue towards plots located beyond the nearby topographical barrier.

enclosed plots in the total cultivated areas of the village becomes smaller the farther away someone is from the latter (Ron 1977b: 223, Grossman 1994: 22). It would be worthwhile to re-examine the Arab villages of the early 20th century and follow the British authorities’ efforts to assist the local peasants. The government understood that the quality of the lands differs from one region to another, and surveys were conducted to evaluate their qualities. The authorities sought to collect taxes based on the quality of the land, meaning the owners of fertile lands would pay higher taxes than the owners of poor lands (Grossman 2004: 197). Although deducing from such advanced practices on the reality in Antiquity might seem difficult, throughout the present study it often became clear that there are substantial differences among enclosed plots adjacent to settlements. Such a case was found among the plots of Kh. Kesfa (Site 13; Safrai and Fixler 1993).

Besides analyzing the terrain and measuring the walking time—based mainly on roads and boundary markers—a computer program was used for examining walking distances in the most convenient topographic conditions (Fig. 5.2). The map drawn in the field and the computer-generated map showed great similarity, with only minor differences discerned mainly in the agricultural plot of H. Te’ena, in the west part of the study area. This discrepancy is not surprising, as the vague natural boundaries of this area difficult the accurate mapping of the settlements. The boundaries of this site are marked by boundary markers on the east, north, and west. No boundary marker was found in its southern part, where the agricultural plot is delimited by the wadi. A large dividing wall was discerned on the wadi’s southern bank. In contrast, the eastern part of the study area consisted of a rocky area bisected by deep wadis. However, as noted above, in the site of H. Yeqavim—whose western boundary was further westward—no significant changes were observed between the two maps.

The boundaries of the settlements were examined based on the latter’s agricultural-roads network, boundary markers (see below), and the topography of the area. This model has been already implemented in other regions, such as H. Sumqah in the Carmel (Dar 1998: 222–250). The topography of the area dictated, to some extent, the distribution of land in the area. The wadis and steep slopes forced the settlement to own specific lands. According to Safrai and Fixler

Fig. 5.2. Estimated boundaries of the Byzantine settlements. The dotted and full lines represent the boundaries as reconstructed by the author; the colored areas are the boundaries generated by the computer-generated map.

113

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods

Fig. 5.3. Upper right: an ancient boundary marker of an agricultural plot. Upper left: a boundary marker of the 1949 ‘Green Line’. Bottom: an ancient boundary marker of a village (A. Shadman).

Boundary Stones

the Nessana papyri, where a meticulous land division is described (Kraemer 1958: 95–101; Evenari, Shanan and Tadmor 1971: 120–125; Stroumsa 2008).

The supervision and control of lands have been significant issues throughout history. During the Roman period, for example, under the reign of Diocletian, boundary markers were placed in large rural lands in the Golan and the Hula Valley for the purpose of collecting taxes (Sion and Hartal, 2003: 233–239).3 A study conducted in the agricultural plot of Kh. al-Burk, in the Samaria Hills, shows that this phenomenon was also prevalent in rural areas comprising many plots (Dar 1982a: 113, photo 62). Land divisions and their marking are also mentioned in written sources such as

Within the study area, boundary markers were used to separate enclosed plots, while in the buffer zones they marked the limits of the settlements (Fig. 5.3). Boundary markers were mainly found in large plots, each made of slightly worked medium- to large-sized fieldstones. Most of them were trapezoidal and were integrated into dividing walls or placed near them. Several scholars have suggested that these stones marked boundaries between plots belonging to local farmers. Safrai and Fixler (1993: 210) found stones standing in the vicinity of Kh. Kesfa and suggested these signify the intersection of field fences. Placing

3  Unlike the stone described by Sion and Hartal, the stones in our region are uninscribed.

114

Summary

prominent stones in the rural landscape apparently indicated the delimitation of a given area.4 Large- to very large-sized stones (some 1.50 × 1.90 m) were set in many other places along the boundaries between villages and rural roads, possibly designating the boundary of a certain settlement. For example, the large stone discerned at the edge of the main road (Road No. 1), between H. Te’ena and Kh. Umm alHammam, represented, in fact, the boundary between the two communities, with the plots of H. Te’ena to its east and the extensive agricultural plots of Kh. Umm alHammam to its west. Another example was found west of both H. Yeqavim and Kh. Kesfa, where a boundary marker was placed next to an agricultural road that extended west of the settlement.

Studying the rural landscape requires a research method that takes into consideration both the archaeological findings and the different landscape components. It has become clear that the methods used in urban areas, including tells, are not applicable in open areas. It should be emphasized, however, that this method does not replace previous ones, including surveys and archaeological excavations, but rather integrates different disciplines: geography, hydrology, geology, botany, and others. Establishing new towns in Israel, and expanding others like Rosh Ha‘ayin, generates many opportunities for archaeological research. Most times, the ancient landscape is preserved in these areas, enabling scholars to examine them as they were prior to modern development. Although the discipline of archaeology is not new, landscape archeology seems to be still in its infancy.

In a study conducted in Midrakh ‘Oz, in the western Jezreel Valley, an attempt to define the boundaries of a given settlement was made (Tepper 2009). Tepper described rock-cut crosses and suggested that these were boundary markers placed between the agricultural plots or signs of ownership. Similar studies were carried out in Roman-period rural settlements in England (Fowler 1975: 131), but the stones in these sites were much higher than those in the Samaria Foothills.

4  Surprisingly, and probably coincidentally, trapezoidal concrete pillars are still present in our region, marking the pre-1967 ‘Green Line’. The similarity between the boundary markers in Antiquity and the 20th century is quite striking.

115

CHAPTER 6

ROADS AND AGRICULTURAL SYSTEMS

Rural roads—connecting the settlements to heir fields (henceforth, for the purposes of this study, ‘agricultural road’). Local roads—connecting villages and towns (henceforth ‘rural road’). Main roads—connecting the various cities. Imperial Roman roads—the international imperial road system.

Roads Rural and agricultural roads played an important role in transporting the crops from our region to the neighboring areas, and perhaps even to the large cities on the Coastal Plain. Rural roads were frequently connected to crossroads, connected in turn to the ‘national’ road network which was increasingly being developed during the Roman period.1 Roll (1996: 151– 156) divides the roads into four types:

Fig. 6.1. Agricultural systems in the study area. 1

For further bibliography see: Avi-Yonah and Safrai 1966; Aharoni 1974; 1987; Roll 1976; Tsafrir 1984: 49; Karmon 1984; Kasher 1988; Dan 1990a: 107; ; Roll 1996: 1166; Neeman 1995; McCollough and Edwards 1997; Roll 2000; Tal 2006: 12.

117

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods International Roads

and build bridges over the streams of Naḥal Hadera, Naḥal Alexander, Naḥal Poleg and the Yarqon, some of which survived to this day (Roll 1976: 48; Ziegelmann 1982; Safrai et al. 1990: 249–261). Most parts of the road did not survive, as these were based on sandy soil, but some remnants are still standing: in Caesarea’s vicinity (Avi-Yonah 1959: 35–46), and a bridge over Naḥal Hadera. According to several milestones discovered along the road, it was probably constructed in the 1st century CE (Roll and Ayalon 1987: 150). Two other segments of the road were found near H. Balkiah and Tel Qasile beneath cement remnants dated to the Byzantine period (Ayalon 1987: 9–12), attesting to the road having functioned during a long period.

Two important international roads crossed the Land of Israel and its neighboring regions. The first is the ‘Via Maris’ (see below), used throughout most historical periods, and the second is the ‘King’s Highway’, running from today’s Syria, through Jordan, to Aqaba (Isaac and Roll 1982: 3; Van Beek 1958: 141–152; Avi-Yonah and Safrai 1966: Map 57; Aharoni 1987), also known in the Bible as ‘Hamesilah’ (Number 20:19), and during the Roman period as the ‘Via Nova Traiana’. Although this road is far from the present study area, it should be mentioned as it was part of the road network of the Land of Israel. ‘The Via Maris’

Due to its importance, the road was described by ancient historians. Josephus (Wars IV, 11,5) described Titus’ campaign, from Egypt to Caesarea (69 CE), noting that he had passed through the coastal cities but without mentioning the historical Aphek-Antipatris Pass. Thus, it is likely that this campaign used the road under discussion. In the Peutinger map (Tabula Peutingeriana; Finkelstein 1978b: 71–79; Dilke 1985),2 the road is shown as part of the Land of Israel’s road network. Safrai (1981: 334) saw a great similarity in the roads on the Peutinger map and the routes used in the Roman military campaigns and argued that the map reflects the road system in the Land of Israel during the period preceding the Bar Kokhba Revolt. Theodosius, a 6th century Byzantine pilgrim, also mentioned this road (Wilkinson 1977: 63, Map 19).

The study area is confined on the west by the main road Yavneh–Lod–Antipatris–Caesarea. However, this road was part of the international route that connected Egypt and Syria. Although many sections of the road are not located along the coastline, it has been generally referred to as ‘The Via Maris’. The name “Way of the Sea” is present in the Bible, in the book of Isaiah (8:23), whereas in the book of Exodus (13:17), this route is named “the way of the land of the Philistines” (Aharoni 1987). It is during the Roman period that it received its Latin name ‘Via Maris,’ by which it is identified today. The road originated in the Eastern Delta of the Nile, known in Egyptian sources as Sile, 3 km east of modern Qantara (Gardiner 1947: 202), the departure point for military campaigns and commercial convoys. The northern section of the Delta was named by the Egyptians the ‘Ways of Hor’, along which they built fortresses and stations for the convoys (Aharoni 1987). This road ran close to the shoreline up to the Ono Valley, in the Lod area, where, because of the Yarqon barrier, it turned east to Aphek-Antipatris. In the gap between the Samaria foothills and the springs of the Yarqon near Tel Aphek-Antipatris, was a strategic historical pass evidenced, inter alia, by the itinerary of Pharaoh Thutmose III, who conquered Aphek on his way north to Yehem in the Sharon Plain (Simons 1937: 117). The road continued beyond Aphek toward the Jezreel Valley, the Transjordan, and Syria. Presumably, the Roman-period route followed this ancient one, but this is not evident (Roll and Ayalon 1987: 155). Roll and Ayalon (ibid.) suggest that the section north of Antipatris served the Jewish population who could thus travel between the Galilee and Judea bypassing the Samaria hills which were blocked to them.

2. Antipatris–Caesarea Road. This road was constructed to connect Caesarea to Lod and Jerusalem through Antipatris, which was the major city in southern Sharon Plain. North of this city, the road ran through the swampy areas of Naḥal Alexander and Naḥal Hadera, crossed by bridges built by the Roman authorities (Roll and Ayalon 1987: 152–153). A milestone, along with remains of the road’s pavement discovered on the modern road between Netanya and Tulkarm (Dar and Applebaum 1973: 93), enabled to trace its route. South of Antipatris, the road bifurcated: one road went toward Lod, and the other toward Jerusalem along the Samaria foothills.

‘National’ Longitudinal Roads (Fig. 6.2)

Based on historical sources, the road seemingly functioned as a principal route through history. Josephus (Wars II, 1) mentions that during Gallus’ campaigns to suppress the Great Revolt (66 CE), his troops marched from Caesarea to Antipatris via this road. Then he describes a Jewish settlement that overlooked the road, in the vicinity of Midgal (Migdal

1. Caesarea–Apollonia–Jaffa Road. The increasing importance of the coastal cities led to the construction of a road to connect them. The road ran close to the seashore, and its constructors had to overcome swamps

2  The original Puettinger map was drawn in the 4th century, and perhaps even in the 2nd century CE, by the Roman cosmographer Castorius. It describes the entire Roman Empire as well as its road network, including the Land of Israel. The map was copied in the 12th century and was discovered again in the 16th century by Puetinger.

118

Amit Shadman: Chapter 6, Roads and Agricultural Systems 2. Neapolis–Jaffa Road. This road descended from Neapolis to Naḥal Alexander along a route that sometimes overlaps the modern road. Near Kafr Lakf, the road turned southwest and continued on the topographical ridge of Ḥableh toward the Coastal Plain (Roll and Ayalon 1983: 132-133). Once reaching the Yarqon River, the road crossed it on a bridge marked on the Jacotin map (Karmon 1960: 155–156). From there the road continued southwest to Jaffa. Several milestones were found on various spots along the road (Roll and Ayalon 1983).

Afeq, see Site No. 14). However, despite Josephus’ description, it is not clear whether the road was paved during this period (Safrai et al. 1990: 253). Puetinger’s map shows the road connected Caesarea directly to Lod. The traveler from Bordeaux (333 CE) describes the road and mentions the springs of Ras el-‘Ein (IshShalom 1966: 217). The Jewish rabbinical source ‘Pirkei Avot’ of Rabbi Natan (1885) describes the road as “the aristrea from Lod to Caesarea,” and Roll and Ayalon (1987) suggested this term should be actually the word ‘strata’ that referred to Roman imperial roads. Caesarea and Antipatris were rebuilt during Herod’s reign; Caesarea became an important port that connected the Land of Israel to the Roman world, and Antipatris, due to its strategic location, became an influential city that controlled the sources of the Yarqon River and the Afeq Pass to its east. Due to the important role of these two cities, the road between them became the most important route in the Sharon Plain during the Roman and Byzantine periods (Zohar 1991: 65).

3. Neapolis–Apollonia Road. This road was actually a northern branch of the Neapolis–Jaffa road discussed above. This branch turned northwest in the area of Kh. Ras ‘Atiyeh and ran through Nebi Yemin, where it intersected the Antipatris-Caesarea road, from there continuing westward and crossing the kurkar ridge toward Apollonia via Kefar Saba and H. Sabiya (Roll and Ayalon 1986b: 116–117). Remnants of rock-cutting, a pavement, and a fragment of a milestone were found along the road between Kh. Ras ‘Atiyeh and Ḥableh (Roll and Ayalon 1983: 134).

3. Jerusalem–Neapolis Road. This road was one of the main transportation routes in the central mountain region during the Roman and Byzantine periods. It started in Be’er Sheba and ran for a distance of 159 km through the main cities along the Judean and Samaria hills (Hebron, Jerusalem, Neapolis), ending in Jenin (Dorsey 1991: 132–136). Due to the topography of the Samaria Hills, the road was divided into two branches (Zertal 1992; Sion 2001: 22): The western branch passed through Neapolis and Sebaste and the eastern one through Tubas. Latitudinal roads running from the Coastal Plain to the Jordan Valley connected to this road (see below).

4. Antipatris–Apollonia Road. Although the route of this road is not archaeologically evidenced, it is assumed that such a route existed because of the importance of these cities in the Roman and Byzantine periods (Roll and Ayalon 1987: 158). Its existence is further supported by being included in the regional road network and by its connection to the Jerusalem– Gophna–Antipatris road (see below). 5. Ḥaris–Antipatris Road. This road was the western branch of the main Jerusalem–Neapolis road. It turned west near Kafr Yassuf and descended toward Kafr Hares, intersected the Neopolis–Lod road and continued toward Kafr Qassem, reaching Antipatris (Roll and Ayalon 1983). The road was preserved mainly west of Kafr Ḥares, where retaining walls, remains of pavement and rock-cuttings were discerned (Roll and Ayalon 1986b: 117). Topographically, this is a fairly convenient route, as the area—consisting of moderate slopes and saddles—did not create barriers. It should be noted that this is the route which the Philistines used in their Shiloh campaign (1 Samuel 4), as evidenced by the Iron Age fortresses in the region (Kochavi 1977).

Latitudinal Regional Roads 1. Sebaste-Apollonia Road. Starting at Sebaste, this road turned southwest to Naḥal Shechem, and then to Bet Lid (Roll and Ayalon 1983: 131). From Bet Lid, the road continues on a topographical ridge to Ẓur Natan, and then southwest to Tira. Near Tira, the road intersected the Antipatris–Caesarea road (see above), and continued to the swampy area north of Herzliya, reaching Apollonia. Dar (1981: 376–382), who investigated the eastern part of the route, noted that the road was 8 m wide, bordered by curbstones with built-up and hewn steps. The segment he excavated revealed the road’s 13 m wide and c. 0.5 m thick bedding. Segments of this road underlay now modern roads (Roll and Ayalon 1983). The presence of fortresses along the road indicates that it was probably used from the Hellenistic through the Roman and Byzantine periods (Dar 1981). During the Ottoman period, the road served pilgrims who traveled to the Sidna ‘Ali mosque near Apollonia.

6. Neapolis–Lod Road. The road departed from Neapolis and first ran along the Michmetat Valley to the Ḥuwarah area. The route intersected the Antipatris–Neapolis road (above) in the Kafr Ḥares area and then crossed Naḥal Shiloh, intersecting with the Antipatris–Gophna road (below) west of al-Luban. From this area, the road descended southwest to Kh. ‘Al and Kh. el-Bireh (Dar and Safrai 1984), continuing from there toward Tirat Yehudah and Lod. The road was well preserved between Naḥal Shiloh and Rantis, where 119

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods

Fig. 6.2. The main roads in the region (after Tsafrir, Di Segni and Green 1994).

curbstones, rock-cuttings, and pavement remnants were discerned (Roll and Ayalon 1983: 136). However, Safrai’s study (1980a: 55–64) is worth noticing, as he offers an alternative southern route to Lod, along the Samaria foothill.

period, the importance of the cities mentioned above. The road turned northwest in the area of Gophna and continued toward Bir Zeit and Umm Safa, then west to Timna. Descending toward Rantis, the road took advantage of the moderate range sloping westward to Rantis (Finkelstein 1977). From the Rantis area, it continued on the steep southern bank of Naḥal Shiloh to the boundary between the Samaria foothills and the Coastal Plain.

A probe excavated east of el-Bireh revealed that this section of the road was founded on bedrock, overlaid by a 60-cm thick fill of medium-sized stones. The road was paved with flat slabs (Roll and Ayalon 1986a: 38). Several milestones were found in the vicinity of Rantis, one of them inscribed and mentioning Hadrian. The section of the road running between Lod and Rantis is mentioned in historical sources. Eusebius mentioned the names ‘Armthem Seipha’ (Rantis) and Diospolis (Lod), as well as Timna, all associated by him with the territory of Lod (Roll and Ayalon 1983, note 32). Jerome, in the early 5th century, also referred to this section of the road while describing Paula’s pilgrimage to Eretz Israel (Roll and Ayalon 1983, No. 33).

The first to study this road were the British surveyors (SWP), who briefly described it (Conder and Kitchener 1882: 300, 310). Based mainly on the SWP findings, Thomsen (1917: 76) described four milestones, one (from the area of Gophna) indicating a distance of 13 miles from Jerusalem. Avi-Yonah (1984: 82) suggested that the milestone discovered near Shu‘afat and bearing the name of Emperor Nerva (96–98 CE), points to the construction date of the road, i.e., after the Great Revolt. Another milestone, dated to Elagabalus’ reign (219 CE), was discovered near Tel Aphek-Antipatris (Roll and Ayalon 1983: note 37). A structure overseeing the road was excavated on top of a hill above the mouth of Naḥal Shiloh, south of Migdal Afeq (Bahat and Bucheri 1963). It was identified as a Roman fortress, and a coin minted in Neapolis and dated to the time of Severus (200 CE) was found there. Presumably, the

7. Gophna-Antipatris Road (Fig. 6.3). This was one of the main latitudinal roads, connecting the Samaria Hills to the Coastal Plain (Finkelstein 1977: 171; Tsuk 1982: 14). This route passed through the cities of Gophna and Timna and significantly shortened the way from Jerusalem to Caesarea, by that increasing, in the Roman 120

Amit Shadman: Chapter 6, Roads and Agricultural Systems

Fig. 6.3. Gophna-Antipatris Road. milestone found in the excavations (courtesy of O. Sion).

fortress was associated with the road at least during the Roman period. Tsuk (1982: 14) identified a segment of the road that included a retaining wall and a milestone.

the Roman usage, went in the front of the army after a decent manner, and marched through Samaria to Gophna…” (Wars V, 2, 1). In the New Testament, after Paul was captured by the Roman army, he was taken from Jerusalem to Caesarea, stopping overnight at Antipatris (Acts 23: 31–30). Eusebius mentioned Timna on the road from Lod to Jerusalem (Onomasticon No. 486).

The section between the Rantis junction and ‘Abud was investigated by Roll and Ayalon (1983), who recorded some curbstones and a retaining wall. A probe excavated by them in H. Leved revealed that the road was constructed on a fill of soil and small stones (60 cm thick) which overlaid the bedrock. Flat slabs overlaid a second, higher soil layer (Roll and Ayalon 1986a: 36–37). During the intensive excavations conducted at H. Anusha and H. Leved on behalf of the IAA, another segment of the road was found (Sion, ‘Ad and Haiman 2007: Fig. 14). It became clear that the road (8.5 m wide) was constructed of flat field stones laid 1 m above the bedrock. In several other spots, the road was hewn in the bedrock (5.7 m wide).

Rural Roads (Fig. 6.4) The terms ‘rural road’ and ‘agricultural road’ were established by Dar (1982a: 197–229) in the framework of his extensive research in the Western Samaria Hills. The rural roads—average width 3–4 m, with some 5–6 m-wide exceptions—connected the settlements to the agricultural roads so the farmers could commute to their fields, and stretched from a few hundred meters to several kilometers. The road’s margins, up to two– three courses high, were mostly built of mediumand large-sized fieldstones with mud in between. Sometimes, one side of the road served as a double retaining wall. The road’s pavement was usually made of a bedding of small-fieldstones with mud in between. These roads were often founded on bedrock, and in several instances, the latter was hewn to moderate the route’s slope. Potholes were leveled using local soil and stones. The construction techniques of the agricultural roads were similar to those of the rural roads but were usually shorter. Agricultural roads were discussed in

Despite its importance in the Roman and Byzantine periods, the road is only rarely mentioned in historical sources (Finkelstein 1977: 173). The earliest reference is in Josephus’ writings, who described the campaign of the Roman army led by Titus: “Now, as Titus was upon his march into the enemy’s country, the auxiliaries that were sent by the kings marched first, having all the other auxiliaries with them; after whom followed those that were to prepare the roads and measure out the camp. Now Titus, according to 121

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods

Fig. 6.4. Rural and agricultural roads.

122

Amit Shadman: Chapter 6, Roads and Agricultural Systems Chapter 3, as part of the description of the agricultural plots of each village.3

3, and then ran north on a leveled plateau, continuing northwest for c. 500 m and reaching the site of Naḥal Shiloh site (Site No. 4). At map ref. 198770/665400, the road entered one of the tributaries of Wadi Susi and continued between its agricultural terraces, up to its northern mouth. In the junction point of Wadi Susi and Wadi al-Bureid, the road passed on the eastern slopes of a hill, and at map ref. 198500/666260, it met rural Road No. 2 that ran northeast–southwest. The road continued northward to Kh. Dayyar and apparently joined the Antipatris–Ḥares road (see above). In addition, at the southern end of the road, there was a steep and winding ascent leading south from Naḥal Shiloh, where it met the Gophna–Antipatris road.

Rural roads were found also in studies (mainly surveys) conducted in the Samaria Foothills, (Kochavi and Beit Arieh 1994), in the northwestern Samaria Hills (Dar, Tepper and Safrai 1986: 95–101), in Ramat Beit Shemesh (Dagan 2010), in the Judean Shephelah (Dagan 2006), and in the Central Judean Hills (Weiss, Zissu and Soleimany 2004). The roads observed by 19th-century travelers (see Dar 1998: 29) are also noteworthy. It should be emphasized that it is not always possible to discern the differences between a rural road and an agricultural road, and often they are defined only after analyzing a given region. As noted above, rural roads were integrated into the regional road network and were controlled by the local authorities (Dar 1982a: 197). Ten rural roads were identified in the study area. These were examined during salvage excavations and development surveys or during our survey, or through aerial photographs and maps from the 1940s, 1960s, and 1970s.

2. Rafat–Kh. Kesfa–H. Te’ena–Migdal Afeq Road. The road runs for about 6 km between Kh. Kesfa and H. Te’ena, and for 6.5 km between Kh. Kesfa and Migdal Afeq. It was identified by Kochavi and Beit Arieh (1994: Sites 111, 148), and selected segments were excavated in salvage excavations along modern route 513. Its origin is unclear, but maps from the 1930s and 1940s, and aerial photographs from the 1960s show that it departed from Rafat and ran westward. The road then passed through enclosed plots located east of Kh. Kesfa and connected to an additional route that was discerned south of the settlement (map ref. 200000/666600). From Kh. Kesfa, the road continued southwest to Kh. Umm el-Bureid (Site No. 7). Between Kh. Kesfa and Kh. Umm el-Bureid are enclosed plots and agricultural installations, such as field towers and rock-cut wine presses. Southwest of

1. The Naḥal Shiloh-Kh. Dayyar Road. Starting at Naḥal Shiloh (map ref. 198968-662547), this 6-km long road turned northeast and climbed a relatively steep slope where it was leveled by cutting the bedrock (Fig. 6.5). After 800 m, it reached Kh. Ta‘amur (Site No. 19), and about a kilometer to the north it reached the agricultural plots west of H. Yeqavim (Site No. 9). At map ref. 199260/664500, it intersected rural Road No.

Fig. 6.5. Road No. 1. Naḥal Shiloh–Kh. Dayyar (A. Shadman). 3  The agricultural roads were discussed in Chapter 3 as part of the description of the agricultural plots of each village.

123

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods that overlooked H. Yeqavim to its north. East of H. Yeqavim, the route passed between enclosed plots that apparently belonged to this settlement. The section of the road east of H. Yeqavim that approached Deir Balut still serves the residents of the latter. Probes excavated by us along the road showed that its width varies, in

K. Umm el-Bureid (map ref. 198500/666260), the road intersected Road No. 1. About 200 m southwest of this intersection, it turned south and split into a western and a southern branch. The southern branch (2–3 m wide) turned towards a farmhouse (Site No. 2) and continued south for a distance of c. 400 m to the agricultural plots to the west and north of H. Te’ena. This segment was excavated, revealing that the road was built of two parallel walls (0.5–1.0 m wide) that were preserved up to two courses high, with the better-preserved one on the western margin. The walls were alternately built of medium- or large-sized fieldstones with mud in between. The road turned west at its southern end, where several probes were excavated (Figs. 6.6.1, 6.6.2, 6.7), revealing that its bedding was made of a layer of fieldstones (10–12 cm) with mud in between (L7085). Persian- and Byzantine-period potsherds were found in the excavation. The western branch (2–3 m wide) first ascended unto a rocky plateau east of Qasr es-Sitt (Site No. 1). Its extension toward Migdal Afeq did not survive but was discerned in ancient maps (see below). 3. H. Te’ena–Dir Balut Road. A two-kilometer long road that began in the northern slopes of H. Te’ena (map ref. 198570/665180; see Chapter 3, Site No. 3), continuing southeast up a moderate hillside where the bedrock was hewn and leveled. From here, the road continued eastward, intersected Road No. 1 and slopped down southeastward, reaching a hill

Fig. 6.6.1. Road No. 2. Plan of the southern section.

Fig. 6.6.2. Road No. 2. Plan of the southern section.

124

Amit Shadman: Chapter 6, Roads and Agricultural Systems

Fig. 6.7. Road No. 2, the southern section (Survey of Israel).

5. H. Tukkim–Deir Balut Road. A one-kilometerlong road that departed east of H. Tukkim (map ref. 199100/663030) to the northeast, intersecting Road No. 1 after 300 m. It continued northeast, south of a small wadi, and after 500 m intersected the agricultural Road 19b (see Chapter 3, H. Ta‘amur) which approached it from the south. Subsequently, after a few hundred meters, it intersected rural Road No. 3. The general direction of the road shows that it connected H. Tukkim with sites located to its east and northeast.

some parts reaching 7 m. Maps from the 1940s (Fig. 6.8) show the road continued north to Qurant al-Haramiya and Migdal Afeq, but no evidence for this was found during excavations at the site. 4. Migdal Afeq–H. Yeqavim Road. A three-kilometerslong road, first discovered by Kochavi and Beit Arieh (1994, Sites Nos. 193, 168), beginning at Migdal Afeq on the west and ending southwest of H. Yeqavim (Fig. 6.9). The road (width 3–6 m) was built of medium- and large-sized stones and was preserved to a max. height of 1.0–1.5 m. Field towers, stone clearance heaps and agricultural terraces were discerned along the route. The road’s starting point at Migdal Afeq is unclear, but according to the topography, it apparently continued southeast to the northern bank of Naḥal Shiloh, and then east to the opening of one of its tributaries. At map ref. 197750/664600, the road passed north of a farmhouse (Site No. 17) and continued eastward along a gentle slope on the southern bank of the wadi. At map ref. 198200/664420, the road ascended on a steep slope toward a ridge where it split into two branches. The southern branch reached a farmhouse (site No. 21) located in a tributary of Naḥal Shiloh, and the main route continued on a plateau, reaching H. Yeqavim.

6. Kh. Umm al-Hammam Road–Road No.2. A twokilometer-long road (Fig. 6.10) that connected the longitudinal Roads No. 2 (the southern branch) and No. 7. The road (width c. 4 m) was built of medium- and large-sized fieldstones to a height of one or two courses. At certain spots, its margins were built of double walls. Several sites, including H. Te’ena, Naḥal Shiloh, and Naḥal Shiloh (West) were located near its western end, while its eastern end was on a plateau east of Kh. Umm al-Hammam. The eastern section of the road intersected Road No. 1, where a farmhouse (Site No. 4) was located. Here, a subsidiary northward-branch of the road was 125

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods

Fig. 6.8. Map of the area in the 1940s.

Fig. 6.9. Road No. 4. Migdal Afeq - H. Yeqavim Road. Looking west (A. Shadman).

126

Amit Shadman: Chapter 6, Roads and Agricultural Systems

Fig. 6.10. Road No. 6. Kh. Umm al-Hammam—Road No. 2 (Sky View, A. Shadman).

8. Wadi al-Bureid–Road No. 1 Road. This twokilometers-long road ran along the southern bank of Wadi al-Bureid. Due to its poor state of preservation, it could only be traced through aerial photographs. A segment was discovered by Kochavi and Beit Arieh (1994: Site No. 125), who reported that some sections of the road were supported by a wall built of mediumand large-sized fieldstones. Based on the topography of the area and the orientation of the preserved sections, the road seemingly connected both Kh. Kesfa and Rafat with Road No. 1.

discerned, next to field towers and agricultural terraces. The road continued eastward passing two dammed wadis until it descended into a tributary of Wadi alBureid, where it intersected rural Road No. 7 (see below), and continued southeast of Kh. Umm al-Hammam and north of a wadi built with agricultural terraces, ending at map ref. 199900/66540. Its easternmost section is not evident, but according to aerial photographs, it seems to have continued toward a large wadi (see Road No. 8), possibly reaching Rafat. 7. Wadi al-Bureid–H. Yeqavim Road. This road was constructed in a wadi descending toward Wadi alBureid (Fig. 6.11). It is generally oriented north–south and, presumably, mainly served to connect between Kh. Umm al-Hammam and H. Yeqavim, possibly continuing southeastward to Deir Balut. The road originated in Wadi al-Bureid (map ref. 199125/666060), from where it continued southward, turning east after 200 m between two hills. Further down, it turned south again and climbed through the wadi to the west of Kh. Umm al-Hammam (width 5 m at this point). At map ref. 199125/666060, the road intersected Road No. 6, continuing southward and ascending in curves through the wadi. Its next section was not preserved, but in an aerial photograph, it reached a large enclosed plot northeast of H. Yeqavim.

9. Kh. Umm al-Hammam–Deir Balut Road. This 1.5 km-long road originated east of Elevation Point 185, at the intersection with Road No. 6. It was generally oriented northwest–southeast, running on a slightly sloping plateau. The road ran southeastward and passed between cultivated areas that belonged to the settlement of Kh. Umm al-Hammam. At a distance of about another 800 m, the road turned southward to enclosed plots that apparently belonged to the settlement of Deir Balut. The road’s presence attests to a direct connection between Deir Balut and Kh. alHammam, at least during the Byzantine period, when the latter was first occupied.

127

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods

Fig. 6.11. Road No. 7. Wadi al-Bureid – H. Yeqavim. Looking north (A. Shadman).

2. A considerable number of roads led to settlements occupied in the Roman and Byzantine periods (Kochavi 1972: 26–27). 3. Potsherds dating from these periods were found in most excavated rural roads (Dar and Safrai 1981: 376– 382), although earlier sherds were also occasionally found (Dar, Tepper and Safrai 1986: 96). 4. An organized and complicated road network attests to a cooperative society, such as the Jewish community of these periods (the Mishnah and Talmud periods), and not to a sporadic occupation. Although it may be argued that such rural road system could have been established by the authorities that ruled the Land of Israel after the Byzantine period, there is no historical basis for this assumption (Safrai 1990a: 161).

10. Kh. Ta‘amur–Deir Balut Road. This 1.5 km-long road connected Kh. Ta‘amur with nearby Deir Balut. The road is poorly preserved, making it difficult to trace its remains. It would seem that the road originated east of Kh. Ta‘amur and continued northeastwards on the western curved bank of Naḥal Shiloh. After one kilometer it turned north toward Road No. 3 where it joined Road No. 5, leading together to Deir Balut. Dating the Rural Roads International and regional roads can be dated based on historical sources, such as the accounts of the traveler from Bordeaux (Ish-Shalom 1966) and Eusebius (Melamed 1950). However, rural roads are usually not cited, and milestones, which may also provide dating data, are hardly found along these roads (Tsafrir 1984: 51; Tepper 2013: 14). Since excavations carried out in rural roads in our area yielded only a few worn potsherds dating from several periods, their construction date is unclear. Therefore, in dating the rural roads, one should rely on archaeological evidence from sites that used these roads.

Most rural roads surveyed in the present study apparently led to settlements founded in the Byzantine period (see Chapter 3, Kh. Kesfa, Kh. Umm al-Hammam, H. Yeqavim and Kh. Ta‘amur), and that some continued to exist until the Early Islamic period. No roads leading to settlements dated earlier than the Byzantine period were discerned, except for Migdal Afeq, which was occupied through almost all historical periods. It should also be noted that only five Roman sites were discovered in the area, as opposed to 20 Byzantine sites. Therefore, it is unlikely that a network of ten rural roads served the few Roman settlements. Nevertheless, the sites of Rafat and Deir Balut, located east of the study area, were approached by three rural roads coming from Migdal Afeq, H. Tukkim and H. Te’ena (Roads Nos. 2, 5 and 3, respectively).

It is generally accepted the rural roads should be dated to the Roman and Byzantine periods, and this for several reasons: 1. The distribution of settlements and their economic status significantly increased throughout the Land of Israel during these periods (Safrai 1986a: 43; Tsafrir 1996: 269; Bar 2001: 70–71). 128

Amit Shadman: Chapter 6, Roads and Agricultural Systems The survey at Rafat and Deir Balut showed that these sites were occupied in Iron Age II, and in the Byzantine, Middle Ages and Ottoman periods (Kochavi 1972: 231). Since most sites in our region date from the Byzantine period, it can be assumed that the roads were mostly used during this period, although the multiperiod sites at Rafat and Deir Balut difficult arriving at this conclusion.

this period. This suggestion is supported by studies conducted at the sites of Deir Sama‘an (Magen 2008a: 185–189; 2012: 9–106) and Deir Qal‘ah (Magen 2008a: 189–194; Magen and Aizik 2012: 107–156), east and southeast of Deir Balut, where extensive occupational activities had occurred in the Roman, Byzantine and Early Islamic periods. Road No. 4. This road ran between H. Yeqavim on the east and Migdal Afeq on the west. While the latter was occupied almost through all historical periods, the first was established in the Byzantine period and was active until the Early Islamic period. It can thus be concluded that the road was built in the Byzantine period. However, it should be considered that this road was also a main connecting route between Migdal Afeq and Deir Balut prior to this period. Thus, the area southwest of H. Yeqavim must also be taken into account as it includes enclosed plots that clearly appear to be later than the nearby Road No. 4. In light of the importance of Migdal Afeq and Deir Balut, it is assumed that they were interconnected by a road most likely built during the Roman period. Only later, during the Byzantine period, were the enclosed plots of H. Yeqavim built along the road.

The evidence for dating nine roads is as follows:4 Road No. 1. This road was connected to the Gopnha– Antipatris road (Safrai 1995c: 41). Excavations (Sion, ‘Ad and Haiman 2007: 110–111) yielded no dating evidence, but according to historical references, this road was constructed in the Roman period and continued at least until the Byzantine period (Finkelstein 1977). Therefore, it seems that Road No. 1 was already used in the Roman period and reached its activity peak in the Byzantine period. The fact that it crossed the territories of Kh. Ta‘amur, H. Yeqavim and Kh. Dayyar, which were established during this period, emphasizes its main period of use. Road No. 2. This road connected the area of Kh. Kesfa and Kh. Umm al-Bureid, founded in the Byzantine period, to the settlements to its east. Excavations in the western sections of this road yielded potsherds from the Persian and Byzantine periods. A Persian farmhouse (Site No. 2) was situated near the road, but the excavations showed that these two features were not interconnected. Therefore, the road should be dated based on other sites in the area. The sites on both ends of this road—Rafat on the east and Migdal Afeq on the west—were occupied in several periods. However, the sites along the road (except Site No. 2): Kh. Kesfa, Kh. Umm al-Bureid, and H. Te’ena were established in the Byzantine period, presumably the time when the road was also constructed.

Road No. 5. This road connected H. Tukkim to the area to its east. This site was first occupied in the Byzantine period and continued into the Early Islamic period. The end of the road was not clearly identified, but it is evident that it reached the plots of Kh. Ta‘amur and H. Yeqavim, continuing towards Deir Balut. Since no potsherds dated earlier than the Byzantine period were discovered in H. Tukkim, it would seem the road was probably not founded prior to this period. The road might have been constructed during the Early Islamic period, but this assumption is less likely. Road No. 6. This route connected Kh. Umm al-Hammam (Site No. 11) to the central longitudinal Road No. 1 that crossed the study area from south to north. Kh. Umm al-Hammam was founded in the Byzantine period and continued into the Early Islamic period. Presumably, the road was founded in the Byzantine period and was also used in the following period.

Road No. 3. This road runs from Deir Balut to H. Te’ena. The excavations we conducted in H. Te’ena indicated that the site was founded in the Byzantine period (5th century CE) and continued into the Early Islamic period (7th century CE). Therefore, it is obviously unlikely a road leading to H. Te’ena would be earlier than the Byzantine period. Although Deir Balut was occupied in the Iron Age and in the Roman, Byzantine, Middle Ages and Ottoman periods (Kochavi 1972: 231), it is clear that there was no road between Deir Balut and H. Te’ena during the Iron Age, as the latter was not occupied then. Although there are references to roads in the Bible, surveys and excavations showed most were developed in the Roman period (Avi-Yonah 1954: 712–713). Two Byzantine settlements existed on both ends of the road, and it seems that the road was used mainly during 4 

Road No. 7. This road connected Kh. al-Hammam to H. Yeqavim, 1.5 km to the south. These two sites were first occupied in the Byzantine period and therefore this is probably the date of construction of the road used also in the Early Islamic period until the site was abandoned. Route 9. This road connected the sites of Deir Balut and Umm al-Hammam. As the latter was not occupied before the Byzantine period, there was no reason to construct the road prior to this period. Thus, it is suggested that the road was built in the Byzantine period and was probably used also in the Early Islamic

There are no sufficient data for dating Road No. 8.

129

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods period. It is plausible that this road was continuously used also in later periods, since both Deir Balut and Kh. Umm al-Hammam (on a hill above the Byzantine site) were occupied in the Middle Ages and the Ottoman period (Site No. 11; Kochavi 1972: 231).

households (Safrai 1990b: 110; Safrai 1994b: 224). Thus, it is clear the economy of the rural settlements was depended to a certain extent on the road network that served them (Dar 1982a: 225). Studies carried out in Italy, for example, showed that, in the Roman period, there was a correlation between the density of farms in a certain area and the level of the road network, and that the number of farms increased the more this network was developed (White 1970: 386; Dilke 1971: 88–89, 93, 134–158). Moreover, the study of Roman roads shows that they directly influenced the way the Empire and its colonies were administered (Chevallier 1976: 202–209).

Road No. 10. This road connected Deir Balut to Kh. Ta‘amur. The latter was first occupied in the Byzantine period and ceased to exist during the same period. Thus, this road was constructed during this time. It should be mentioned that a few Roman potsherds were found in Kh. Ta‘amur, but these were too meager to indicate a settlement requiring the construction of a road. Rural Roads and Field Towers

Transporting agricultural products and tax collection were efficiently carried out thanks to the developed road network. The emperors, among them Julius Caesar and Augustus, also understood the importance of agricultural roads and plots adjacent to them. These were measured by the method known as ‘limitatio’, which determined their boundaries, especially in rural areas. Julius Caesar determined that the width of an agricultural road should be about 1.5 m, while Augustus determined the standard dimensions of the Cardo and Decumanus, as well as the secondary roads.

During the excavations and the surveys, an interrelation was discerned between the rural roads and the field towers. Dar (1982: 213–216, photo 64) noticed this connection during his study of rural roads in the Western Samaria Hills. During our survey, field towers were discerned throughout the research area, in many cases located on the margins of the roads or adjacent to them. Several field towers were identified along Road No. 1, mainly between H. Te’ena on the south and Kh. Umm al-Bureid on the north. The field towers were located west and east of the road, on the banks of Wadi Susi, where it ran between farming terraces. A similar picture was observed in some sections along Road No. 2, between Kh. Kesfa and Kh. Umm al-Bureid, where several field towers and winepresses were discerned. Additional field towers were excavated east and west of this road running between Wadi al-Buried (Site No. 2) and H. Te‘ena.

Our study area is characterized by a developed network of rural and agricultural roads where seemingly a routine and efficient trade was conducted. The vast number of agricultural terraces, winepresses, and oil presses indicates that these were not used only for domestic purposes, and it is clear that surplus agricultural products were marketed to neighboring communities and perhaps even beyond. Thus, in the absence of a road network, only small villages and farmhouses could exist, but certainly not a multilateral marketing system with the neighboring settlements (Clark and Haswell 1967: 179–199). Another factor that impacted the marketing of crops is the Mediterranean warm climate, because of which the transit time of agricultural crops from the fields to the markets had to be significantly shortened (Dar 1982a: 227). Fresh crops such as grapes and figs had to be collected as quickly as possible, and Taaffe and Gauthier (1973: 34–36) argued that the development of certain species resulted from their transporting feasibility along the rural roads.

Other field towers were excavated along the western branch of this road, where a cluster of many winepresses was found. An interesting picture emerged along the northern branches of Road No. 6 (Kh. Umm alHammam–Road No. 2), where a large area comprising field towers and winepresses was surveyed. Excavation of the roads and the field towers showed that the two features were interrelated, however, the roads did not actually approach the towers, and no entrances from the roads into the towers were found. Presumably, the entrances were at a higher level than the margins of the roads. Rural Roads and their Contribution to the Development of the Rural Countryside

Rural Roads and Transportation Means

Agricultural products were the main economic resource of the rural countryside. The crops were first transported from the fields to storage facilities (field towers, farmhouses) and then to the markets, whose profitability depended on the roads leading to them (Frayn 1979: 27). Obviously, some part of the crops was for domestic consumption, i.e., of the farmers’

The use of wagons as a mean of transportation was common in the Byzantine period when most roads in the region had been built. This is evidenced in the 5th century CE Theodosius Codex (VII, 5, 8, 17, 28, 30, 47), which examined the weight permitted for wagons (Chevallier 1976: 180). However, it seems that most

130

Amit Shadman: Chapter 6, Roads and Agricultural Systems of the traffic between settlements, especially in rural areas, was done with by donkeys or camels and without a towed wagon (Safrai 1994b: 234; Dar 1982a: 234).

The road network reflects organized communities that maintained advanced social and commercial relations and operated independently, providing all their needs. Observing the road network, it would seem that, at least during the Byzantine period, several settlement clusters can be defined. On the west—the central settlement of Migdal Afeq and its surroundings, which controlled the Via Maris and the Antipatris–Gophna road; in the middle of the region— Kh. Ta‘amur, H. Yeqavim, Kh. Umm al-Hammam and H. Kesfa; on the east—Deir Balut and Rafat. Nevertheless, it is clear that the associated area was controlled by a large governmental center located outside the immediate area.

Donkeys, mules, and camels were the main transportation means, appearing also as terms in Jewish rabbinical sources which describe convoys between various villages (Safrai 1990b: 114; Avital 2014: 68, Figs 144–145). Donkeys have been used since the beginning of history as a beast of burden that can easily carry about 75 kg; a mule can walk up to 60 kilometers daily and carry 150 kg, while a camel can also travel tens of kilometers daily and carry about 300 kg or more (Avitsur 1968: 5–14). These figures show the importance of these animals in ancient trade. The construction of rural roads in the area required considerable investment and planning, but perhaps most were plausibly not intended for wagons. The roads examined during our research were founded on rocky terrain, with rough surfaces, and often in steep slopes which do not always allow moving a beast of burden and a wagon.

Three main roads surrounded the study area from the north, west, and south, and led to various governmental centers. The rural roads were linked to the regional roads in several places: In the north, rural Road No. 1 connected to the Antipatris–Hares road; in the west, two rural roads (Nos. 2, 4) connected to the Via Maris which passed at the foothill of Migdal Afeq (no connection between Migdal Afeq and the Via Maris was found, but it probably existed); in the south, the southern part of rural Road No. 1 connected to the Antipatris–Gophna road (see also Amit and ‘Ad 1999: 90–91). Based on the rural roads and their connections to the international and regional roads, it appears that the area was apparently associated with the large city of Antipatris, seen as all the main roads passed through it (see Chapter 7).

The Study Area and its Main Rural Roads The study of the rural roads in the study area revealed that they were arranged according to the ‘network model’ (for other models see Safrai 1983b: 270). This model is composed of several settlements (‘rural towns’ in our case) connected by an extensive road network. The regional network indicates that H. Yeqavim (Site No. 9) was the most interconnected among the neighboring settlements. This site was directly connected to Kh. Ta‘amur (Site No. 19), Kh. Umm al-Hammam (Site No. 11), H. Tukkim (Site No. 18), Migdal Afeq (Site No. 14), H. Te’ena (Site No. 3) and Deir Balut. The second most connected site was Kh. Umm al-Hammam. Although only two roads connected it to other large settlements in the area (H. Yeqavim and Deir Balut), it was well linked to the longitudinal routes of the area. Based on the road network, Kh. Ta‘amur seemingly had also commercial and other relations with Deir Balut and H. Yeqavim. Safrai (1997a: 233) argued that, in the Byzantine period, Kh. Ta‘amur was a satellite village of Deir Balut.

Summary Ten rural roads were traced in the study area, totaling 30 km in legth. The main route was Road No. 1, running south–north, with additional roads connected to it: H. Tukkim–Deir Balut (No. 5), Midgal Afeq–H. Yeqavim (No. 4), H. Te’ena–Deir Balut (No. 3), a route northeast of H. Te’ena (No. 6), and a route at the junction of Wadi Susi and Wadi al-Bureid (No. 2). In the lack of evidence, Road No. 8 can only be assumed to have connected to Road No. 1 near Wadi al-Bureid. The connection of Road No. 1 to the regional latitudinal roads north and south of the study area suggests that the agricultural products were transported via this road to the main cities (see below). Two latitudinal roads led to Antipatris, the largest city in southern Sharon Plain, and to the Samaria Hills and the main roads there. In addition, latitudinal rural roads leading to Rafat and Deir Balut were discerned on the east, as well as the road leading to Migdal Afeq on the west.

Another satellite village was apparently at H. Tukkim, which was connected by two roads to Kh. Ta‘amur, H. Yeqavim, and Deir Balut. Special ties existed between Deir Balut and H. Te’ ena, which were connected by a rural road. Although the latter was occupied only by a small monastery during the Byzantine period, the presence of the road indicates that it had a significant status in the eyes of the residents of Deir Balut. The other Byzantine sites and those from earlier periods were connected (not always) to the road network via one route which was often remote from the sites themselves, e.g., Kh. Kesfa, which overlooked the Rafat–Migdal Afeq road.

The roads were carefully designed in accordance with the topography of the area. In most cases, they were founded on moderate ridges or in wide 131

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods wadis. In case of steep slopes (e.g., the ascent of Naḥal Shiloh), the roads were constructed using sharp curves to overcome the obstacle and ease the route.

the western and eastern settlement strips. The farmers commuted daily from these large settlements (Migdal Afeq, Deir Balut, and Rafat) to their fields, and it is therefore likely that the rural roads (mainly Nos. 1, 2, 4) were constructed for this purpose. Due to the good road network that developed during this period, the residents could seemingly commute and transport agricultural products from the fields to local markets. During the Byzantine period, when the dramatic increase in the number of settlements occurred, most roads visible at present were constructed or renovated. The largest sites in the area, including Kh. Kesfa, Kh. Umm al-Hammam, H. Yeqavim and Kh. Ta‘amur, were interconnected by a complex network of carefully constructed rural roads. These settlements and the monasteries in their vicinity were founded during the Byzantine period and continued to exist in the Early Islamic period.

Rural roads were constructed without mortar. Based on their nature and quality, they seem to have served mainly pedestrians and loaded animals. A great investment was observed in the construction of the roads’ curbstones, which were built up to two courses high. Frequently, the roads passed between privately owned enclosed plots, and the built-up curbstones prevented deviation, either deliberate or not, from the road. It should also be noted that roads, and their margins, apparently had a legal status intended to define jurisdictions and the roads’ rules of use (HarpazSegal 2012: 219). Presumably, rural roads, in contrast to agricultural roads, were supervised by the local government or its agencies.

Agricultural Installations

Various installations were identified and excavated along the roads. Field towers were used to store agricultural products until being sold and were therefore preferably located close to the rural and agricultural roads. Winepresses were discovered next to the towers, as the grapes had to be quickly transferred to a cool, shady place. On the other hand, oil presses were not found near the roads but rather inside the settlements. Water cisterns were scattered along the roads, but following no specific distribution pattern (see below). The agricultural-economic development of the rural settlements in the periods under discussion depended on its road network. The ties between large communities created an open and diverse market, and the authorities/communities maintained the roads and the basic needs of the users. Clear rules regarding the roads and their measurements were established by the rabbinical sources of the Mishna and the Talmud. In many cases, these correspond to the archaeological finds in the study area (see Krauss 1924: 114–213; Safrai 1965: 16–119; Dilke 1971: 93; Safrai 1990a).

Introduction Agriculture was the main source of income in antiquity, and naturally, the environmental conditions largely determined the agricultural crops raised by the farmers (Dar 1986c: 142). Thus, land quality was an important component of the entire rural system (Garnsey and Saller 1987: 44; Pastor 1997: 1), and the Land of Israel was not blessed with high-quality lands (Bar 2008: 105) and water sources. Suitable areas for agriculture can be found in the valleys and wadis where alluvial soil was accumulated. Governing the land, and its occupation, generated severe struggles among people, resulting in legal measures taken to maintain the local social order (Safrai 1998: 37). For example, the Feudal Law—the Colonatus system—was introduced in the Land of Israel at the end of the 4th century and became an important law during the Byzantine period (Dan 1990b: 188–190; Sion 2004: 193, see also Chapter 10 below). This law prevented vassals from freely moving out of their living places.

Dating the roads was one goal of the present study. Excavating selected parts of a particular road should apparently provide information regarding its architectural and chronological aspects, but while the excavations provided us with clear information about the building methods of the roads, they failed to date them. The many excavated probes yielded worn and eroded potsherds from a variety of periods, and no specific date could be inferred from these. We have suggested dating the roads based on the sites they served, but here, too, no decisive date was obtained. Unsurprisingly, not a single road approached a Hellenistic farm, since it is unlikely to build and maintain a road leading to a small isolated farmhouse.

With the rising population growth, especially during the Byzantine period (Broshi 1982: 442–455, see also Chapter 9), the demand for agricultural areas significantly increased (Rubin 1990: 16–17; Tsafrir 1996), and lacking other alternatives, mountainous regions were also cultivated (Dagan 2006: 21). The dramatic increase in the number of agricultural terraces in the Land of Israel indicates that a skilled population with impressive engineering abilities built these systems. The farmers had to carefully select the areas where terraces would be built, based on their stability in supporting the soil behind them. The terraces and the dams comprised three main components: 1. The fence—a retaining wall between one terrace and another. These walls were built of fieldstones with no mortar, allowing for water seepage and preventing

During the Roman period, the population of the region lived in large settlements west and east of 132

Amit Shadman: Chapter 6, Roads and Agricultural Systems potential pressure from the accumulated soil behind them. The width and height of the fence depended on the cultivated area and the incline on which it was located.

bedrock, seldom on a thin layer of alluvial soil. Terraces of this type were constructed wherever possible and were discerned in most of the study area. Terraces were not discovered on the steep banks of wadis, where agriculture could not be maintained.

2. The terrace’s layer—most of the area of the terrace used for cultivation.

Dams

3. The terrace’s height—“the short cliff that separates between layers” (Ron 1977b: 210–211).

A massive wall usually built within a riverbed or wide gully (Fig. 6.13). The dams were built as a double wall (max. width 2 m) and had several courses (max. height 0.5–1.0 m). They were constructed of fieldstones of various sizes without mortar. The dams’ face was tilted in accordance with the natural slope of the area. Dams in steep wadis were built at a sharp incline of about 90°, while on the more moderate wadis they decreased to an incline of about 45°–70°. The length of the dams varied from tens to hundreds of meters, depending on the width of the gully. Frequently, the dams were built in a curved shape to reduce pressure on the wall.

The locations for building agricultural terraces also depended on their degree of exposure to sunlight, which affected soil moisture. In his study conducted in Jerusalem hills, Ron (1977b: 214–215) found that the slopes facing south are dryer than those facing north. However, no orientation preference was noted in our region for terraces. Thus, these created leveled and convenient surfaces for cultivation (probably also cleared of stones: De Geus 1975: 67) which prevented erosion. Rainwater draining during winter was seemingly also addressed in advance (see below).

It should also be noted that dams were not discerned in large streams such as Naḥal Rabah, Wadi al-Bureid and Naḥal Shiloh, apparently due to the intensive winter flow. Even if dams existed in these streams, they were most likely swept away over the years. Figure 6.14 shows that almost no dams were traced in streams with large quantities of runoff water (dark blue). Approximately 1,000 dunams of dammed wadis are estimated to have existed in the study area, which extends over 30 sq km (Fig. 6.15).

Soil availability was also considered by the farmers, and wherever a soil shortage existed, especially in the bare mountains, soil was brought from nearby areas or preserved in its natural spot. Farmers were already aware, in antiquity, of the fact that the quality of the soil adjacent to settlements was higher (due to nitrogen and organic materials, Dagan 2006). For example, studies conducted in the vicinity of Jericho showed that soil from the ancient tell was transported to nearby fields to fertilize them, apparently during the Byzantine period (Kenyon 1957a: 45).

Enclosed Plots

A similar phenomenon was noted in Egypt (Gardiner 1961: 207–208). Thus, the agricultural terraces were part of extensive agricultural systems that characterized most of the central mountainous regions in the Land of Israel and in the Mediterraneanbasin mountains (Gibson and Edelstein 1985: 143–144; Gibson 2001b: 113–146; Treacy and Denevan 1994: 91–110; Rubin 1998: 68–74; Haiman 2007: 131–146; 2012a: 403–425; 2012b: 1–7). In the present study, we distinguish between agricultural terraces and dams (see below).

These plots were found in the vicinity of Kh. Kesfa, H. Yeqavim and H. Ta‘ana, occupied mainly during the Byzantine period. Surprisingly, no such plots were delineated in Kh. Umm al-Hammam. The plots were built of medium- and large-sized fieldstones, usually in straight lines. In many cases, there were built double walls, over 1 m wide and up to two courses high. The plots were planned mainly to define their area and ownership. This construction method attests to an organized society that took care of the plots and their access to the agricultural roads, and especially to the rural roads, which extended between the settlements. Marking plots had been done before the Byzantine period, for example, in the Roman period, when plots were measured and marked by surveyors who operated on behalf of the army.

Agricultural Terraces Two main types of farming terraces were discerned. The first type, massive and similar to dams (Fig. 6.12, see below), was located on wide mountain slopes and carried a large amount of soil. Many of these terraces were in fact a double wall (width 1–2 m) serving as an agricultural road or a passage between plots. Such terraces were identified and excavated east of Migdal Afeq and east of H. Te’ena. The second type of farming terraces was constructed in a simpler technique and it usually composed of a wall founded on terraced

The engineers attached to the Roman legions visited various provinces throughout the empire and assisted in the planning of cities, roads, and other (Safrai and Fixler 1993: 215). Measuring the areas was done using the Centuria method that created equal squared plots. This division was noted mainly in the area of H. Yeqavim, whereas in other places the shape 133

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods

Fig. 6.12. A dammed gully and a terraced slope west of H. Yeqavim (map ref. 198730-664100).

Fig. 6.13. A dam between H. Te’ena and H. Yeqavim (map ref. 198860-664740; A. Shadman).

134

Amit Shadman: Chapter 6, Roads and Agricultural Systems

Fig. 6.14. The drainage basin of the region’s streams.

135

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods

Fig. 6.15. Dammed wadies.

136

Amit Shadman: Chapter 6, Roads and Agricultural Systems and size of the plots varied but were almost always designed in symmetrical straight lines. For example, around Kh. Kesfa, the plots’ areas ranged from a few to dozens of dunams. Differences in the sizes of the plots were evident also in Europe during the classical period. Plots measuring 20–40 dunams were documented in classical Greece, whereas in Rome they measured approximately 12.5–25 dunams (Garnsey 1988: 46).

In recent years, another dating method has been introduced: OSL (Optically Stimulated Luminescence), which measures the accumulated signal in quartz crystals (Davidovich et al. 2011: 481). Several agricultural terraces in the study area were dated using this method (see below). Many probes were excavated in different components of the agricultural systems to trace datable finds. Since in most cases only worn potsherds of various periods were uncovered, it appears that dating agricultural systems should be based mainly on the dates of the neighboring settlements and the rural and agricultural roads. For example, Finkelstein (1981: 338–339) approached this issue by examining the relationship between agricultural systems and the Roman Antipatris–Gophna road. Obviously, it is not always possible to rely on the road network and/ or the surrounding sites, e.g., no roads leading to the Hellenistic farms were discovered, making it impossible to use this method for dating them. Similarly, the almost total absence of Roman settlements in the study area does not allow dating the agricultural systems based on the settlement.

Dividing Walls and Enclosures Dividing walls were used to mark boundaries between plot clusters or between village territories. Unlike enclosed plots, dividing walls were always built as double-walls and considerable effort was put into their construction, which included slightly worked stones. The length of these walls varied from tens to hundreds of meters, as discerned north and east of H. Te’ena, where a wall was apparently built to separate the monastery from its surrounding area. Another characteristic of these walls is the lack of a clear line, sometimes following the natural topography. Dating the Agricultural Systems

In examining the dates for the agricultural systems, we refer to the four ‘settlement strips’ that comprised the area. ‘The western strip,’ which includes the sites of Migdal Afeq and Qurnat al-Haramiya, occupied from the Iron Age through the Ottoman period. ‘The farmhouses strip,’ settled from Iron Age II to the Hellenistic period, coexisted with ‘the western strip’ while its inhabitants cultivated the nearby fields (within a half-an-hour walking distance). It seems, therefore, that the agricultural systems in these two strips were probably constructed in Iron Age II and also operated in the Persian period. Unfortunately, these systems were heavily damaged and are not visible today. During the Hellenistic period, agricultural terraces were built mainly in the farmhouses strip, adjacent to the farms and their surroundings, some of which are even visible today. Detecting the terraces was possible only through the use of OSL testing, carried out near the farmhouse at Site No. 5, active from Iron Age II through the Early Hellenistic period (Fig. 6.16; Table 6.1). The test revealed that the lower part of the terrace was probably built during the Hellenistic period (3rd– 2nd centuries BCE) and underwent changes, during the 2nd century CE, which included the removal of soil and renovations. Excavating the terrace, however, yielded worn potsherds dating from various periods.

As mentioned in Chapter 5, it is difficult to date agricultural installations used in several periods, especially those located in rocky areas like the Land of Israel’s central mountainous region. As noted, the first major occupation of the area took place in Iron Age II and reached its peak in the Byzantine period. Thus, in the present study, it is argued that the agricultural systems (or some of them) were initially built in Iron Age II. Agricultural systems from this period were found in various areas in the Land of Israel (e.g., Dar 1980: 97–100; 1986b: 36; Gibson 1995: 165–175; 2001; Paz, Mizrachi and Grossman 2015: 91–98; Gadot et al. 2015: 118–142; Gadot et al. 2016: 397–417). Despite the above, in many regions in the Land of Israel, a different picture seems to emerge, tending, quite often, to date the agricultural systems to the Late Roman and Byzantine periods (Bar 2008: 113–114). These dates are based on the significant increase in settlement distribution during these periods and the development of agro-mechanical capabilities (Feliks, 1982: 439–440) which naturally caused a shortage in agricultural lands (Tsafrir 1996). For example, Dagan (2006: 21–22) suggested that, considering the large number of Byzantine sites in the Amazya Map, in the Judean Shephelah (more than 1500 sites—Dagan 2006: 35), the agricultural systems in this region were built in the same period. The agricultural systems in the Negev Highland were dated to the Byzantine and Early Islamic periods, as were the farmhouses near them (Haiman 1986: 11–33). A similar dating was also proposed in other regions of the country (Dan 1984: 13; Tsafrir 1984: 347).

During the Roman period, and compared to the previous one, there was a dramatic decline in the number of settlements. Most of the open areas in the western part of the region were apparently used as agricultural systems by the residents of Midgal Afeq and Quarnat el-Haramiya of ‘the western strip.’ The eastern part of the area apparently served the inhabitants of 137

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods

Fig. 6.16. OSL tests. Above: A probe northeast of Midgal Afeq. Below: A probe near a farmhouse at Site No. 5.

138

Amit Shadman: Chapter 6, Roads and Agricultural Systems Table 6.1. OSL tests results. Lab code

ROH-1

ROH-2

ROH-3

ROH-4

ROH-5

0.27

0.57

0.83

0.35

0.73

Water

(%)

17

16

19

17

16

K

(%)

0.61

0.67

0.77

0.67

0.83

U

(ppm)

1.75

1.7

1.93

1.75

1.78

Th

(ppm)

9.1

9.1

10.5

9

9.6

Ext. α

(μGy/a)

10

10

11

9

10

Ext. β

(μGy/a)

730

776

863

764

881

Ext. γ

(μGy/a)

651

673

749

660

726

Cosmic

(μGy/a)

233

205

192

224

194

Total dose (μGy/a)

1624±62

1664±64

1814±73

1657±64

1812±70

No.

Aliquots

18/18

18/18

17/17

18/18

15/18

OD

(%)

23

12

7

8

15

De

(Gy)

1.29±0.33

3.12±0.42

3.11±0.62

2.96±0.27

3.85±0.37

Age

(years 2015)

770±50

1860±90

1710±80

1780±80

2120±100

ROH-1

ROH-2

ROH-3

ROH-4

ROH-5

b.

Lab code

IAII farm house

(m) 

Qaser el Sit - E

Depth

0.27

0.57

0.83

0.35

0.73

Water

(%)

17

16

19

17

16

K

(%)

0.61

0.67

0.77

0.67

0.83

U

(ppm)

1.75

1.7

1.93

1.75

1.78

Th

(ppm)

9.1

9.1

10.5

9

9.6

Ext. α

(μGy/a)

10

10

11

9

10

Ext. β

(μGy/a)

730

776

863

764

881

Ext. γ

(μGy/a)

651

673

749

660

726

Cosmic

(μGy/a)

233

205

192

224

194

Total dose (μGy/a)

1624±62

1664±64

1814±73

1657±64

1812±70

No.

Aliquots

18/18

18/18

17/17

18/18

15/18

OD

(%)

23

12

7

8

15

De

(Gy)

1.29±0.33

3.12±0.42

3.11±0.62

2.96±0.27

3.85±0.37

Age

(years 2015)

770±50

1860±90

1710±80

1780±80

2120±100

b.

‘the eastern strip of settlements’—Deir Balut and Rafat (see below). Although Roman potsherds were collected in the agricultural systems, they do not necessarily date them. However, due to the lack of settlements from this period, it was difficult to assess which of the systems should be attributed to this period. This issue was addressed by doing OSL tests c. 2 km east and northeast of Midgal Afeq. The results showed that these two agricultural terraces were built during the 1st and 2nd centuries CE, meaning that terraces were built also in the Roman period, some from scratch in this period and others based on earlier Hellenistic ones.

IAII farm house

(m) 

Qaser el Sit - E

Depth

Two key sites were in the study area during the Byzantine period: Migdal Afeq on the one hand, and H. Te’ena, to the east of ‘the farm strip, on the other. Large plots to the west and east of H. Te’ena, some of them excavated, were, in fact, part of an extensive system associated with this site (evidenced by the rural and agricultural roads that connect the site with its fields). The excavations at H. Te’ena showed that the site was founded in the 5th century CE and reached its peak in the 6th century CE. Therefore, it is most likely that many agricultural systems in ‘the western strip’ were actually associated with H. Te’ena and are 139

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods thus to be dated to the Byzantine period. However, it is also plausible these agricultural systems were based on earlier systems dated to Iron Age II through the Hellenistic period. Finally, it is possible to deduce from the road network and its dating that Deir Balut and Rafat were occupied during the Byzantine period, including their neighboring developed agricultural systems.

(CAM) after removing distinct outliers. All samples show recycling ratios within 6% of 1.0 for most aliquots and negligible IR depletion ratios. Alpha, beta and gamma dose rates were calculated from the radioactive elements measured by ICP MS (U&Th) or ICP-AES (K). Cosmic dose rates were estimated from the current burial depths. OD – Overdispersion. Aliquots used – the number of aliquots used for the average De out of the aliquots measured.

OSL summary of method – Noami Porath

The Plots’ Outline as a Dating Marker (Fig. 6.17)

88-125 μm quartz was extracted and purified by wetsieving to the selected grain size, dissolving carbonates by 8% HCl, removing heavy minerals and most feldspars by magnetic separation, and dissolving the remaining feldspars and etching the quartz with 40% HF (for 40 min), followed by soaking in 16% HCl to dissolve any fluorides which may have precipitated. Dose recovery tests over a range of preheats showed that a recovery of 95-100% can be obtained using a preheat of 10s @ 260oC , a test dose of ~4.6 Gy and a test dose preheat of 5 s @ 200 oC. Moisture contents are as measured. De was measured on 2 mm aliquots using a modified single aliquot regenerative (SAR) protocol. The average De and errors were calculated using the central age model

It would seem that dating the enclosed plots is more complicated than dating the other agricultural installations in the rural landscape. While collecting potsherds and carrying OSL tests are effective in dating agricultural installations, these methods are not applicable in dating enclosed plots that were often founded on bedrock. Although in Chapter 5 it was suggested to date these plots based on the road network, here it is proposed to date them based on their outline. For this purpose, we examined the difference between the enclosed plots of the Arab villages and those adjacent to the ancient sites. Some Arab farmers still cultivate their fields using traditional techniques,

Fig. 6.17. Enclosed plots. The asymmetrical plots (upper right) belong to modern Deir Balut; the symmetrical plots belong to the Byzantine settlement of H. Yeqavim.

140

Amit Shadman: Chapter 6, Roads and Agricultural Systems and their plots are often not uniform, with rounded, asymmetrical, or amorphic outlines, e.g., in the vicinity of Deir Balut.

treading floors, and occasionally a small channel was cut through which the must flew from the treading floor to the collecting vat.

In his comprehensive study of land administration in the Land of Israel, Granowsky (1949) noted that, in the vicinity of Arab villages, the lands are not unequally divided among the farmers (fellahin). These plots, divided and subdivided, caused unproductive use of the land. In addition, as a side effect, the plots are irregularly shaped and hinder the economy of the area (Granowsky, 1949: 179–180). On the other hand, the Roman agronomist Columaella, in his de re rustica, argued that plots can be of different sizes and shapes, but almost always symmetrical. Indeed, as one proceeds from Deir Balut westward toward H. Yeqavim (a Byzantine site), the plots change their outlines from amorphic to square-shaped. As noted above, it was common in the Roman and Byzantine periods for plots to be symmetrical. Hence, the remarkable differences in the outline of the enclosed plots may serve as a dating indicator, or at least to distinguish between the plots of the traditional Arab village and the ancient ones.

3. Simple winepress with a settling pit The settling pit was usually much smaller than the collecting vat and was mostly rectangular. The settling pit was connected to the collecting vat through a hole cut in their shared wall. 4. Simple winepress with a mosaic These winepresses were usually discovered near small settlements, such as farmhouses or satellite villages. Although it is clear that considerable efforts were put into building these winepresses, they were quite small. The winepresses included a well-hewn treading floor on which an ‘industrial’ white mosaic was laid. The settling pits and the collecting vats were plastered, and a mosaic was placed on their floors. The must flew from the treading floor into the collecting vat through a narrow, hewn channel. Various components were added to these winepresses, e.g., the wall that bordered the treading floor west of H. Te’ena (Fig. 6.20).

The Wine Industry

5. Large winepress with mosaic

Sixty-one winepresses were discerned in the study area, thirty of which were excavated. Generally, the winepresses were found in the agricultural plots of the settlements, and only ten were discovered in or near the settlements. The latter are larger than those found in the open areas, most of these located on bare bedrock and usually bear no traces of plaster (Fig. 6.18, Table 6.2). The winepresses were divided into five main types:

These large winepresses comprised various components: a large treading floor (the largest, at Kh. Kesfa, measured 7 ×7 m) paved with ‘industrial’ white mosaic, sometimes with a cut depression in the middle of the floor to accommodate a perforated stone that housed a pressing device. There were large settling pits and collecting vats adjacent to the winepresses that could store thousands of liters. Based on the large winepress at Kh. Kesfa, these winepresses seem to have been used by large villages with dozens of households.

1. Simple irregular winepress

The treading floors (Fig. 6.21)

These winepresses, many of them excavated, are rather simple, poorly hewn installations, generally comprising irregular treading floors and collecting vats. The treading floor was cut in the sloping bedrock and the must was directly channeled into the collecting vat (Fig. 6.19). These installations were found throughout the Judean and Samaria Hills (e.g., H. Hermeshit—Greenhut 1998: 144–146). Their small dimensions may fit the biblical term ‘pora’ (‫ ;הרופ‬Haggai 2: 16; Isaiah 43: 3). In Akkadian, this term refers to some kind of stone bowl, and possibly was a small, often mobile, installation (Avshalom-Gorni, Getzov and Frankel 2008: 65–67).

The measures presented here were collected by Kochavi and Beit Arieh (1994) and were presented in their Rosh Ha‘ayin Map Survey: Sixteen treading floors measured 1–2 × 1–2 m; fourteen measured 2–3 × 2–3 m; eleven measured 3–4 × 3–4 m; two measured 5 × 5 m and one, larger treading floor measured 7 × 7 m. The collecting vats (Fig. 6.22) Most of the collecting vats were small, as were also other parts of the installations. Ten vats contained less than 500 liters, and 15 additional vats contained up to approximately 2000 liters. Only four winepresses (probably from the Byzantine period) could store approximately 4,500–12,000 liters, compared to 450,000 liters by the winepress discovered at Byzantine H. Zikhrin to the south of the study area (Fisher 1985: 119).

2. Simple asymmetrical winepress Many winepresses of this type were discovered in the study area. These comprise a treading floor and a collecting vat which were usually meticulously hewn in a straight outline. Some winepresses had rounded 141

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods

Fig. 6.18. The distribution of winepresses in the study area.

142

Amit Shadman: Chapter 6, Roads and Agricultural Systems

Fig. 6.19. Four irregular simple winepresses.

The accumulated capacity of the winepresses does not allow calculating the quantity of wine produced in the region over the periods, and this for two main reasons: [1] not all the dimensions of all the collecting vats are known; [2] it is impossible to know how many times the collecting vats were filled in each harvest season. Moreover, since the number of Roman settlements is very small, there is insufficient

information regarding the quantities of wine stored in the winepresses during this period. Similarly, although many winepresses were dated to the Byzantine period, only a few have been excavated, so no reliable estimation of the wine production capacity is available. Thus, since ‘the farmhouses strip’ was meticulously examined, an attempt was made to calculate the wine production capacity during the Hellenistic period. 143

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods

Fig. 6.20. A simple mosaic winepress, west of H. Te‘ena.

About 30 winepresses, half of them surveyed and most of them excavated, were discovered in ‘the farmhouses strip.’ The capacity of wine production measured in half of the winepresses in the area was about 28,855 liters. Assuming that the rest of the winepresses yielded a similar volume, it is estimated that in the Hellenistic period, and perhaps even earlier, the residents of the area were able to store c. 40,000 liters of wine at any given time. In addition, the large sites of Migdal Afeq and Qurnat al-Haramiya and their winepresses should also be taken into account.

horizontally on the rocky surfaces (roglit in Jewish rabbinical sources). Dating the winepresses Dating the winepresses and other installations in mountainous regions is not an easy task. Dar (1982a: 236), who surveyed hundreds of winepresses in the Western Samaria Hills, addressed this issue and noted that the different components of the winepresses were used over long periods of time. Hirschfeld (1981: 383– 390) surveyed 15 winepresses east of Emaus and divided them into four chronological groups: 1. A simple rock-cut winepress, with no built-up additions. 2. A complex winepress which included a device for squeezing grapes’ waste. 3. A simple, hewn winepress with mosaic floors and built-up segments. 4. A complex winepress with mosaic floors, cupmarks, and an elaborate pressing installation.

A rock-cut vineyard One method of growing vines in ancient times was planting them in hewn shafts or cut ditches (see Dar 1986a: 12; Getzov 2005: 49–104). Zertal and Mirqam (2000: 461–467) surveyed bare bedrock areas where many shafts (0.5 × 0.5 m; depth c. 1 m) had been hewn and suggested that these were, in fact, vineyards. The cupmarks near the shafts would, therefore, be bases for wooden posts which supported the climbing vines. Ditches next to the shafts served to pour water into the latter (Zertal and Mirqam 2000: 463). Cupmarks often found near the winepresses were used to place poles for some kind of roofing extending above the winepress during the hot summer. Small rock-cut shafts (diam. c. 0.50 m) were found in the study area, especially in its rocky sections. The shafts were hewn deep into the soil, creating a proper bedding for planting vines (Fig. 6.23). It was not evident which method was used in these vineyards, but it is plausible the vines grew

However, Ahlstrom (1978: 19–49), who investigated over a hundred winepresses in the Jenin-Megiddo region, believes that winepresses cannot be dated based on their shape. Frankel (1984) suggests it is possible to date winepresses based on the date of the nearby settlement, as long as this is not a multi-period site. Ayalon (1997b: 150) argued that dating Byzantine winepresses depended on several parameters, such as the existence 144

Amit Shadman: Chapter 6, Roads and Agricultural Systems Distribution of the winepresses according to the dimensions of their collecting vats 2–3 meter 5%

3–4 meter 3%

0–1 meter 0–1 meter 32%

1–2 meter 2–3 meter

1–2 meter 60%

3–4 meter

Distribution of the winepresses according to the dimensions of their treading floors 5–6 meter 5% 4–5 meter 5%

6–7 meter 9%

1–2 meter

1–2 meter 19%

2–3 meter 3–4 meter 3–4 meter 29%

4–5 meter

2–3 meter 33%

5–6 meter 6–7 meter

Fig. 6.21. Distribution of the winepresses according to the dimensions of their collecting vats and treading floors.

Number of winepresses

Capacity of the collecting vats (liters)

Fig. 6.22. Distribution of the winepresses according to the capacity of the collecting vats.

145

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods of a later technique, of a wooden screw installed in the center of the treading floor, used for secondary extraction.

periods, the adjacent installations had presumably operated during the same periods. A similar picture was observed south and east of this area, near the sites of Naḥal Shiloh (West) (Site No. 5) and Site No. 4, respectively.

Excavations of winepresses in the study area yielded mostly undatable potsherds, sometimes dating from several periods. Even when plaster remains were traced on the walls of a winepress (or any other plastered installation), an element which could point to a certain period (Porath 1989: 69–76; Porath 2002: 25–36), it did not mean that the winepress had not been built and used earlier.

In contrast to the simple winepresses, the larger and more sophisticated winepresses were generally found within the boundaries of the Byzantine and later settlements, and their agricultural plots were in ‘the rural-towns strip’. In fact, complex winepresses, Hirschfeld’s groups 3–5, were discerned from the area nearby H. Te’ena (dated to the Byzantine period) eastward. The most outstanding evidence in ‘the rural-towns strip’ is the sharp decline in the number of irregular-shaped winepresses. It seems, therefore, that the simple winepresses belonging to groups 1 and 2 probably operated from Iron Age II through the Hellenistic period, while winepresses with mosaic floors (groups 3–5) operated during the Hellenistic (less likely), Roman and Byzantine periods. The large winepress—the only one within the study area where a screw device was discovered—apparently operated in the Byzantine period, the main occupational period of the site. It is worth noticing that, sometime in the Early Islamic period, this winepress went out of use and was converted into a quarry.

The location of the winepresses as a dating tool and as a settlement marker The distribution pattern of winepresses and their location in the rural area may point to their date and their relation to the nearby settlement. These issues are being addressed, in this study, by examining ‘the farmhouses strip’ and ‘the rural-towns strip’ to its east. The surveys showed that most of the simple winepresses had been hewn on the gently sloped bedrock outside of the settlements, in the open areas near ‘the farmhouses strip’. Excavations of some 30 winepresses showed that most of those located near farms are dated from the Iron Age to the Hellenistic period and can be attributed to Hirschfeld’s groups 1 and 2—simple irregular and simple symmetrical winepresses, respectively. A particularly outstanding cluster of such winepresses was found between Qasr es-Sitt (Site No. 1) and Wadi el-Bureid (Site No. 2, Fig. 6.18). Since these two sites were active mainly during the Persian and Hellenistic

The location of winepresses within the landscape In the Land of Israel, harvesting took place during the hot summer—between June (mainly in the south) and September (in cooler areas, Hadas 2007: 82). This factor affected the location of winepresses in the rural landscape. Unlike oil presses, which operated mainly

Fig. 6.23. A rock-cut vineyard.

146

Amit Shadman: Chapter 6, Roads and Agricultural Systems Table 6.2. List of the winepresses, their locations, and dimensions. No.

No. in Kocavi and Beit Arieh 1994

Dimensions

Volume/ liters

Notes

1.

70

Treading floor: 1.0 × 1.1 m. Collecting vat: 0.55 m deep; 0.75 m. ∅

96

2.

87

Treading floor: 2.4 × 2.7 m. Collecting vat: 0.8 × 1.2 m.

87

3.

89

Treading floor: 3.5 m. ∅ Collecting vat: 0.8 × 1.2 m

Another press nearby.

4.

91

Treading floor: 2.2 × 2.5 m. Collecting vat: 1.5 m. ∅

Plaster remains on the vat.

5.

92

Treading floor: 2.6 × 2.6 m. Collecting vat: 1.4 × 1.5 m; 0.7 m. ∅

1470

Trapezoid floor

6.

93

Treading floor: 2 × 2 m. Settling pit: 1.4 × 1.8 m; 1 m. deep. Collecting vat: 1.0 × 1.6 m; 0.3 m. deep.

480

Another press nearby.

7.

95

Treading floor: 3.7 × 4.0 m. Settling pit: 0.65 × 0.70 m; 0.3 m. deep. Collecting vat: 1.1 × 2.2 m.

8.

101

Treading floor: 2.15 × 3.20 m. 0.15 deep. Collecting vat: 1.7 m. ∅; 0.95 m. deep.

c. 2000

9.

102

Treading floor: 3.6 × 3.9 m. Settling pit: 0.65 × 0.85 m. Collecting vat: 1.06 × 1.50 m.; 0.95 m. deep.

2250

This press was built over an older one.

10.

107

Treading floor: 1.2 × 1.9 m. Collecting vat: 0.4 × 0.6 m.

11.

114

Treading floor: bedrock. Collecting vat: 1 × 1 m.

12.

117

Treading floor: 2.5 × 2.5 m. Collecting vat: 1.0 × 1.4 m.; 0.6 m. deep.

840

Oval vat.

13, 13a

120

Treading floor: 7 × 7 m. with 0.4 × 0.4 m. central pit. Settling pit: 1.0 × 1.3 m. Collecting vat: 2 × 4 m.; 1.5 m. deep.

c. 12000

Large press at Kh. Kesfa. Another press was found east of the site.

14

131

Treading floor: 3.0 × 3.5 m. Collecting vat: 1.4 m. ∅; 1.1 m. deep.

1700

Byzantine sherds.

15.

133

16.

139

Treading floor: 2.0 × 2.3 m. Collecting vat: 1.7 × 2.0 m.

Floor on worked bedrock.

17.

143

Triangular treading floor. Collecting vat: 1.7 × 2.0 m.

Oval collecting vat.

18.

149

Treading floor. 2.5 × 2.5 m. Collecting vat: 1.0× 1.5 m.

c. 2000

The collecting vat is plastered with gray plaster.

19.

158

Treading floor: 3 × 4 m. Collecting vat: 1.8× 2.3 m.; 1.7 m. deep.

c. 2000

Near Kh. Umm al-Hammam.

20.

159

21.

165

Treading floor. 2.4 × 3.0 m. Collecting vat: 0.65 × 1.2 m.; 0.2 m. deep.

22.

168

Treading floor. 1.75 × 2.0 m.

One of the collecting vats may serve as settling pit.

Rock-cut press north of Migdal Afek.

A rock cut press belonged to Wadi al-Bureid (south). 156

A cistern nearby. Damaged.

147

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods Table 6.2 (cont.). List of the winepresses, their locations, and dimensions. No.

No. in Kocavi and Beit Arieh 1994

Dimensions

Volume/ liters

Notes

23.

169

Collecting vat: 1.5 m. ∅.

24.

188

Treading floor: 3.1× 3.2 m. Settling pit: 0.55 × 0.55 m. Collecting vat: 1.1 × 1.2 m.

25.

190

Collecting vat: 0.8 × 1.1 m.; 0.3 m. deep.

264

26.

195

Treading floor: bedrock. Collecting vat: 0.6 × 1.0 m.; 0.5 m. deep.

300

27.

197

Treading floor: 1.0 × 1.7 m. Collecting vat: 0.7 × 0.8 m.; 0.5 m. deep.

280

28.

205

29.

225

Treading floor. 1.5 m. ∅.

A cistern nearby.

30.

253

Treading floor: 4 × 7 m. Collecting vat: 1.2 × 1.2 m.

Bedrock treading floor.

Irregular press. Rock cut installation, may also use for oil.

31

Treading floor: bedrock. Collecting vat: 0.5 m. deep.

32.

Treading floor: 1.85 × 2.6 m. Collecting vat: 1.00 × 1.25 m.

33.

Treading floor: 1.8 m. ∅. Collecting vat: 0.78 × 0.90 m.; 04 m. deep.

280

Irregular press.

34.

Treading floor: 3.5 × 3.5 m.; 0.2 m. deep. Collecting vat: 1.5 × 1.5 m.; 2 m. deep.

4500

Remains of plaster on the floor and vat. Byzantine sherds.

35.

Treading floor: 2.5 × 2.5 m.; 0.4 m. deep. Collecting vat: 1.0 × 1.5 m.; 0.8 m. deep.

1200

36.

Treading floor: 2.8 × 3.1 m. Settling pit: 0.5 × 1.0 m.; 0.2 m. deep. Collecting vat: 1.1 × 1.1 m.; 1 m. deep.

1000

37.

Treading floor: 2 × 2 m. Collecting vat: 0.78 m. ∅; 0.62 m. deep.

296

38.

Collecting vat: 0.72 m. deep.

Irregular press.

39.

Treading floor: 1 × 1 m.

Small press with irregular vat.

40.

Treading floor and Collecting vat: 1.6 × 3.5 m.

Irregular vat.

41.

Treading floor: 2 m. ∅. Collecting vat: 1 m. ∅; 1.5 m. deep.

42.

Treading floor: 1.75 × 1.88 m. Collecting vat: 0.8 m. ∅; 0.6 m. deep.

43.

Treading floor and Collecting vat: 1.56 × 2.6 m.

44.

Treading floor: 2 × 2 m. Collecting vat: 1 × 1 m.; 0.8 m. deep.

c. 800

45.

Treading floor: 1.0 × 1.8 m. Collecting vat: 0.5 × 1.0 m.; 0.5 m. deep.

c. 258

46.

Treading floor: 1.75 × 1.90 m. Collecting vat: 0.5 × 0.5 m.; 0.5 m. deep.

c. 125

47.

c. 1150

Oval floor and vat. Cupmark on the floor.

Irregular press.

Irregular press.

Incomplete press.

148

Amit Shadman: Chapter 6, Roads and Agricultural Systems Table 6.2 (cont.). List of the winepresses, their locations, and dimensions. No.

No. in Kocavi and Beit Arieh 1994

Dimensions

Volume/ liters

Notes

48.

Treading floor: 1.5 × 2.0 m. Oval collecting vat.

49.

Treading floor: 1.5 × 2.0 m.; 0.3 m. deep. Collecting vat: 1 × 1 m.; 0.7 m. deep.

700

A cup mark in the center of the floor.

50.

Treading floor: 3.34 × 3.92 m. Settling pit: 0.95 × 1.06 m.; 0.29 m. deep. Collecting vat: 1.36 × 1.38 m.; 1.3 m. deep.

2400

The largest excavated press, dated to the Byzantine period.

51.

224

Treading floor: 4.5 × 4.5 m. Collecting vat: 2 m. ∅.

West of the central buildings at Kh. Tukkim.

52.

Treading floor: 3.0 × 3.8 m. Settling pit: 0.8 × 0.9 m.; 0.7 m. deep. Collecting vat: 2 m. ∅; 1.9 m. deep.

5500

Bell shaped vat, probably used as cistern between harvesting seasons.

53.

Treading floor: 3.3 × 3.35 m. Collecting vat: 0.95 × 1.30 m.; 1.2 m. deep.

1480

Hydraulic plaster on the vat.

54.

Treading floor: 3.4 × 3.7 m. Settling pit: 0.70 × 0.73 m. Collecting vat: 1.00 × 1.55 m.; 1.3 m. deep.

2020

On the northern edge of H. Te’ena, secondarily used as a quarry.

55.

Between Kh. Kesfa and Kh. Umm al-Buried.

56.

Treading floor: 2.9 × 5.0 m. Collecting vat: 1.3 × 1.4 m.

Irregular press. White tesserae and Byzantine sherds.

57.

Press I Treading floor: 1.8 × 2.3 m. Collecting vat: 0.9 × 1.20 m.; 0.8 m. deep. Press II Irregular press, was replaced by press I.

Two adjacent presses. A rounded wall above the floors. Many Byzantine and early Islamic sherds.

58.

Treading floor: 1.7 × 1.9 m. Collecting vat: 0.6 × 1.0 m.

North of Site No. 4.

59.

Treading floor: 1.5 × 1.6 m. Collecting vat: 0.8 × 1.0 m.

60.

Treading floor: 2.0 × 2.2 m. Collecting vat: 0.85 × 1.00 m.; depression in the vat 0.2 × 0.2 m.; Ditch 0.12 m. wide, 0.16 deep.

61.

A simple press on a terraced slop.

A large press northwest of the church at H. Yeqavim.

within the settlements, winepresses were located close to the vineyards (Frankel 2009: 2). Since the grapes are sensitive to the summer heat, they had to be processed and stored in a cool place as quickly as possible. The fact that the winepresses within the settlements were generally larger and more complex than those in the open areas, brings up the question why, at a later stage, in our region, probably at the beginning of the Byzantine period, the winepresses had been moved into the large settlements or near them.

The answer possibly lies both in the development of the winepresses and the improved means of transportation. Increasing the size of the winepresses meant additional manpower during the harvest season (similar to oil presses). This manpower was naturally present within the settlements. At the same time, the complex network of rural and agricultural roads in the area significantly reduced the transporting time from the vineyard to the large winepress within the settlement. Therefore, this road network may have contributed to 149

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods moving the winepresses to the settlements. However, the small, simple winepresses clearly continued to operate within the vineyards.

be kept in mind that the winepresses in the ‘farmhouses strip’ were used also in later periods, sometimes as small reservoirs. Caution should be exercised when calculating the quantity of wine production in the region. Although it is possible to calculate the volume of the collecting vats at a given time, it is impossible to know how many times the collecting vats were filled during the harvest season.

Summary Grapes (Vitis vinifera) were the main ingredient in ancient diet besides grain and olive oil. These basic crops were an important component of the agricultural landscape through history. Wine, in all its types, is often mentioned in the Talmud (Frankel 1984) and the classical literature (Frankel 2009: 1). Particularly noteworthy are Roman researchers and historians whose studies are still used as a source of information on the ancient wine industry. This is particularly the case of Columella, who described, in his de re rustica, the different areas suitable for vineyards. Cultivating vines was done using several methods (Ayalon, Frankel and Kloner 2012: 18) also described by ancient writers. For example, the 1st century CE historian, Pliny the Elder, reported five methods of growing vines (Naturalis Historiae, 164–166). Nevertheless, there are only a few historical sources that mention vineyards in the Samaria Hills, the earliest dating from the Crusader and Ayyubid periods (Amar 2000: 103–108), such as a document dated to 1165 and mentioning vineyards at Mirabel-Migdal Afeq.

Field Towers The field towers in western Samaria Hills were discussed at length by Dar (1982: 149–195; 1986b: 40–51). Onehundred field towers were recorded in the study area, while some additional ones were not included in the various studies and/or were destroyed over the years. Dozens of these towers were excavated (Fig. 6.24; Table 6.3), and most were built in a similar or identical fashion (Fig. 6.25.1; 6.25.2). The towers were square/ rectangular or rounded (Fig. 6.26), and both types were built of large- to very large-sized stones (1.5 m or more), especially their roofs. Most towers were square (c. 65%; Fig. 6.27) and only 28% were rounded (the rest 7% were of various shapes). The square/rectangular towers measured 3 × 3–5 × 5 m, and the rounded ones minimum 3–7 m in diameter (Fig. 6.27). Unsurprisingly, most rounded towers were better preserved, as their roofing is more easily built and rather stable. In several cases, the rounded towers were built of double or triple walls, a technique identified also in the Northern Samaria Hills (Shadman 2008).

Wine industry was an important component in the economy of rural settlements. Most of the 61 winepresses recorded in the study area (and there are probably more) were outside the settlements, on bare rocky areas suitable for quarrying. These winepresses were simple and included a treading floor and a collecting vat. Larger and more complex winepresses, compared to those near the farms, were discovered within the large settlements (e.g., Kh. Kesfa, H. Te’ena, and Kh. Umm al-Hammam) and they could store a vast number of grapes during the harvest. There is no consensus among scholars on how to date the winepresses, although it would seem that, as stated by Hirschfeld, the more complex and sophisticated the winepress was, the shorter the time it served, meaning that the simple, hewn winepresses were used for a longer time (Hirschfeld 1981: 389).

The square towers were usually built of a single wall, and the fact that they seemingly did not have a second floor allowed for their exterior walls to be thinner (see below). The entrances to the towers were mostly similar: A 1-m wide doorway, with two stone doorjambs, and a large stone lintel. The bedrock floor was leveled and (or) filled with local soil. Roofing was apparently a great challenge for their builders since it is very difficult to roof such a structure. The roofs were laid on corbelled vaults whose stone slabs are placed one on top of the other leaving a protrusion. Using this technique enabled to quickly roof the tower, but it was architecturally unstable and, in most cases (especially in square towers), the roofs collapsed into the tower. Presumably, a wooden post was placed in the center of the tower to stabilize the roof. Often, small stones and soil were laid on the roofs, so much so it looked like a stone clearance heap. Covering the towers probably aimed to isolate them, as much as possible, and thereby create a cooler place, devoid of animals, that could support a second floor (see below). Although the outer contours of the towers were quite large, their interior was too small to host a human being. It would seem, therefore, that these facilities were used, at least in our area, to safely store agricultural products.

To date the winepresses, it is essential to examine the nearby settlements and not the specific winepress itself. However, analysis of the shape of the winepresses and their locations led to the conclusion that the construction dates of the simple irregular winepresses discovered in the ‘farmhouses strip’ are probably the same as their operating period, i.e., from Iron Age II through the Hellenistic period. This type of winepresses was hardly present in the vicinity of Byzantine settlements where larger and improved winepresses were generally found. However, it should

150

Amit Shadman: Chapter 6, Roads and Agricultural Systems

field towers

Fig. 6.24. Distribution of field towers.

151

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods

Fig. 6.25.1. Plans of selected excavated field towers.

152

Amit Shadman: Chapter 6, Roads and Agricultural Systems

Fig. 6.25.2. Plans of selected excavated field towers.

Fig. 6.26. A rounded field tower east of H. Te’ena.

153

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods Distribution of field towers according to their dimensions. 7–8 meter 3%

Up to 2 meter 3%

Up to 2 meter 2–3 meter

5–6 meter 13%

2–3 meter 22%

3–4 meter

4–5 meter 14%

4–5 meter 5–6 meter 3–4 meter 45%

6–7 meter 7–8 meter

Distribution of field towers according to their outlines. unknown 7% round 28%

square round

square 65%

unknown

Fig. 6.27. Distribution of field towers according to their outline and dimensions.

Another, less common type of tower had no inner space. It should be emphasized that these structures were not stone clearance heaps, but rather another type of field towers, usually smaller, square, and without an opening. Their interior was filled with small- and medium-sized stones up to the top level of the outer walls and formed a leveled surface. It seems, therefore, that these were a kind of base (a podium) on which a shed was erected. A few such towers were built within enclosed plots of H. Te’ena and H. Yeqavim. Much larger field towers were discovered in other parts of the Samaria Hills, some of them twostories high (e.g., Dar 1982a: 9, photo 46). The modern Arab villagers still use field towers as shelters and watching towers (Ron 1977a), as seen, for example, in tower L2084 (Fig. 6.25), where a spiral staircase leads from its base to the roof. Sheds made of perishable

material are erected on top of these towers (Ayalon 2012a: 93, photo). Dating the field towers As with other agricultural installations, the field towers yielded only worn and eroded potsherds that cannot be used to date them. In a few towers, such as Tower 88, fragments of a bag-shaped Byzantine jar were found on the floor, while in other towers, other pottery body sherds of this period were also recovered. Thus, the towers seem to have been used mainly during the Byzantine period, when the area was intensively cultivated. The fact that the field towers were located a few hundred meters from the Byzantine settlements may attest to their use in this period since the settlements used their own storage 154

Amit Shadman: Chapter 6, Roads and Agricultural Systems facilities and the towers stored agricultural products in the fields. However, towers were often discovered in close proximity (sometimes only a few meters) to Iron Age II and Hellenistic farmhouses, meaning that they operated also during a later period, as there was no apparent reason to build them adjacent to a farm containing storage facilities within it.

the must (see also Ayalon 2012b: 90–91), which required a cool and dark place. Dar tested the temperature in one of the field towers at Qarawat Beni Hassan and found it to be up to 15ͦ C cooler than its surroundings on a hot day. In addition, the optimal conditions for storing wine were specified already by the 1st-century Roman historian Pliny the Elder.

Dar (1982a) reported that the field towers had been already built in the Hellenistic period and continuously used from this period through the Roman period. In his excavations of several field towers in the Modi‘in area, ‘Ad (2016) reached a similar conclusion.

The towers in the study area were located at a maximum distance of 100 m from the winepresses. One hundred field towers and 61 winepresses were found in the study area, i.e., a ratio of slightly over 1.5 towers per winepress. Figure 6.28 shows that wherever some winepresses were discovered (e.g., north of Kh. Ta‘amur, and east and northeast of Migdal Afeq) relatively few field towers were also discerned. This could result from the fact that these few winepresses were of the simple type dated to Iron Age II–Hellenistic period when field towers were not widespread. Alternately, it is possible contemporary works destroyed the towers. Although the first reasoning is more likely, one should remember that simple winepresses had a long duration. Field towers with nearby winepresses were recorded in an extensive study conducted in the Modi‘in area (Golani 2005: 74, Fig 1). It seems, therefore, that this phenomenon extended over large parts of the central mountains, and even to other areas in the Land of Israel (e.g., Peleg and Feler 2004b; Sion, Haiman and Artzi 2008; ‘Ad 2011).

Excavations near field towers in the Judean Hills revealed Iron Age II potsherds, but it is not specified whether the towers were used in this period or not (Peleg and Feller 2004a). Zertal and Mirqam (2000: 489–490) dated the field towers discovered near the site of Al-Ahwat to Iron Age I. Other field towers excavated in this area were dated to the Roman period (Shadman 2008; H. Torga, personal communication). Tens of field towers excavated in the Negev were exclusively dated to the Byzantine period, based on pottery and a few 5th–6th centuries CE coins (Nevo 1991; Fabian and Goldfus 2004: 3; Haiman 2007: 137). The use of field towers in the southern Land of Israel ceased in the Early Islamic period (Nevo 1991, Haiman 1995: 5–29), possibly because of the Islamic prohibition of alcohol consumption.

Summary Field towers were widespread in vast areas of the Land-of-Israel central mountains, especially in the Western Samaria Hills and even in other regions such as the Carmel. Nevertheless, there are large areas, such as the Judean Hills, where only a few towers were found. For example, recent studies conducted in the Bet Shemesh area have shown that field towers are not widespread in that area (in many other areas even a single tower is rare). It is possible that this absence results from the soft rock formation that characterized many areas of the Judean Hills, not suitable for roofing the towers, perhaps replacing these with subterranean basements. Another reason for the lack of towers could be a lack of interest in them from the part of the later inhabitants of the area (perhaps of the Ottoman period), who converted them to other uses such as limekilns or a source of stones for building terraces.

Field Towers as Mausolea Field towers were often found in conjunction with burial systems, and those particularly impressive were even renamed ‘mausoleum’. For example, in a survey conducted by Gophna and Porath (1972: 201, 223, 230) in the Northern Samaria Hills, a field tower with two sarcophagi was recorded, together with bones and potsherds dating to the Byzantine period. A square field tower south of the village of Qarawat Bani Hassan (map ref. 209800/669800) was also defined as a mausoleum. However, it seems that, despite the few cases where field towers were used for burial purposes, they mainly functioned as storage facilities for agricultural products. None of the excavated towers yielded burial remains but were rather built within an agricultural context and as an integral part of the whole rural system. Nevertheless, the burial grounds found in the study area, such as the mausoleum at H. Yeqavim (Site No. 9) and the burial systems adjacent to Kh. Kesfa (Site No. 13), presented a different pattern.

As already stated, the field towers at the Samaria Foothills are smaller than those located on the east, in higher locations. In the many studies conducted in these areas, it was ascertained that many towers had been massively constructed (especially the square ones), often having a second floor. It is possible that the outstanding size of these towers was due to the security situation in the area or to the large quantities of

Winepresses and Field Towers Dar (1982a: 174; 2000: 165) has pointed to a link between the winepresses and the field towers and argued that the latter provided a suitable place for fermentation of 155

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods

Fig. 6.28. Distribution of the field towers and winepresses.

156

Amit Shadman: Chapter 6, Roads and Agricultural Systems Table 6.3. List of the field towers, locations, and dimensions. No.

No. in Kocavi and Dimensions Beit Arieh 1994

Preserved Height

Shape

Notes

1.

76

2.5 × 3.0 m.

2–3 courses.

Square.

2.

76

4 m. diameter

2–3 courses.

Rounded.

3.

77

2.5 m. diameter

4 courses.

Rounded.

4.

79

4 × 5 m.

5.

82

2.5 × 2.5 m.

6.

83

2.8 m. diameter

Rounded.

7.

89

3 × 3 m.

Square.

8.

59

3.5 × 3.7 m.

9.

59

5.5 × 3.5 m.

Square.

10.

59

2.0 × 2.5 m.

Square.

11.

59

3.2 × 3.5 m.

Square.

12.

74

4.3 × 5.8 m.

2–3 courses.

Square.

13.

74

4.6 m. diameter.

2–3 courses.

Rounded.

14.

75

3 × 3 m.

15.

75

5 m. diameter.

2–3 courses.

Rounded.

16.

93

4 × 4 m.

1 course.

Square.

17.

97

4.0 × 4.5 m.

1 course.

Square.

18.

99

3 × 4 m.

2 courses.

Square.

19.

100

4 × 4 m.

Square.

20.

101

3 × 4 m.

Square.

21.

104

4 m. diameter

2 courses.

Rounded.

22.

104

5 × 6 m.

1–2 courses.

Square.

23

106

2.5 × 3.2 m.

3 courses.

Square.

Pavement around the tower.

24.

108

4 × 4 m.

Square.

Poor remains.

25.

109

4 m. diameter

1 course.

Rounded.

26.

110

3 m. diameter

2 courses.

Rounded.

27.

113

3 m. diameter

2 courses.

Rounded.

28.

114

3 m. diameter

2 courses.

Rounded.

29.

115

2 × 3 m.

1 course.

Square.

30.

115

3 × 3 m.

5 courses.

Square.

31.

115

2.5 × 2.7 m.

0.7 m.

Square.

32.

116

3 × 3 m.

33.

119

3 × 3 m.; opening 0.6 m. wide.

2 courses.

34.

123

3 × 3 m.

4 courses.

Square.

35.

124

3 × 4.7 m.

4 courses.

Square.

36.

124

2 courses.

Square.

37.

126

3.7 × 3.7 m.

38.

144

6 m. diameter

39.

146

4 × 4 m.

Square. 3 courses.

2–3 courses.

Square.

Square.

Square.

4 neighboring towers.

A cluster of towers south of Kh. Dayyar.

Built of 1 × 1 m. stones.

Built of 1 × 1 m. stones. Walls 0.7 m. wide.

Square. 1 m.

Rounded. Square.

157

Built of roughly worked stones.

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods Table 6.3 (cont.). List of the field towers, locations, and dimensions. No.

No. in Kocavi and Dimensions Beit Arieh 1994

Preserved Height

Shape

4 courses.

Rounded.

Notes

40.

6 m. diameter; 2.8 m. inner diameter.

41.

6 m. diameter; 2.1 m. inner diameter.

Rounded.

Built of 0.3 × 0.5 m. fieldstones.

42.

5 m. diameter

Rounded.

Walls 2 m. wide.

Square.

43.

149

3 × 3.5 m.

44.

150

3.5 × 3.5 m.

3–4 courses.

Square.

45.

150

3.5 × 5.4 m.

3 courses.

Square.

46.

152

3.5 m. diameter

Rounded.

47.

178

3 × 3 m.

Square.

48.

181

3.5 × 4.5 m.

49.

184

2.7 m. diameter

50.

186

3.0 × 3.5 m.

51.

191

3 × 5 m.

52.

191

2.4 × 3.0 m.; opening 0.8 m. wide.

53.

191

54.

2 courses.

Square. Rounded.

2 courses.

Square. Square.

A cluster of 7 towers. Byzantine sherds.

1.5 m.

Square.

Opening .8 m. wide.

2.5 × 4.0 m.; opening 0.8 m. wide.

1.5 m.

Square.

191

5 m. diameter.

1.5 m.

Rounded.

55.

191

3.0 × 4.5 m.; opening 0.8 m. wide.

Square.

56.

191

4 × 4 m.

Square.

57.

191

3 × 3.5 m.

Square.

58.

192

2.5 m. diameter.

Rounded.

59.

192

2.7 × 3.0 m.

2–3 courses.

Square.

60.

192

2.5 × 3.0 m.

2–3 courses.

Square.

61.

199

3.15 × 4.3 m.; opening 0.5 m. wide.

62.

202

3.9 m. diameter.

63.

202

3.4 × 5.4 m.

Square.

64.

203

4.6 × 5.4 m.

Rounded.

65.

204

4.0 × 4.4 m.

Square.

66.

252

3.1 × 3.4 m.

2 courses.

Square.

Byzantine sherds.

67.

4 × 4 m.

1 course.

Square.

Hellenistic and Roman sherds.

68.

4 × 4 m.

1 course.

Square.

69.

2.5 × 2.5 m.

1 course.

Square.

Built of 1 m. stones.

70.

3.8 × 3.8 m.; opening 0.6 m. wide, 0.8 m. high.

4–5 courses.

Square.

Built of 1.4 m. stones.

71.

3.5 × 5.0 m.; opening 0.6 m. wide, 1.2 m. high.

1 course.

Square.

Built of 1.5 m. stones.

72.

2.5 × 2.5 m.

73.

3.5 × 2.5 m.; opening 0.6 m. wide.

2 courses.

Square.

74.

2 × 2 m.

2 courses.

Square.

Square. 4 courses

Square.

Square.

158

A wall is attached to the tower.

Amit Shadman: Chapter 6, Roads and Agricultural Systems Table 6.3 (cont.). List of the field towers, locations, and dimensions. No.

No. in Kocavi and Dimensions Beit Arieh 1994

Preserved Height

Shape

Notes

75.

2.5 × 3.0 m.; opening 0.5 m. wide.

1 course.

Square.

76.

3.5 × 3.5 m.; opening 0.5 m. wide.

3 courses.

Square.

77.

4 × 5 m.

2 courses.

Square.

78.

6 m. diameter; opening 0.5 m. wide.

1 m.

Rounded.

79.

7.1 m. diameter.

1 course.

Rounded.

80

3.5 m. diameter; opening 0.6 m. wide.

2 courses

Rounded.

81.

4.5 m. diameter; opening 0.4 m. wide.

2–3 courses.

Rounded.

82.

4 m. diameter; opening 0.5 m. wide.

2 courses.

Rounded.

83.

8 m. diameter.

4 courses.

Rounded.

84.

5 m. diameter.

3 courses.

Rounded.

85.

3.1 × 3.5 m.; opening 0.5 m. wide.

2–3 courses.

Square.

Byzantine sherds.

86.

4 × 6 m.

4 courses.

Square.

Built of 0.6 × 0.7 m. fieldstones.

87.

2 × 2 m.

Square.

A winepress (No. 56) nearby.

88.

7.5 m. diameter; opening 0.5 m. wide.

4 courses.

Rounded.

Fragments of a Byzantine jar on the floor.

89.

4.0 × 4.5 m.; opening 0.6 m. wide.

3 courses.

Square.

90

5 m. diameter; opening 0.7 m. wide.

5 courses.

Rounded.

A rounded wall on its east.

91.

Unknown.

92.

Unknown.

93.

Unknown.

94.

Unknown.

95.

Unknown.

96.

3 courses.

Square.

97.

3 courses.

Square.

98.

2.0 × 2.1 m.

3 courses.

Square.

99.

2.4 × 3.2 m.

3 courses.

Square.

100.

Rounded.

agricultural products that required immediate storage. The average ratio of field towers/area in the study area is one tower per approximately 330 dunams, but their distribution is uneven, being distributed above the alluvial areas (Fig. 6.24; note the topography), on moderate ridges, or in wadis outlets.

number of winepresses in the area. The winepresses within these areas were mostly located near buildings or inside them, making the field towers unnecessary since the agricultural products were transferred directly to storage facilities within the villages. Dating the towers, like other agricultural installations, is somewhat problematic. Most excavations conducted in towers yielded no diagnostic finds. Based on a few identified potsherds, it would seem that the field towers

A few field towers were discovered in the vicinity of ‘rural towns’, and this may relate to the small 159

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods The cisterns in ‘the farmhouses strip’

operated mainly during the Roman and Byzantine periods. However, it is possible that these facilities had been used already in earlier periods.

In most sites, except for Qasr es-Sitt, a single cistern was discerned near each farm. Since most farms in the region were founded during Iron Age II, the cisterns were hewn in this period, as it is unlikely that a farm could survive even one day without water. The farms continued into the Persian and Hellenistic periods when the cisterns were renovated and covered with plaster, enabling their dating (Porath 1989).

Water Cisterns As previously noted, the local population in antiquity relied almost entirely on rainwater collected in cisterns and reservoirs. Although the study area is totally lacking in natural water sources, productive communities lived here for thousands of years. The area’s proximity to the sources of the Yarqon ensured the inhabitants a stable water source within a short walking distance during a drought. Over 100 water cisterns were recorded in the area, and there are clearly many more (Fig. 6.29). Water reservoirs were also discovered, particularly in Byzantine sites, which significantly increased the storage volume. Naturally, not all the cisterns were measured and their total volume can only be estimated. The volume of the thoroughly examined reservoirs, however, could in most cases be calculated (Table 6.4). The following discussion follows regional subdivisions and chronology. Most water cisterns in our region are ‘bellshaped’, as discerned also in a comprehensive study conducted at the nearby Kh. Zikhrin (Tsuk 1994: 133). The cisterns had a circular or square opening (diam. c. 1.0–1.5 m). Many of them still have their capstones, in situ. Some cisterns are particularly large, such as the one located west of Site No. 20. The shafts’ depths varied and could be c. 2 m deep or more, but also much shorter. The cisterns were approximately 3–6 m deep and 4–8 m in diameter, yet the accumulated soil at the bottom should be also taken into consideration.

The cisterns in ‘the rural towns strip’ The settlements in this strip were founded in the Byzantine period and continued into the Early Islamic period. It became evident that no earlier settlement existed in these sites, except for Kh. Ta‘amur and H. Yeqavim where a few Roman sherds were found. The many water cisterns found in this strip were hewn when the settlements were first established, in the Byzantine period. The cisterns were either incorporated into the buildings or located near them. Remains of red plaster characteristic of the Byzantine period were discerned on the walls of the cisterns, and sometimes also ‘speckled’ plaster, attributed to the Early Islamic period (Porath 1989: 74–75). It can be concluded, therefore, that the water cisterns in the large Byzantine settlements (Kh. Kesfa, Kh. Umm alHammam, H. Yeqavim and Kh. Ta‘amur) were hewn during this period, and continued to be used, after being replastered, in the Early Islamic period, when no additional cisterns were seemingly hewn because of the probable decline in population (see Chapter 9). Similarly, several cisterns were recorded at H. Te’ena, averaging c. 3 m in depth and coated with light reddish plaster. The excavations showed that this site was also founded in the Byzantine period and continued into the Early Islamic period. Accordingly, it seems that the cisterns at this site were hewn in the Byzantine period and used, with no plaster repairs, in the Early Islamic period, when the site was abandoned.

Dating the Cisterns Like other agricultural facilities in the area, water cisterns were also used for a long period, turning their dating into a difficult task. Thus, dating them was done, like with other facilities, by relating them to ‘the farmhouses strip’ and ‘the rural towns strip’. Dating a cistern is easy when discussing a single-period site, as it was likely hewn when the site was occupied. However, when a particular site was continuously occupied during several periods, the task becomes more difficult. For example, cistern No. 106 was located a few hundred meters southeast of H. Te’ena, near an Iron Age II– Persian period farmhouse. At some point, probably during the Byzantine period, the cistern was covered by an agricultural terrace and went out of use. It should, therefore, be concluded that the cistern was hewn during Iron Age II when the farm operated. Attention should also be given to the plaster, its thickness, colors and texture, which may shed light on the period of use of the cistern. Nevertheless, as before, this does not tell on when the cistern was quarried.

Cisterns and rural roads The construction of agricultural and rural roads was a communal enterprise and perhaps a project initiated by the formal authorities. The roads were maintained and renovated and their users were provided road services, including water supply (Safrai 1990a: 167). Since our study showed that cisterns were not always located close to settlements, it would seem that they were hewn in relation to the road network and concomitantly with it. Following is a short description of the roads along which water cisterns were identified (Fig. 6.30):

160

Amit Shadman: Chapter 6, Roads and Agricultural Systems

Fig. 6.29. Distribution of cisterns in the study area.

161

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods Road No. 1. Several cisterns were recorded along this main road. A cluster of cisterns (Nos. 100–105) was discerned south of Kh. Ta‘amur, whose inhabitants probably also used them. A cistern (No. 64) was found northwest of H. Yeqavim, and two others (Nos. 35, 94) were identified along the road in the area of Wadi Susi. Cisterns No. 89 and 93 were identified by the side of the road.

Summary The absence of stable water sources in the area forced the ancient inhabitants to ensure water supply by quarrying water cisterns. The initial phase of occupation in our region was in Iron Age II when quarrying cisterns was a common phenomenon. The farms in ‘the farmhouses strip’ continued being active for the next hundreds of years and continued using the same method for their water supply. However, unlike the water consumption pattern in this strip, in ‘the rural-towns strip’ many water cisterns were discovered both within buildings and in open areas. It is estimated that in every ‘rural town’ with a built-up area of over ten dunams, there were about 20 water cisterns or more. It was also obvious that the water cisterns had been hewn alongside the rural roads, possibly as an endeavor initiated by the local government.

Road No. 2. Four water cisterns (Nos. 31, 90–92) were identified along the western branch of the road leading to Migdal Afeq. Road No. 4. Three water cisterns were discerned along this road running from Migdal Afeq to H. Yeqavim, fairly distanced from each other. The eastern cistern (No. 60) was identified at the foot of the site of Naḥal Shiloh (Southwest); the second cistern (No. 58) was a kilometer to the west, and the third one (No. 53) was discovered further west toward Migdal Afeq.

‘Stone Clearance Heaps’

Roade No. 7. This road passed through a tributary of Wadi Al-Bureid. Three water cisterns were identified along the road (Nos. 37, 38, 66), between Kh. Umm alHammam and Wadi al-Bureid (South).

Excavating dozens of heaps (Fig. 6.31) during the present study has raised many questions regarding their role in the rural landscape. The stone clearance

Fig. 6.30. Water cisterns and rural roads.

162

Amit Shadman: Chapter 6, Roads and Agricultural Systems Table 6.4. Water cisterns, locations and dimensions. No.

No. in Kocavi and Beit Arieh 1994

Dimensions

Site/Road

Notes

1.

75

1.5 × 1.5 m. opening.

Near Kh. Dayyar.

Square opening.

2.

87

3.

88

1.1 m. opening ∅; 1.5 m. deep shaft.

North of Qurnat alHaramiya.

4.

89

5 m. deep.

Qurnat al-Haramiya.

A total of 18 cisterns.

5.

89

1.1 m. opening.

Qurnat al-Haramiya.

Square opening.

6.

89

Qurnat al-Haramiya.

7.

89

1.1 m. opening ∅; 1.5 m.; 1.2 m. ∅ Qurnat al-Haramiya. capstone; 0.5 m. ∅ central hole.

8.

91

Qurnat al-Haramiya.

Fragments of capstone nearby.

Qurnat al-Haramiya.

Two openings.

9. 10.

8 m. ∅ at the bottom; 4 m. deep; 1.8 m. ∅ capstone; 0.6 m. its thickness; 1.5 m.; 0.5 m. ∅ central hole.

Qurnat al-Haramiya.

11.

93.

1.2 m. ∅ opening.

Qurnat al-Haramiya.

12.

102

0.6 m. ∅ opening.

Qasr as-Sitt.

13.

102

3 m. visible depth; 1.65 m. ∅ capstone; 0.6 m. ∅ central hole.

Qasr as-Sitt.

14.

108

Kh. Umm al-Bureid.

Inside the site near an oil press.

15.

120

Kh. Kesfa.

4 cisterns within the builtup complex. Additional cisterns known (see Chapter 3).

16.

120

Kh. Kesfa.

17.

120

Kh. Kesfa.

18.

120

Kh. Kesfa.

19.

128

Migdal Afek.

20.

131

21.

132

Migdal Afek.

22.

132

Migdal Afek.

23.

134

24.

136

25.

136

26.

136

27.

136

1.3 m. ∅ capstone; 0.5 m. ∅ central hole.

4.3 × 5.3 m. opening; 4 m. deep.

At least 11 cisterns were discerned.

Migdal Afek.

North of Migdal Afek. Migdal Afek.

2.5 m.

Broken capstone.

Migdal Afek. Migdal Afek.

7 m. ∅; 4 m. deep.

Square opening.

Migdal Afek.

28.

Migdal Afek.

Bell-shaped cistern with rounded capstone.

29.

136

Migdal Afek.

A basin nearby.

30.

137

Migdal Afek.

Collapsed ceiling.

31.

138

32.

140

Still in use. 3 m. deep.

Rafat–Migdal Afek road.

163

Steps leading into the cisterns.

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods Table 6.4 (con.). Water cisterns, locations and dimensions. No.

No. in Kocavi and Beit Arieh 1994

33.

141

34.

142

35.

144

Dimensions

Site/Road

Notes Half capstone.

Farmhouse 740.

Opening plastered in the Ottoman period. A capstone in situ. Bottle-shaped.

36.

H. Te’ena

At the site foothill; three steps.

Wadi al-Bureid (east).

Bell-shaped cistern with capstone overlaid with trough.

37. 38.

155

1.5 m. ∅ opening.

39.

155

Wadi al-Bureid (east).

40.

158

Kh. Umm alHammam.

41.

158

Kh. Umm alHammam.

42.

158

Kh. Umm alHammam.

43.

158

Kh. Umm alHammam.

44.

158

Kh. Umm alHammam.

45.

158

Kh. Umm alHammam.

46.

158

Kh. Umm alHammam.

47.

158

Kh. Umm alHammam.

48.

158

Kh. Umm alHammam.

49.

158

Kh. Umm alHammam.

50.

158

Kh. Umm alHammam.

51.

159

Wadi al-Bureid.

52.

165

53.

168

Migdal Afek–H. Yeqavim road.

54.

170

Migdal Afek–H. Yeqavim road.

55.

171

0.7 × 1.0 m. opening.

56.

172

1 m. ∅ opening.

57.

176

Khallat as-Sihrij

58.

178

Naḥal Shiloh (2); Migdal Afek–H. Yeqavim road.

59.

182

60.

190

11 cisterns inside buildings.

Southeast of an oil press.

Bell-shaped.

0.4 m. opening; 6 m. deep.

Bell-shaped.

Capstone in situ. Migdal Afek–H. Yeqavim road.

164

Capstone in situ, and a trough nearby.

Amit Shadman: Chapter 6, Roads and Agricultural Systems Table 6.4 (con.). Water cisterns, locations and dimensions. No.

No. in Kocavi and Beit Arieh 1994

61.

195

62.

196

63.

Dimensions

Site/Road

Broken capstone. 1.2 m. ∅ opening. 1 m. ∅ opening.

Umm al-Hammam– Deir Balut road.

Bell-shaped.

Naḥal Shiloh–Kh. Dayyar road.

Secondarily used as a tomb.

14 cisterns in buildings.

64.

200

65.

205

66.

205

67.

205

H. Yeqavim.

68.

205

H. Yeqavim.

69.

205

H. Yeqavim.

70.

205

H. Yeqavim.

71.

205

H. Yeqavim.

72.

205

H. Yeqavim.

73.

205

H. Yeqavim.

74.

205

H. Yeqavim.

75.

205

H. Yeqavim.

76.

205

H. Yeqavim.

77.

205

H. Yeqavim.

78.

205

H. Yeqavim.

79.

1.3 m. ∅ opening; 1.75 m. deep shaft; 3.5 m. deep.

Naḥal Shiloh (3).

Capstone nearby. 5 cisterns.

80.

224

H. Tukkim.

81.

224

H. Tukkim.

82.

224

H. Tukkim.

83.

224

H. Tukkim.

84.

224

H. Tukkim.

85.

72

Kh. Taha.

86.

1.2 m. ∅ opening.

Rafat–Migdal Afek road.

87.

Qasr as-Sitt; Rafat– Migdal Afek road.

88. 89.

Notes

Hewn cavities, possibly cisterns. 1.5 m. ∅ opening.

Rafat–Migdal Afek road.

90.

A lime kiln nearby.

91.

1 m. ∅ opening.

92.

1.0 × 1.5 m. opening.

93.

0.4 m. ∅ opening.

A winepress nearby. Naḥal Shiloh–Kh. Dayyar.

94.

In area of agricultural terraces.

95.

A lime kiln and a few winepresses nearby.

165

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods Table 6.4 (con.). Water cisterns, locations and dimensions. No.

No. in Kocavi and Beit Arieh 1994

Dimensions

Site/Road

Notes

96.

0.5 m. ∅ opening; 0.2 × 0.2 m. hewn basin.

North of Migdal Afek.

97.

0.5 × 0.5 m. opening.

Migdal Afek.

Modern iron cover.

98.

1.5 m. ∅ opening.

Kh. Ta‘amur.

7 cisterns.

99.

253

Kh. Ta‘amur.

100.

253

Kh. Ta‘amur.

101.

253

Kh. Ta‘amur.

102.

253

Kh. Ta‘amur.

103.

253

Kh. Ta‘amur.

104.

253

Kh. Ta‘amur.

105.

253

Kh. Ta‘amur.

106.

A cistern in a terraced wadi; a channel diverted runoff into it. A trough nearby.

heaps had different plans and dimensions and comprise several elements:

3. The piled, cleared stones were medium- to very-small-sized. It is unlikely that such stones may affect routine cultivation of the fields. – 4. The soil was intentionally accumulated on the heaps. It could not have naturally eroded as many of the heaps were found in higher locations than their surroundings. – 5. Chives usually grow on most of the heaps (Fig. 6.32), but it is unlikely the heaps were piled for this purpose. However, one should be aware of the fact that the heaps provided a convenient and airy bedding for vegetation (see below). The question is, therefore, what was the purpose of the stone clearance heaps? Columella wrote, in his de re rustica, about clearing fields and accumulating the stones on their margins, but it seems that he referred to large stones that disrupted cultivation.

1. An outer wall, built up to four courses high. This wall was in most cases rounded or elliptical, but two walls forming a corner were also discerned. 2. An inner wall, usually circular, commonly located in the center of the heap. These walls were built to a height of two–three courses, probably for stabilizing the heap or to create an interior space of unclear purpose (perhaps for planting?). 3. Small-sized (fist size) to tiny (gravel) stones mixed with dirt. The excavations showed that besides the ‘cleared stones,’ large quantities of dirt were brought to the site, probably together with the stones. The heaps were found on the bare bedrock or were piled against an agricultural terrace and division walls, while almost none were discovered in alluvial areas. Excavating the heaps yielded worn potsherds dating from a variety of periods. As of today, no reliable interpretation seems to have been offered in archaeological research regarding the purpose of these heaps.

Modern scholars brought up hypotheses, two of which are cited here:

several

[1] The stone clearance heaps in the Negev, often arranged in groups in a particular area (Tsafrir 1984: 351), help to better drain runoff water (Evenari, Shanan and Tadmor 1980). However, the stone clearance heaps in our area were scattered, with no clear pattern. A similar interpretation was suggested by Hammer (2014: 276) regarding stone heaps in southeastern Turkey.

This confusion can be explained, at least in our region, by five possible factors: – 1. The fields are not completely devoid of stones, and it appears as if they had not been cleaned in antiquity. – 2. The outer walls of the heaps were sometimes built up to a height of two–four courses, and it is unlikely they were aimed at preventing deterioration of the heaps.

[2] The second hypothesis, based on the Arabic term ‘Tuleilat al-Enab’ (mounds of grapes) associated with the stone clearance heaps in the Negev, suggests that these heaps were used for growing vines (Negev 1983: 214). This idea should be seriously considered as vines can survive almost in any type of soil, but a bedding of small limestone and/or gravel better supports its 166

Amit Shadman: Chapter 6, Roads and Agricultural Systems

Fig. 6.31. Selected plans of excavated stone clearance heaps.

167

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods

Fig. 6.32. A stone clearance heap with chives (A. Shadman).

growth (Sneh and Hochberg 1976: 177). The presence of Chives, mainly, and sometimes exclusively on the stone heaps, may indicate that they served as a convenient bedding for agricultural crops. Besides growing vines and olives, evidenced in the archaeological findings, other crops, such as fruit trees and vegetables are known to have been grown in antiquity (Kraemer 1958: 95–101; Tsafrir 1984: 354). To conclude, the stone piles were used as a bedding for some agricultural crop, probably vines. A similar technique was observed in contemporary Cappadocia, on the Anatolia plateau, where vineyards grow on piles of stones (Fig. 6.33). It is possible that this is a long-standing tradition that has been preserved in that region which can perhaps illustrate an agricultural method that was prevalent in our region.

type of agriculture, but on the other hand, they were suitable for piling stones. 2. The second interpretation may relate to the numerous limekilns built into the ruins, probably because of the easily available stones. Accordingly, piling stones near the kilns assisted in their operation. Dating the stone heaps by the potsherds poses many difficulties. The piled heaps also included potsherds dating from various periods, both from the time of the farms and from the clearance itself. Thus, three periods are suggested for the formation of the clearance heaps: – The post-Hellenistic period, when most of the farms ceased functioning. – The Roman and Byzantine periods, when the farms were no longer used for residential purposes. – The Ottoman period, when stone heaps were piled in conjunction with the limekilns of that period that cut into the ruins.

Stone clearance heaps on farmhouses Most excavated farmhouses were overlaid by large stone clearance heaps. Piling the stones was done only over the rooms and not over the adjacent courtyards. Thus, it can be concluded that the courtyards of the earlier buildings were also used in the Roman and Byzantine periods, probably for agricultural purposes. If so, why and when did the local inhabitants pile these heaps over the farmhouses?

Cultivation Potential of Grapes, Olives, and Grains in the Area This discussion deals mainly with the Byzantine period, although certain patterns certainly prevailed in earlier periods, focusing on the location of vineyards, olive plantations, and grain fields (mainly wheat and barley). It is clear that other crops were grown throughout history (Kraemer 1958: 95–101; Tsafrir 1984: 354), but these are not evidenced in our region.

Two possible interpretations can be brought forth: 1. It is possible that these piles, like all clearance heaps in the area, were aimed for agricultural crops. On the one hand, these ruined buildings did not allow for any 168

Amit Shadman: Chapter 6, Roads and Agricultural Systems

Fig. 6.33. Cappadocia. Vines growing on stone heaps.

1982a: 82–131; 1986c: 142–169) seems to be one of the most comprehensive. One of his conclusions refers to the relationship between field towers, the wine industry, and the vineyards, as noted also for our region. He examined the lands of Kh. al-Burak and Kh. Najjar and concluded that most of the area of the first site was used for dry-farming (470 dunams), while 360 dunams were allocated to vineyards and only 230 dunams to olive groves. This division largely depended on regional topography, geology and terrain characteristics (see below). In recent years, additional studies have been conducted aiming to examine the location of agricultural crops within large areas (Ramat Beit Shemesh—Kolska-Horwitz 2011: 180–190; Bonfil and Hadas 2011: 202–204).

The 1st-century CE Roman agronomist Columella (de re rustica) talked about an ecosystem and the mutual interaction between environment and life. He specified three factors that must be taken into account when considering living areas: forest, cultivation areas, and pasture. It seems that Columella was well familiar with the rural countryside. Naturally, it has been sometimes impossible to determine the land’s potential in antiquity, either due to the lack of archaeological findings, or the damage caused to the historical landscape by contemporary development. Even little survived of the natural vegetation of the area, although not directly related to the present discussion. Although it can be argued that the operators of the Ottoman limekilns (some 100 kilns over c. 30 sq km) are the ones who damaged the vegetation, it is more likely that the Byzantine inhabitants, who founded most of the agricultural systems in the area, are to blame (Lev-Yadun 1997: 93, 100). For cultivating the area on a large scale, they (and their predecessors, albeit on a smaller scale) cleared and prepared vast areas, thus dramatically changing the landscape.

Aviam (1994: 129), who studied the Christian settlements in Western Galilee, argued that the valleys and wadis in the area were used for growing grain, whereas the rocky areas were used for vineyards and olive plantations. This suggestion fits the situation in our area. The present situation of the Arab village of Deir Balut can shed some light on the potential of crops in the past. Topographically, the village is situated higher than its surroundings, with rocky hills

Of the several studies published on this issue, Dar’s work in the Western Samaria Hills (Dar 169

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods encircling it on all sides but the eastern one. These areas are occupied today by many olive groves, bordering on the south Naḥal Shiloh, and to the north the olive groves of Rafat. East of Deir Balut lies a wide valley, characterized by high-quality alluvial soil used for growing grain. All the communities in the study area dominated large areas in their vicinity, more or less suitable for the various crops. Vines are very durable plants that can survive in fairly harsh conditions, but increasing its crops requires a deeper soil with low levels of limestone. Small limestone and gravel are actually good for vineyards (Sneh and Hochberg 1976: 177) and fits the nature of our region. Similarly, it was found that winepresses in the southern part of the Samaria Foothills were scattered on the slopes of the rocky hills, and many of these areas were apparently used for vineyards (Greenhut 1998: 151–153). Thus, this characteristic was not a local phenomenon.

alluvial lands, whereas on the other sides were rocky areas that had been utilized for agriculture by building farming terraces. These characteristics dictated, in fact, the location of the crops: wheat was grown in the alluvial areas to the west of the site since it cannot grow in other types of land. The large threshing floor found on the northern slopes of the site attested to growing grain in this area. Vines and olive trees grew in the rocky areas around the site. The fairly large winepress excavated in the southwestern part of the site and three others to the east, north, and northwest of the settlement, together with the many field towers, point to the existence of vineyards in these areas. The olive groves were mainly in the southern parts of the site where agricultural terraces and rocky areas are located. The absence of winepresses and field towers in these southern areas is worth noticing. In another site, Kh. Ta‘amur, located above Naḥal Shiloh, most lands are suitable for vineyards and olive groves rather than wheat. The archaeological finds at the site confirmed this observation. Located above a small wadi, Kh. Kesfa dominated large areas in its vicinity. South of the site was a gentle plateau suitable for most agricultural crops. West of the site, the harsh rocky area quite inhibited cultivation. Consequently, it appears that wheat was grown in the plain areas, while vineyards and olive groves were grown in the rocky areas. Each settlement thus seemingly evaluated the potential of the lands and their disposal and their suitability for the various crops. Wherever a certain crop was not grown, the inhabitants could buy it elsewhere.

Olive trees can grow in almost any terrain and type of soil, except for not properly drained areas (Lavi 1976: 202; Ben-David 1998: 46). This is evident also today, in traditional Arab agriculture. Based on the numerous discovered Byzantine oil presses, it is clear that the oil industry was a large agricultural branch. Since olive trees and vines can grow almost everywhere, it is possible that both vineyards and olive groves grew in the same area, presumably in separate plots. Therefore, it is difficult to trace a specific crop in these areas. This is not the case for grain crops, which require specific conditions, including suitable passage areas for an animal and a plow. Wheat and barley were probably the most important crops in the area, as reflected in the many threshing floors found in the study area.

In conclusion, after calculating land use in the study area, it would seem that, in a 20 sq km area, c. 4290 dunams were used for olive groves, c. 5830 dunams for vineyards, and c. 2820 dunams were used for growing grain (Figure 6.34). As raising cattle and grazing were also confirmed in the area, it seems to have been practiced on steep slopes and harsh areas, where cultivation was not possible.

Growing grain requires fertile and welldrained soil (Pinthus and Amir 1994: 321), with an advisable depth of over 30 cm, although it can also be cultivated in shallower lands. Accordingly, the Samaria Foothills were suitable for grain crops. However, since most of this area is rocky and with limited alluvial soil, the farmers, in antiquity, were forced to use these lands for grain rather than groves. Even today, large areas of the Samaria Foothills are used for growing grain, e.g., east and west of H. Te’ena and the vicinity of Kh. Umm al-Hammam. When comparing our area with the Sharon Plain to the west, it becomes clear that more extensive cultivation of cereals took place in the latter, despite the drainage problems. Traditional Arab agriculture in the Sharon Plain was also characterized by growing grain in the heavy, alluvial soil (Roll and Ayalon 1989: 207–208).

It is our belief that the location of agricultural crops in the rural landscape should be part of the following disciplines: Archaeology. Locating archaeological findings, including winepresses, field towers, and threshing floors, may demonstrate the extent of local agriculture. Rocky areas, where no winepresses and field towers were found, had been probably used for olive groves. Oil presses should be excluded, as they were usually located within the villages, and thus cannot attest to the location of the groves. However, their number provides some information on the extent of the olive groves in a given area (Ben- David 1998: 48–52).

H. Te’ena, which dominated relatively large areas in its vicinity, can illustrate this situation. West of the site was a large area characterized by fertile deep

170

Amit Shadman: Chapter 6, Roads and Agricultural Systems

Fig. 6.34. The location of agricultural crops in the study area.

Geography–Topography. The area’s characteristics dictated, to a large extent, which crops were located in each region.

Agricultural Systems—Summary This part of the Samaria Foothills is eminent for its developed agricultural systems, similar to those in many other parts of the central mountains in the Land of Israel. Surveys and excavations showed that the ancient farmers efficiently used the area despite the lack of natural water sources. The exploitation of the land and its maintenance through agricultural terraces and dams were, therefore, a daily necessity. This resulted from a chain of events (historical and others) occurring the Land of Israel, especially in the Roman and Byzantine periods. The increase in settlement distribution created a settlement wave into areas not before occupied, apparently into the eastern part of our region. Large villages—‘rural towns’—as well as monasteries and farms were established during this period, supported by the developed agricultural systems.

Agronomy and Climatology. Every agricultural crop requires certain optimal conditions. Grapes and olive trees can stand harsh terrain, but grain requires slightly different conditions, which had to be addressed by preparing the lands, mainly wide wadis for growing it. History. Historical sources and travelers have often described the landscape of the Land of Israel and its agriculture. Unfortunately, only a few such sources relate to agriculture in our region. Ethnography. Much can be learned from traditional agriculture, still maintained in certain Arab villages. However, contrary to areas where traditional agriculture continued to exist, such as the Sharon Plain and Western Samaria Hills (Roll and Ayalon 1989: 207– 208; Dar 1986c), this is not so in our region.

The location of the various crops is often unclear (see also chapter 7). Examining traditional 171

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods agriculture in Arab villages showed that vines and olive trees can be grown in almost every area. These crops (see also London 2009: 19–110) can be observed mainly in the rocky areas of Arab villages, but also in dammed wadis. Wheat, on the other hand, requires a deeper soil and suitable conditions for working with animals. Therefore, it is almost impossible to grow wheat in rocky areas and steep slopes. Many scholars have dealt with dating the agricultural systems in various parts of the Land of Israel. As excavating these systems, in our region, yielded only worn undatable potsherds, the agricultural systems were dated, in the present study, in relation to the surrounding settlements, the road system, and the results of OSL tests. Agricultural activity in the area during Iron Age II and the Persian–Hellenistic periods was quite limited, especially compared to the following periods. The area was occupied during these periods by several small farms, one fortress, and one satellite village whose inhabitants cultivated the area. Although the inhabitants of these periods probably built agricultural systems, most of those visible today seem to date from the Late Roman period, and especially

from the Byzantine period. Although the number of sites during the Early Islamic period had significantly declined, the area was seemingly still being cultivated, but to an unclear extent and with no major changes. The results of the OSL tests properly reflect the settlement distribution in the region. However, one should be aware of the limits of this method, which sometimes does not allow tracing the changes made over the years to terraces (see also Gibson 2015: 295–314). Therefore, dating the agricultural systems should be based on several factors: understanding the nearby settlements, analyzing the roads connecting the settlements to the fields, tracing the correlation between the size of the settlement and its adjacent fields (see also Chapter 9), and OSL tests. Whenever one or more of these factors is not available, then the prevalent method of excavating probes and collecting potsherds should be applied, nevertheless taking into account that this method is problematic since it can only reveal the periods during which the terraces had been active.

172

CHAPTER 7

SETTLEMENT PATTERNS

Introduction

this size (c. 1 dunam) in our region. This estate was probably owned by the landlord of the Kh. Umm alHammam village or someone on his behalf. As a rule, an estate—or a villa—was located a few hundred meters from the village or within it, as with H. Yeqavim (Site No. 9). Percival (1940: 5) defined the estate as “a farm which is integrated into the social and economic organization of the Roman world.” The term farm, as used in the present work, relates to the main residential building (the only building in most cases), the storage room and the agricultural plot (for other definitions see Zissu 2001: 252–253).

In his 4th-century CE Onomasticon, Eusebius describes the differences in architecture and extension of the various settlement types and defines 155 settlements as ‘villages’ and another 35 as ‘large villages’ (Melamed 1950: 1; Safrai 1994a: 26–27). Modern studies have also discussed the definition of settlement types in the Land of Israel and elsewhere, at length (Table 7.1; Dan 1976, Dar 1982a, 1998; Dar, Tepper and Safrai 1986; 1995a, 1997b, 1998a, 1994b; Hirschfeld 1997; Hingley 1990; Zissu 2001: 249–270; Sion 2001; Safrai and Sion 2007; Leibner 2009, Taxel 2010).

‘Farmhouse’al is usually defined by scholars as a single, simple residential building designed to function as needed by the agricultural activity in its vicinity (Dar 1982a: 9). Sometimes, the farmhouse comprised two or three small structures close to each other, as in Site No. 6 (Wadi al-Bureid [east]). Safrai and Dar (1997a: 8, table) describe the difference between an estate and a farmhouse, defining the latter as “A simple residential structure with no luxurious facilities.” In historical sources, this is called “a single farm”— apoikia or Katemata (Dan 1976: 207–214)—referring to a farmhouse owned by one person, or as the Mishna states: “A city of one individual” (Mishnah, Eruvin 5:11).

Settlement Types The analysis of the sites described in Chapter 3 revealed the existence of the following five main settlement types: 1. Farms and Estates The differences between these two settlement types are not evident, and their definitions vary among scholars. The only site defined in this study as an ‘estate’ (Site No. 10, Wadi al-Bureid [south]) is characterized by a fine ashlar construction not common in buildings of

Table 7.1. Most common terms for settlement types (after Amitzur 2006: 12). Term in this Work Village/town

Hebrew

Greek/Latin

‫עיירה‬/‫ עיר‬komai megistai

Description Mid-sized rural settlement with no polis status.

Satellite village

‫ כפר‬komai

Small-sized settlement associated with a neighboring village or a larger settlement.

Small village

‫ כפר‬komai

Independent small settlement

Farmhouse Estate

agridion

A simple domestic building.

‫ עיר של יחיד‬villa

An isolated luxurious manor.

Hamlet Private village

A cluster of several farmhouses katemata

173

A settlement of tenant farmers owned by a landlord.

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods The study of farmhouses in the Samaria Foothills began in the early 1960s with Yeivin’s and Edelstein’s excavations at Tirat Yehuda, where a large farmhouse (four dunams) dating from the Iron Age through the Hellenistic period was discovered (Yeivin and Edelstein 1970: 56–67). The excavations yielded an oil press dated to the Hellenistic period (Hestrin and Yeivin 1977: 29–31; see also Faust 2006: 482) and coins from the time of Alexander the Great until the Seleucid dynasty (Rahmani 1970: 68–69). This large farmhouse was destroyed during the Hasmonean Revolt. Finkelstein (1978, 1981) dated the farms that extended between Naḥal Qana on the north and Ayalon Valley on the south to the 8th–7th centuries BCE, after the Assyrian conquest. He suggested that farms were established following a land shortage in the Samaria Hills which forced refugees to settle in the study area, where lands were available (Finkelstein 1978: 71). He also suggested that this course of events was made possible thanks to the good prevailing security conditions (Finkelstein 1978: 70) since people in an unstable condition naturally prefer to concentrate in large settlements. Dar (1982a 1982b: 47–60), who surveyed farms in the area of Naḥal Beit ‘Arif, in the southern part of the Samaria Foothills, proposed that each farm was owned by a large family that cultivated a 100–150-dunam area. A comparison with the traditional Arab village in the mountainous regions of the Land of Israel shows that each household owned a 30–50-dunam area (Ben-Herut 1977: 167).

In summary, after reviewing the results of the many studies carried out on the farmhouses phenomenon, it has become clear that this type of settlement was extremely widespread in many regions of the Land of Israel (Dar 1978; Magen and Finkelstein 1993; Leibner 2004; Zertal 1996; Sion 2001; Kochavi and Beit Arieh 1994; Finkelstein and Lederman 1994; Gophna and Bei Arieh 1997; Govrin 1992), and in our region in particular (Harpaz-Segal 2012; Amit 1996a, 1996b; Amit and Zilberbod 1996a: 99-100; Zelinger and Amit 2001: 56–58; Taxel 2006). 2. Satellite Villages The expansion of a particular settlement also increased its agricultural areas which, over time, turned into a multi-crop system. Since this new system led farmers to spend more time in their fields, they sought to dwell closer to these and so have more convenient working conditions and avoid unnecessary transportation expenses. This was achieved by establishing satellite villages, whose foundation depended on several conditions. Safrai (1986a: 31) argued that two preconditions were, first, that security must prevail, and second, that the village must be able to maintain an independent reasonable lifestyle, including public services. Satellite villages identified in our area may, perhaps, meet the definition of a ‘hamlet’—a cluster of farmhouses generally close to each other and densely built (Hirschfeld 1997: 77). On the other hand—though not in our area—there were small villages dispersed over a wide area (‘scattered villages’), such as in the Yattir region (Govrin 1992: 103).

The excavation results of the farmhouse at Site No. 2 (Wadi al_Bureid [510]; (Shadman 2014, Tendler and Shadman 2015; Hadad et al. 2015) enabled reaching several conclusions:

The circumstances and reasons for establishing hamlets are recorded in a 10th-century CE document known as ‘agradion’ (‘small field’; Safrai 1994b: 69–70). Hamlets were conceived as a single farmhouse apparently owned by a nuclear family. It later expanded from a few rooms to several dozens. This type of settlement is represented by the sites of Qurnat al-Haramiya, Khallat es Sihrij, and Kh. Tukkim. The first was defined in past studies as a large site, extending for over 40 dunams, but recent excavations revealed it did not exceed 10 dunams in size (AvnerLevi and Torge 1997, see Site No. 15). Its proximity to Migdal Afeq suggests that it was under the patronage of the latter, at least from the Hellenistic period onward. During the Iron Age–Persian period, Qurnat al-Haramiya seems to have been more important than Migdal Afeq. Two additional settlements northeast of Quenat al-Haramiya probably maintained commercial and social relations with it (Kh. Taha—Finkelstein 1978a: 19, Site 67/3; 1981: 344–345; Kochavi and Beit Arieh 1994: Site 68, and an unnamed site—Finkelstein 1978a: 19, Site 66/9, Table 20; 1981: 344–345; Kochavi and Beit Arieh 1994: Site 87).

A. This administrative building was founded during the Persian period, as opposed to most farmhouses in the area, founded in Iron Age II. B. This was a symmetrical building, constructed as a single, uniform unit bearing no modifications or changes. C. The building comprised two courtyards, one serving as a marketplace, while other farmhouses in the area usually had a single courtyard. D. The building was located on a flat plateau overlooking Wadi al-Bureid and Wadi Susi. The other farms were usually located on moderate and rugged slopes, raised above the alluvial areas. E. This administrative building and the farmhouses in its vicinity were built in a plan that included a central structure, a simple winepress and a water cistern (usually one). More installations were, obviously, in the adjacent open areas. 174

Amit Shadman: Chapter 7, Settlement Patterns The second site, Khallat es Sihrij, located 1.5 km east of Migdal Afeq, was also occupied from the Iron Age through the Hellenistic period and may thus be considered as a satellite village of Migdal Afeq, which was also inhabited in these periods. The third site, Kh. Tukkim, located in the southern part of the study area and extending for c. 3 dunams, had several interconnected building complexes and was most probably a cluster of farmhouses. Based on the rural road-network, this site apparently maintained regular contact with Kh. Ta‘amur to its east. 3. ‘Rural Towns’—Medium-large-Sized Settlements

of the Roman-Byzantine Empire, e.g., England (Hingley 1990: 100). 4. Fortresses Two fortresses were present in the study area: Migdal Afeq (Site No. 14), which controls the Via Maris and the nearby ‘Afeq Passage,’ and ‘Naḥal Shiloh (Southwest) (Site No. 12). Although the archaeological finds discovered so far in Migdal Afeq do not point to the existence of a fortress during the periods under discussion, the historical references and the preserved name ‘tower’ point to the nature of the site. It is, therefore, clear that since its establishment and until at least the Crusader period, the site comprised a fortress and an adjacent civil settlement. In the second site, ‘Naḥal Shiloh (Southwest)’ (Finkelstein 1978: 30, Site 84/1, Table 13, Kochavi and Beit Arieh 1994: Site 189), the square fortress (22 × 22 m) took advantage of the natural terrain and jutted out from the farmhouses, which were usually not built according to a uniform pattern.

Rural

This category has been included because of the confusion regarding the definition of a settlement visà-vis its built-up area, with this issue being quite often based on modern concepts (Hirschfeld 1997: 75–129). Thus, this category relates to the size of the built-up area and the agricultural plot of a site as compared to other sites in the area. Nevertheless, a consensus seemingly exists on the role of the ‘rural towns’ as settlements, where most of their inhabitants made their living off agriculture. This type of settlements exceeded ten dunams in size; thus, a ‘town’ was clearly not defined only by size (Yeivin 1987: 59; see also Taxel 2010: 28–32).

Finkelstein (1978a: 29–30) identified a line of fortresses, most dated to Iron Age II and the Persian period, extending from Naḥal Qana on the north to Naḥal Beit ‘Arif on the south. Each fortress controlled the area between the large streams that flow from east to west (Naḥal Qana, Naḥal Rabah, Naḥal Shiloh, Naḥal Mazor, and Naḥal Beit ‘Arif). The identification of Site No. 12 as a fortress is supported by its similarity to other fortresses, e.g., H. ‘Eres, located on one of the summits of Mt. Ruah (Mazar and Wachtel 2015: 214–244). This fortress, dated to the 4th century BCE, consisted of a square structure (21.8 × 25.5 m) with a central courtyard. Comparing the two sites indicates that they functioned as security observation points overlooking their surroundings.

Four rural towns were present in our study area, all four in its eastern part. The settlements are on moderate slopes or on the margins of small valleys which are less suitable for cultivation. All four settlements were founded in the Byzantine period and were densely built. The northernmost, Kh. Kesfa (Site No. 13), c. 15 dunams in size, is north of Wadi Al-Bureid. South of Wadi al-Bureid is Kh. Umm al-Hammam (Site No. 11), which spreads over 10 dunams. H. Yeqavim—a highly important site in our region—is 1.5 km to the south of Kh. Umm al-Hammam and extends for 15 dunams. Kh. Ta‘amur (Site No. 19), spreading over 10 dunams, is in the southern end of this strip, above the northern bank of Naḥal Shiloh. Each of these settlements maintained commercial and social relations with the surrounding ‘rural towns’ and had satellite farmhouses/estates.

5. Rural Monasteries and Churches Rural monasteries became common in the Land of Israel already in the 4th century CE (Hirschfeld 2002) but were not a new phenomenon. Monasticism, not always religion-based, had been known in Egypt from as early as the 3rd century CE (Hirschfeld 2002; Di Segni 2005: 43). Egyptian monasticism evolved into two main streams. The first, founded by Antony, who lived in the years 251–356 CE, is the eremitic monasticism, i.e., each monk lives separately in his cell and meets his comrades once per day or week. The second stream, relevant to our region and founded and designed by Pachomius (290–346 CE), was the cenobitic monasticism, i.e., monks living in and around the monastery and having communal duties such as common prayers and cultivating the fields. These monasteries were known as Coenobium—a place where monks live communally in a monastic spirit (Patrich 1995: 3; Hirschfeld 2002b: 114). Cenobitic monasticism became widespread

Kochavi and Beit-Arieh (1994: Sites 155, 157, 159, 196; Safrai 1997a: 219, Map 2) surveyed structures in the vicinity of Kh. Umm al-Hammam and associated with this site, and defined them as farmhouses, and a church. An examination of the road network in the area confirms this distinction and the settlement at Kh. Umm al-Hammam seems to have met not only the daily needs of its inhabitants but also maintained mutual contacts with the farms in the vicinity. It is plausible that the wealthy villagers who owned the entire area dwelled in these farms (or estates by a different terminology). Such a model was common in other areas 175

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods across large geographical areas. Hirschfeld (2002b) notes that this phenomenon was prevalent mainly in the desert frontier, but monasteries were also found in the central mountains, the Judean Desert and the Coastal Plain (Taxel 2010: 397). ‘Ad (2000), divided the monasteries in the Samaria Foothills into four groups: 1. Monasteries located within the settlement and serving also as hostels for pilgrims (e.g., H. Zikhrin). 2. Monasteries/churches built at the center of the settlement (e.g., Al-Bireh; Safrai and Dar 1997b). 3. Monasteries built close to the settlement. 4. Isolated monasteries.

1. The presence of fragments of marble, tesserae and other items that point to a Christian population. 2. The general plan of the site is typical of a monastery, i.e., a multi-roomed structure surrounding an inner courtyard. 3. The location of the site—on high cliffs or summits. However, Taxel (2010: 396) noted that these features are also present in sites other than monasteries. Besides the archaeological finds, the preservation of site names and the term ‘Deir’ should be taken into consideration (Elitzur and Ben-David (2007: 13–38), since this term, becoming prevalent probably during the Early Islamic period, refers to a wall or compound. Di Segni (2005: 67; note 8) noted that that term for “monastery” in Syriac is ‘Dyara,’ and the similarity between the two terms is clear. The site of Dayyar/Duweir (Site No. 8) in our area housed a monastery and a church, and its ancient name was preserved. Other similar sites, though not in the study area, are Deir Balut and Deir Qasis, which were part of a large Byzantine settlement complex (see also Chapter 10). It should be noted, however, that there are few exceptional sites named ‘Deir,’ where no monasteries have been identified.

Three sites in the study area were defined as ‘rural monasteries’ or churches. This term is based on the quality of their constructions—fine ashlar and capitals, colored tesserae and roof tiles—and on their similarity to structures excavated in the Samaria Foothills (Dar and Safrai 1984; Greenhut 1998: 121–128; Dahari 1996; 2003; Dahari 2012: 105–124; Dahari and Zelinger 2014: 179–203). Not always can a distinction be made between a small monastery and another building such an estate (Hirschfeld 2002b: 114), and even excavations have not always provided evidence for this distinction. If so, and lacking archaeological evidence, what are the criteria for defining a monastery or church/chapel (Aviam 2003: 41–60) and an estate? While estates were better built and their higher standard clearly distinguishes them from farmhouses, monasteries are located on high hills, c. 1.5 km away from the largest settlement in the area, and are usually interconnected by rural or agricultural roads. H. Te’ena, an isolated monastery, remote from any settlement, is the only one breaking this pattern (see below). Thus, rural monasteries were isolated, larger than churches/chapels, and generally included several wings around a central courtyard (Patrich 1995: 109–110). On the other hand, independent churches/ chapels, such as those located above Kh. Umm alHammam and H. Yeqavim, were small and had a single and meticulously constructed building (sometimes with a peripheral courtyard). These buildings did not provide proper living domestic conditions. In our region, estates were much smaller than monasteries, since monasteries were occupied by more people (some of them apparently temporary) and needed more rooms and wings. In addition, monasteries included churches or chapels, which were relatively larger and differently built, as opposed to the other parts of the complexes. There was no evidence of the existence of a church or a chapel in the estate. However, it is worth noticing that the large estates in Ramat HaNadiv (Hirschfeld 2000) and H. Raqit (Dar 2003) included several features characteristic also of monasteries. Thus, only a spatial analysis of these sites can shed light on this issue. Hirschfeld (1990: 3) noted three markers for identifying monasteries:

H. Te’ena—A Case Study The site is located at the center of the study area. A survey conducted prior to excavations yielded colored tesserae, roof tiles, a pillar for anchoring a beam of an oil press, the tops of well-built walls, and Byzantine sherds. The site is surrounded by a fence, pointing to a possible monastic nature (Chitty 1966: 20). A similar wall is present, for example, at Deir Qla‘ah, which was part of a series of monasteries in the Western Samaria Hills (Hirschfeld 2002a: 157–158). Many remains from the Byzantine period were exposed in the excavations at H. Te’ena, including a basilical church paved with colored mosaic floors with an inscription and geometrical decorations, living rooms, a central courtyard, a stable, water cisterns, and an oil press. Even after the excavations, it was still difficult to define the site as a monastery or as the large estate of a wealthy person which served also as a religious center for the region’s residents. The inscription discovered within the mosaic floor of the church mentioned a priest and not an abbot (Di Segni, oral communication), suggesting the complex was an estate and not a monastery. The rural road from H. Te’ena to Deir Balut is another feature indicating the importance of the site. The site is about 1–2 km away from any other settlement, while churches/chapels that served the local population were found, in surveys, in most of the other large settlements, such as Kh. Kesfa, Kh. Umm alHammam and H. Yeqavim.

176

Amit Shadman: Chapter 7, Settlement Patterns Taxel (2005: 125–126) referred to the difficulty in distinguishing between a mansion and a monastery, even after such a building had been excavated. He correctly notes that it is plausible the cenobitic monasteries were isolated, at some distance from a settlement, as with H. Te’ena. Moreover, the built-up area of the entire site is at least 3–4 dunams, and there is no nearby isolated estate with buildings densely concentrated around a church. For example, in Kh. elBira (Safrai and Dar 1997b), an estate and a church are present, with some other buildings nearby. A different pattern was found in H. Tinshemet, where a church and an oil press, but no domestic rooms, were found (Dahari 1996: 67–68; Dahari 2012: 105–124). Thus, H. Tinshemet should not be considered a monastery but rather a church that apparently served the surrounding communities. Finally, H. Te’ena is situated on a hilltop, characteristic of monasteries, contrasting with the other nearby settlements usually built on moderate sloping ridges.

Finkelstein (1978a, 1981) and Kochavi and Beit Arieh (1994), and later on, through the excavations carried out by the author. The large (non-farm) settlements of this period, such as Migdal Afeq, are still active. Twelve sites were active in the ‘farm strip’ during this period1: eleven continuing without interruption into the Hellenistic period, and one site (No. 16, Elevation Point 136) founded in the latter. Only one site (No. 12, Naḥal Shiloh [southwest]) active during the Persian period ceased in the later period. Most farms operating in the Hellenistic period did not continue into the Roman period, except for Site No. 12, where potsherds from the Roman, Byzantine and Early Islamic periods were found. Settlement Types. During this period, the region was occupied mainly by agricultural farms. Eight sites were identified as farms while the others were most likely larger settlements (perhaps a cluster of farms, as noted at H. Mazor; see Chapter 8). The largest settlements include Migdal Afeq (Site No. 14), Quenat al-Haramiya (Site No. 15) and Kh. Taha (Kochavi and Beit Arieh 1994: Site 68). The built-up area of the farms measured usually 1–2.5 dunams. The buildings differ in plan within each farm but have similar components (such as the width of walls, entrances, and courtyards).

In summary, the case of H. Te’ena shows that excavating a site does not necessarily shed light on its nature. Although we can identify Christian sites through excavations and even surveys, we should be careful when defining them. It seems that H. Te’ena was a monastery, but even if it were an estate, one should consider it as an active unit that cultivated the surrounding lands. Dan (1976: 187) wrote that “… the philosophers of the early monasticism considered the physical work in general, and working the land in particular, of high value to the monks” (Dan 1976: 187), and this is also evident in our study area. Settlement Patterns Through The Persian Period (Fig. 7.1)

the

The buildings were usually constructed of dry masonry, with plaster noted only in one site (No. 2). The walls, especially the exterior ones, were double walls made of medium- and large-sized fieldstones. The buildings’ interior included similar components: the floors were usually made of small-sized fieldstones laid on a bedding of local soil filling the gaps in the bedrock. The rooms were of various sizes (9–90 sq m), with no single pattern. The entrances (width 1.5–2.0 m) were in the outer walls or in partition walls between the residential wings and the courtyards (see below). The inner entrances, between the interior rooms, were smaller than the outer ones (width c. 0.8–1.0 m).

Periods

Thirteen sites from this period were discovered, nine of which were agricultural farms. Except for one site (No. 2) initially founded in the Persian period, all the others were built on Iron Age sites. These rural-agricultural farms had a built-up area measuring about 1–3 dunams. The main crops were seemingly grain, vine, and sheep, and the many millstones found in several farmhouses attest to the production of flour. In Site No. 4 (Naḥal Shiloh). the building contained a large quantity of mortaria bowls used for grinding/crushing. Water cisterns and simple winepresses were also discerned next to the farms. Most sites continued into the Hellenistic period, though the Persian settlement was apparently one of the peak periods in our region (see also chapter 10).

Site No. 5 (Naḥal Shiloh (West) seems to have had a second floor which was probably built from consumable materials (a kind of gallery). The roofs of the farmhouses were also made of consumable materials, as evidenced by the roller designed to compress the roof found in Site No. 20 (Naḥal Shiloh [3]). The use of such a roller in a traditional Arab village was presented by Ilan (1984: 28). Most farmhouses had one or two courtyards (14 × 17.0–25.5 × 26.5 m) either adjacent to the buildings or incorporated into them, and their floors were leveled with local fieldstones and soil. The southern courtyard in Site No. 2 (Wadi al-Bureid) was detached from the

The Hellenistic Period (Fig. 7.2) The agricultural farms of this period were a direct continuation of those of the Persian period. Initial information regarding these farms was obtained by

1 

Three sites were not included in the present study: A farmhouse (Finkelstein 1978: 11, Table 15; Kochavi and Beit Arieh 1994: Site 138), which was poorly preserved, and two additional sites (Kh. Taha and Site 87; Kochavi and Beit Arieh 1994) that were completely destroyed.

177

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods

Fig. 7.1. Settlement distribution in the Persian period.

178

Amit Shadman: Chapter 7, Settlement Patterns

Fig. 7.2. Settlement distribution in the Hellenistic period.

179

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods residential wing, suggesting that it functioned as an animals’ pen or as a local marketplace. The sitting bath discovered in this site may attest to the residents’ economic/social status.

in courtyards within the farms or adjacent to them. 4. Leaving unattended buildings for long periods of time may attract bandits and lawbreakers. Topographic location. Farmhouses were almost always on moderate hillsides surrounding the alluvial terrain. This preference can be understood based on the following factors: 1. This location enabled keeping the alluvial areas for agriculture. 2. This location is benefited by the better drainage of the rainfall into water cisterns. Naturally, hewing these cisterns could be done in the rocky terrain (Finkelstein 1981: 343). 3. The wind disturbs less in the hillsides. 4. Cutting installations, such as winepresses and cupmarks, was an easier task in rocky surfaces.

Abandonment/deportation. Another common feature discerned during the excavation of the Hellenistic farmhouses is the intentional blockage of the rooms’ entrances. This find, and the fact that no signs of destruction and hardly any ceramics were found in situ, indicate that the farms were abandoned in an orderly fashion and that their inhabitants took their belongings with them. Blocking the entrances was probably intended to prevent access to hostile elements, including animals, and to stabilize the walls of the buildings during earthquakes. Seasonal or permanent structures? The issue of how long the farms were occupied during the year raises several questions. There are many possible reasons for their abandonment, such as political events (see Chapter 10) forcing the farmers to leave them. In such a case, it is likely they had time to block the buildings’ openings, perhaps planning to return. Another reason for abandoning the farms could have been an economic crisis and/or drought.

The agricultural plot. Hellenistic-period sites were scattered over an area of about 12 sq km, less than half of the total study area. This area included also the large settlements in the ‘Western Strip.’ The agricultural plot of each settlement in the ‘Farm Strip,’ east of Migdal Afeq during this period, was about 400 dunams. This estimation is based on the walking distances (not exceeding half an hour) from each farmhouse to its fields, and on the relative location of other farms in the area.

It has also been suggested that the farms were actually seasonal settlements and that during the offseason their inhabitants returned to the large settlements in the vicinity. However, several factors suggest that the farmhouses were permanent settlements:

Interconnections with a large settlement. Settlement hierarchy in the area can be divided into three “circles”: 1. ‘The inner circle,’ which included an ‘administrative building’ (see below) and other farms nearby.

1. Their construction required considerable investment. Although the buildings were built of partially worked fieldstones and with no signs of luxury (e.g., as seen in estates), much time had been evidently been invested in building them for the residence of an extended family. In addition, the water cisterns hewn near the farmhouses also indicate a prolonged stay. It is unlikely, therefore, to build such structures for occupying them only a few months per year.

2. ‘The outer circle,’ which included the communities of Migdal Afeq and Qurnat al-Haramiya. 3. ‘The general administrative circle,’ represented by the city of Pegai (Antipatris), which controlled the entire region. The first circle is represented by the structure found in Site No. 2, which was founded in the Persian period and continued into the Hellenistic period. The structure was designed following a unique plan similar to that of administrative buildings and palaces appearing first in the Neo-Assyrian period in Mesopotamia, developed in the Neo-Babylonian period, and used in the Land of Israel during the Persian period (Reich 1992: 119). Its plan, with a larger built-up area, differed from that of other farmhouses in the region. The building was apparently used for administering the region and the farms in it (Hadad et al. 2015: 50–68; Tendler and Shedman 2015: 193–207).

2. Work in the fields was year-round. At the end of the harvest season (grain, olives, and grapes—the main agricultural activities in the area), maintenance works in the fields had still to be carried out to prepare them for the next season. These tasks required much labor force, and it is unlikely for it to be located outside the area, within a walking distance of a few hours. 3. Raising sheep and grazing was an additional central branch in antiquity (Avitsur 1976: 64). Naturally, livestock required annual and continuous work carried out in the open areas, where the farms were located. It should also be noted that no animal pens were identified in the farms’ vicinity, meaning that the animals were stored

The second circle included ‘The Western Strip,’ ‘The Farms Strip,’ and ‘The Eastern Strip’. 180

Amit Shadman: Chapter 7, Settlement Patterns Since ‘The Farms Strip’ was in close proximity to ‘The Western Strip,’ the farms apparently maintained connections with the communities of Qurnat alHaramiya, and especially with Migdal Afeq, which was a major settlement also in the Hellenistic period. The third circle is discussed in detail below, but as our region was close to the Hellenistic city of Pegai (Beck and Kochavi 1993: 70), it may have served as the main agricultural hinterland of this city during the Hellenistic period.

were also discovered. Cooking pots with a wide thin rim and high-quality pots with thin walls were also found. Torpedo jars, originating in the Persian period, and bag-shaped jars with a thick rim were the main jar types. Red- and black-slipped oil lamps were associated with the ‘Attic vessels’ (5th–3rd centuries BCE). Site No. 5 yielded mainly bag-shaped jars, and this site may have been active only during part of the Hellenistic period. Coins dated to the Hellenistic period were found in all agricultural installations and farmhouses. A 4th-century BCE coin minted in Sidon, a 4th-century silver Athenian Tetra-Drachma, and a bronze coin dated to Ptolemy I (305–283 BCE) were found in Site No. 4. A silver coin—Tetra-Drachma—from the time of Alexander the Great (325 BCE) was found in Site No. 5. A coin from the time of Ptolemy I (294–282 BCE), minted in Alexandria, was found in Site No. 2. Other coins, dated mainly to the 3rd century BCE, were discovered in Quenat al-Haramiya.

Chronology. Excavations in some farmhouses yielded results similar to those revealed in the surveys, i.e., they were occupied during Iron Age II and the Persian and Hellenistic periods. In Site 2, for example, the surveyors dated the potsherds to these same periods (Finkelstein 1978a: 11–12, Table 16, 1981: 338–339, Fig. 6; Kochavi and Beit-Arieh 1994: Site No. 145), while the excavations showed that the farmhouse operated mainly during the Persian (5th–4th centuries BCE) and Hellenistic periods (3rd–2nd centuries BCE), while Iron Age potsherds were absent. Our survey of Site No. 5 dated it from Iron Age II to the Hellenistic period, like the previous surveys at the site (Finkelstein 1978a: 11, Pl. 17; 1981: 332, 342–343; Kochavi and Beit Arieh 1994: site no. 142), and the excavations showed that it had indeed been occupied throughout these periods, although a Byzantine limekiln was also found. The excavation of Site No. 4 revealed that it was founded in Iron Age II and ceased to operate at the end of the 3rd century BCE, while a large limekiln was built atop the ruins of the building, probably in the Ottoman period. Kochavi and Beit Arieh (1994) reported that Site 147 (our Site No. 16) was occupied in the Byzantine period, but the excavation showed that the farm had been initially settled in the Hellenistic period (the only site founded in this period), and that during the Byzantine period it was probably not used for residential purposes.

Agricultural Crops. Archaeological excavation results attest to the farmhouses’ inhabitants growing mainly grain, vine, and sheep. Basalt millstones, used to produce flour, were discovered in the excavations, many of them in Sites Nos. 4 and 5. These installations seemingly belong to the ‘Hopper Rubber’ (Alynthus mill) or ‘Saddle Quern’ types and were common in the Land of Israel, especially in Iron Age II (8th century BCE) and the Persian period, probably continuing into the Hellenistic period, as in Greece and Italy (Avitsur 1976: 74–75, Fig. 208; Frankel 2003: 43–60). No circular rotating installations were found. In addition, many cupmarks hewn in the bedrock were exposed inside the buildings. These installations may have also been used for grinding/crushing, as attested by a pestle discovered in one of them. The wine industry was evidenced by simple winepresses found near the farms and comprising a treading floor and a collecting vat. Other winepresses were found in open areas, within the vineyards. In most cases, these winepresses were poorly hewn and irregularly shaped, and remains of plaster were traced only a few times. Presumably, these winepresses originated in the Iron Age, being modified and extended at a later stage, as were also the water cisterns. Raising sheep was undoubtedly an important economic activity in rural communities in the Hellenistic period, as evidenced by the large animal pens annexed to the farmhouses. These pens occupied almost half of the entire built-up area of the farms. Ben-Ari (1989) showed that a 200-sq m pen can accommodate about 130 sheep. In some cases, troughs were cut beside the cisterns, providing water to the flocks. Presumably, producing olive oil was another sector of the local economy, although no large-scale oil presses from this period, like those at Tirat Yehuda (Hestrin and Yeivin

In summary, the settlements occupied in the Hellenistic period were active mainly in its early stages, at the end of the 4th century and the 3rd century BCE, with only a few ceramics dated to the beginning of the 2nd century BCE. It seems, therefore, that the area was not settled in the Hasmonean period. Distinctive pottery and coins. Finkelstein (1978a: 50–51) dated the Hellenistic period in the region to its early stages and emphasized the absence of Terra Sigillata. His dating was based mainly on ‘fish bowls’ and ‘Attic vessels’, including the black bowls that appear in Samaria in the 4th century BCE. A considerable number of ‘Attic vessels’ was found in the excavations conducted during the present project. Black-slipped ‘Attic vessels’ were found in Site No. 2; this type first appeared in the Persian period and continued into the Hellenistic period (3rd century BCE). Similarly, 3rd century BCE red-slipped bowls and local imitations 181

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods 1977: 29–31) or Qasr al-Leja (Dar 1986a: 10–12) were found in the study area. Simple oil presses (bodedah in Mishnaic Hebrew), attesting to oil production, were recorded only at a few sites (e.g., Site No. 4). These may have been founded in Iron Age II (Eitam 1979: 146–155) and did not yield commercial quantities.

rural occupation in the area during the Roman period, its size, its Jewish population and the sequence of events. Josephus noted that the Jewish inhabitants of the area dispersed and that the towns in the area were destroyed It is not clear which communities Josephus speaks of since no destroyed settlements were discerned in our area, neither from the Roman or other periods. Located about one kilometer southeast of Migdal Afeq, Kh. Zikhrin may be mentioned in this context, as it overlooked the Antipatris–Gophna road approaching Jerusalem. The intensive excavations carried out here showed that it was occupied mainly in the Byzantine period (Fisher 1983, 1984, 1985, 1989; Taxel 2005). The site was also inhabited during the Roman period, probably as a farmhouse or road station. No destruction was identified during excavations that could be directly related to the Great Revolt. It seems, therefore, that the residents were forced to leave their home, and it is unclear whether they ever returned. As for Migdal Afeq, according to Josephus, the site was destroyed during the revolt, but no archaeological traces of this have so far been found.

Tombs. Except for two cist tombs discovered southwest of Qasr es-Sitt (Site No. 1), no tombs associated with either the farmhouses nor the large sites of Midgal Afeq and Quenat al-Haramiya (Avner-Levi and Torge 1997; Torge 2007) were discovered. Nevertheless, it is cautiously suggested that the Roman burials found at Migdal Afeq (Tsuk 1993: 67–76) were originally from the Hellenistic period. These included a burial cave and a kokhim cave, which were also common in the Hellenistic period (Tal 2006: 217–269). In addition, the Late Islamic period cemetery excavated north of Migdal Afeq (Shadman 2014) was founded on earlier unexcavated burial caves presumably serving the inhabitants of Migdal Afeq, perhaps even in the Hellenistic period. Cemeteries were usually remote from settlements, because of land shortage or rock formation, as seen in the surroundings of the nearby city of Lod during the Second Temple period (Zelinger 2009: 133–134).

Topographic location. The settlements of the period were built in places overlooking their surroundings, such as Site No. 12, situated on a high steep summit, and Migdal Afeq, also located on one of the high peaks above the alluvial plain, overlooking the area and the Via Maris at its foothill. The preference for high locations in the Roman period was perhaps the result of security instability.

The Roman Period (Fig. 7.3) The architectural, numismatic and ceramic finds of this period are very meager. This phenomenon was also recorded by Kochavi and Beit Arieh (1994: 126, map 3), who traced remains from this period only in the area of Migdal Afeq and Quarnat al-Haramiya. The lack of material culture reflects a relatively small area and points to a clear poor occupation in the study area.

The agricultural plot. The absence of settlements suggests that most of the land was used for agriculture. Excavations in the installations yielded potsherds from the Roman period—although only a few, compared to the Byzantine period—attesting to the agricultural use of the area during this period. The agricultural plot in the Roman period is roughly estimated to c. 16 sq km. The area seemingly served ‘the eastern strip’ withDeir Balut and Rafat, on the one side, and ‘the western strip’ with Migdal Afeq and Qurnat al-Haramiya on the other.

Settlements Types. Besides the main settlements at Migdal Afeq, Qurnat al-Haramiya, H. Yeqavim and Kh. Ta‘amur mentioned above, another site (No. 12) was active on one of the high hills in the area. Based on its location and plan, it was defined as a fortress, overlooking the entire area, and especially its road network. According to the ceramic and burial archaeological evidence from Migdal Afeq and Qurnat al-Haramiya (see Chapter 3 Sites 14, 15, respectively), there were seemingly two large settlements in the Roman period, especially at Migdal Afeq. The area to its east was mostly used for agriculture, but, unlike other periods, no farms or small rural settlements were present in our area during the Roman period.

Interconnections and affinities with the large settlements. The absence of other sites in the area in this period suggests that the main settlement was at Migdal Afeq, whose residents also or mainly engaged in agriculture. Excavations conducted in recent years to its north and northeast (Shadman 2014; Tendler, 2014) showed that the settlement was much larger than it appears today. Tendler (2014) exposed architectural remains northeast of this site and potsherds from various periods, including the Roman, pointing to the extensive area of the settlement during this period. Other Migdal Afeq was, of course, the great city of Antipatris, which was an important urban center (Beck and Kochavi 1993: 70–71).

Abandonment / Deportation. “... when he was informed that there was a great body of Jewish forces gotten together in a certain tower called Aphek, he sent a party before to fight them; but this party dispersed the Jews by affrighting them before it came to a battle…” (Wars II, 19, 1). This citation apparently attests to the 182

Amit Shadman: Chapter 7, Settlement Patterns

Fig. 7.3. Settlement distribution in the Roman period.

183

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods Pottery, a coin, and chronology. A few potsherds (mostly worn) from the Roman period were found in excavations in the agricultural areas east of Migdal Afeq, in the settlement and its immediate vicinity. Potsherds dating to the Roman period were found east of H. Te’ena, including cooking pots and a few amphorae. A few other finds were unearthed in the vicinity of Migdal Afeq and in its burial caves. Potsherds dated to the 1st–3rd centuries CE, and oil lamps dated to the 2nd–3rd centuries CE (Tsuk 1993) were discovered in two of the caves, together with four small fragments of stone ossuaries (see below). Also, a few Late Roman potsherds were discerned in Kh. Ta‘amur (Site No. 19). Tendler (2014: Fig. 4), who excavated southeast of the burial caves mentioned above, exposed cooking pots, jars, and juglets dated to earlier than 70 CE. A single bronze coin, dated to the time of Agrippa I, minted in Jerusalem, was discovered in Site No. 2. The 1st-century CE settlement at Migdal Afeq is evidenced in Josephus’ description of the Jewish population in place during Galus’ campaign (Wars II, 19, 1).

and estates. The archaeological evidence attests to the Christian characteristics of these settlements (see below). The number of sites and their establishment; continuity. Twenty sites were found in the area, reflecting a dramatic increase in the total number of settlements—15 more than in the previous period. Eleven sites were founded in the Byzantine period, five continued from the Roman period and the rest were built atop sites from the Persian or Hellenistic periods. Five types of settlements were identified in the area during this period: 1. ‘Rural towns.’ This type includes four settlements: Kh. Kesfa, Kh. Umm al-Hammam, H. Yeqavim, and Kh. Ta‘amur. These sites are located along longitude 200, and each had an about 10–15 dunams densely occupied built-up area, especially Kh. Ta‘amur. The walls of the buildings, which sometimes survived to a height of several courses, were mostly built of partially hewn stones (some of these 1.5 m long or more), often with drafted margins and characterizing also the Byzantine period. The site’s plans show several complexes, dozens of rooms and other spaces, and many water cisterns. Courtyards integrated into the residential quarters were identified at Kh. Kesfa, Kh. Umm alHammam and H. Yeqavim, and a marketplace and a street at Kh. Ta‘amur. Other than that, no pre-planned streets or other structures were identified, neither were the entrances/doorways leading to the road (H. Ta‘amur and Kh. Umm al-Hammam). It is evident, therefore, that the builders of the ‘rural towns’ had settled in topographically convenient places related to agricultural areas, but there was also some central directive for selecting locations for the settlements (see Chapter 10). Churches were discerned above H. Yeqavim and Kh. Umm al-Hammam and Guérin (1984) mentioned a church at Kh. Kesfa which was not found in the present study.

Stone vessels. Fragments of a base and handle of a ‘measuring cup’ were discovered northeast of Migdal Afeq (Tendler 2014: Fig. 4: 18–21). These vessels point to a Jewish population living there during the Second Temple period and meticulously observing the laws of impurity and purity. Stone vessels were also found in nearby H. Zikhrin (Adler 2011, 370: 370) and Antipatris (Magen 2002c: 168). Tombs. Two tombs dated to the Roman period were excavated by Tsuk (1993) north of Migdal Afeq. The first was a kokhim cave, with an entrance on the east leading through six steps into a central square chamber. Two burial kokhim (length 1.85–2.00 m, width c. 0.5 m) were in each of the northern, western and southern walls of the chamber, and a collecting pit was found cut into the floor. The finds contained glass vessels, an arrowhead, a ring, a bracelet, beads, oil lamps and fragments of stone ossuaries dated to the Second Temple period. The second cave was entered through an elaborate doorway leading to a front room, north of which was a burial chamber. This chamber was mostly destroyed, and no kokhim were discerned, except for an arcosolium which was later converted into a cistern. Although potsherds from the Byzantine and Early Islamic periods were recovered, the cave was dated, on an architectural basis, to the Second Temple period.

2. Satellite villages. The proximity of Qurnat alHaramiya to Migdal Afeq points to former being subjected to the latter. H. Tukkim was a satellite village of Kh. Ta‘amur, interconnected through rural roads. The plan of H. Tukkim shows a small settlement (c. 3 dunams), and based on the presence of oil and wine presses, it was probably self-sustaining. Its wellpreserved buildings with plaster remnants showed high-quality construction.

The Byzantine period (Fig. 7.4)

3. Farmhouses/estates. Three farms and one estate were present in the area during this period. The farms (Sites Nos. 6, 16, 21) were simply built, usually using fieldstones. The estate (Site No. 10), on the other hand, was built of very large high-quality ashlars. Well-cut architectural fragments, including thresholds and lintels, scattered throughout the site, pointed to a

Major occupational changes took place during this period. Large settlements (‘rural towns’) were established, mainly between the ‘farms strip’ and the ‘eastern strip’ which dominated large agricultural areas. Smaller buildings (c. 1–2 dunams) were also constructed, identified as monasteries, farmhouses, 184

Amit Shadman: Chapter 7, Settlement Patterns

Fig. 7.4. Settlement distribution in the Byzantine period.

185

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods magnificent building standing here compared to the rural environment. It should be noted that, even in the nearby large settlement of Kh. Umm al-Hammam, no similar construction was identified. Presumably, a wealthy person, perhaps the master of the village of Kh. Umm al-Hammam, or someone on his behalf, dwelled there.

Yeqavim). The churches/chapels were always located on summits, high above the settlement (north of H. Yeqavim and southeast of Kh. Umm al-Hammam). Monasteries were also on the high hills, somewhat remote from the nearby large settlement (Kh. Umm alBureid overlooked Wadi al-Susi and Wadi al-Bureid; Kh. Dayyar overlooked the area south of Naḥal Rabah; H. Te’ena overlooked the tributaries of Naḥal Shiloh).

4. Monasteries. Three monasteries(?) existed in the area during the Byzantine period: H. Te’ena, (Site No. 3), Kh. Umm al-Bureid (Site No. 7) and Kh. Dayyar (Site No. 8). These sites shared similar characteristics, such as high construction quality, colored tesserae, and roof tiles. The results of the excavations at H. Te’ena support this identification.

The agricultural plot. During this period, most of the area was seemingly used for agriculture, with only a few exceptions, such as banks of steep gullies, where no evidence of human activity was discerned. Many agricultural systems spread over the area, including terraces, dams, and rural and agricultural roads. Despite the difficulty in dating these systems, most were seemingly either built or renovated/expanded in the Byzantine period.

5. Fortresses. Two fortresses were identified: one at Migdal Afeq (Site No. 14) and the other at Site No. 12 (Naḥal Shiloh (Southwest). Both were strategically located, but at Migdal Afeq there was apparently also a large civil settlement that included a church.

The agricultural plots of the ‘rural towns’ of Kh. Kesfa, Kh. Umm al-Hammam, H. Yeqavim, and Kh. Ta‘amur totaled c. 10,000 dunams, and included, in most cases, farmhouses and an estate associated with them (Fig. 7.5). Thus, each ‘rural town’ had an agricultural plot measuring c. 2,500 dunams. Enclosed plots of various sizes were adjacent to most Byzantine settlements, especially in Kh. Kesfa and H. Yeqavim. These plots

Topographic location. Examining the location of the Byzantine sites within the rural landscape showed no clear settlement pattern. The ‘rural towns’ were located on hillsides (e.g., Kh. Umm al-Hammam), on moderate ridges (Kh. Kesfa) and in small valleys (H.

Fig. 7.5. The boundaries of the Byzantine settlements in the study area.

186

Amit Shadman: Chapter 7, Settlement Patterns were seemingly divided among the residents under some central guidance (Figs. 7.5, 7.6). The correlation between the areas of the large settlements and their agricultural plots was 10–15 dunams of built-up area against 2500–3000 dunams of fields. A similar division was noticed in other sites, such as in Kh. Karak in the Carmel (Dar and Ziegelmann 1997: 184), and Susya in the Judean Hills (Baruch 2009: 333), although the built-up area in the latter measured only c. 75 dunams. Finally, west of ‘the rural-towns strip,’ the agricultural plots of H. Te’ena, Kh. Dayyar, Qurnat al-Haramiya, and Migdal Afeq totaled 12,000 dunams. Thus, the total area of the settlements and their agricultural plots in the Byzantine period totaled c. 22,000 dunams.

farmers of the latter. However, it is clear that there was an extensive system of barter, social relations, and religious services (churches) throughout the region. Relative chronology. Based on the ceramic finds, most sites were seemingly founded during the 5th century CE and continued to be active until the 7th century CE. However, it became clear that in Kh. Umm al-Hammam and Kh. Ta‘amur, the Byzantine occupation had begun only during the 6th century CE, also continuing into the 7th century CE. Distinctive pottery and coins. A variety of pottery vessels dated to the 5th–7th centuries CE were found. The common types were bag-shaped jars and Late Roman C bowls, many with cross imprints on their bases. A cooking pot (Kh. Umm al-Hammam), juglets and lamps (Kh. Umm al-Bureid) and kraters (H. Yeqavim) were also discovered. A bronze coin from the time of Justinian I (518–527 CE) and minted in Constantinople was found in H. Te’ena.

Interconnections and affinities with the large settlements. As with the Hellenistic and Roman periods, Migdal Afeq and Antipatris were important factors in the region also in the Byzantine period. These settlements maintained connections with ‘the eastern strip of settlements’—Deir Balut and Rafat—as demonstrated by the rural road network. Based on the road network and the location of the settlements, the region was divided as follows:

Agricultural crops. As in earlier periods in our region, the main crops in the Byzantine period seem to have been vines, olives and grain, and sheep-growing and herding were also practiced. The numerous winepresses present within the agricultural plots of the Byzantine sites, together with the field towers, indicate that vineyards and wine production were widespread. The winepresses found in the settlements and their surroundings were more developed and complex than those in the farms and included mosaics and plaster floors. The largest and most elaborate winepress in the study area was discovered in the western part of Kh. Kesfa.

The relative proximity of ‘the rural-towns strip’ to Deir Balut and Rafat (2–3 km) presumably underline their affinity, and roads Nos. 3, 5, 9, and 10 may support this assumption. H. Te’ena, as a central site located between ‘the rural-towns strip’ and ‘the western settlement strip’, was apparently connected also to Deir Balut, as possibly illustrated by Road No. 3 which connected the two. The settlements in the western part of the study area were apparently associated with Migdal Afeq, whereas the vast area west of H. Te’ena was used by

dunam

Enclosed plots (dunam)

Kh. Umm al-Hammam

Afeq

Kh. Ta‘amur

H. Te‘ena

H. Yeqavim

Fig. 7.6. Distribution of enclosed plots in the Byzantine period.

187

Kh. Kesfa

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods Table 7.2. The distribution of oil presses in the Byzantine sites.

It should be noted that oil production was more widespread than the wine industry. Evidently, the oil industry prospered during this period. A total of 21 oil presses were recorded (Table 7.2). Large olive presses were found mostly in the ‘rural towns,’ some well preserved. Apparently, in each large Byzantine (mainly) and Early Islamic settlement were 3–5 oil presses, as in Western Galilee (Aviam 1994: 129). Several presses were excavated in Kh. Kesfa (Site No. 13) and were dated to the Early Islamic period (Magen 2008b: 275–278). However, since the results of both previous surveys and the present study show that the area was mainly occupied in the Byzantine period, it is reasonable to associate the oil presses with this period. In addition, piers for anchoring beams were found in this site in secondary use, suggesting the oil presses may have gone out of use during the Early Islamic period. A 5thcentury CE oil press excavated in H. Te’ena went out of use at the end of the 6th century CE. Other olive presses were also discovered in the villages at H. Tukkim, Kh. Umm al-Bureid, Wadi al-Bureid (South), and Wadi alBureid (East).

The site

Size (dunams)

Number of Presses per Presses dunam

Kh. Kesfa

15

5

1/3

Kh. Umm alHammam

10

4

1/2.5

H. Yeqavim

15

4

1/3.75

H. Ta‘amur

10

2

1/5

Kh. Umm alBureid (east)

1

1

1/1

Kh. Umm alBureid (south)

1

1

1/1

Kh. Tukkim

3

1

1/3

H. Te’ena

5

1

1/5

Migdal Afek

c. 35

1

?

floor found north of H. Te’ena reflect this reality. The typical threshing floor (c. 12–18 m in diameter) in our region comprised a symmetrical perimeter wall usually built to a height of one course and a surface of small stones and well-beaten earth. Operating the floors used the wind for winnowing grain, thus the location of these floors corresponded to the prevailing winds in our region, following the common ancient practice (Avitsur 1984: 11). In other words, threshing floors were not discovered west of settlements from which wind blew (see Ayalon 1997a: 209–227).

So many Byzantine oil presses raise the following question: Was this a self-sustaining industry or was the oil marketed to the surrounding communities, or even outside the area? For example, in Kh. Kesfa were five oil presses that produced a much larger quantity than needed for local consumption (Dar 1982a; Safrai 1987: 176–182; Mattingly 1988: 177– 195; Ben-David 1989: 88–100). Since oil presses were found in most settlements, including farmhouses, monasteries and an estate, it would seem production was higher and beyond the regional consumption levels. In addition, many oil presses were discovered both east (Western Samaria Hills; Dar 1982a) and west (Coastal Plain; Roll and Ayalon 1989: 199–207) of our area. Therefore, there was no reason to market oil from our area to communities either east or west of it, but it cannot be ruled out that oil produced in our region may have been marketed even overseas.

The existence of animal pens and troughs attest to sheep-growing as another important sector in the regional economy and to the region being a major supplier of meat and milk. Animal pens were recorded near the residential quarters of large settlements, such as in H. Yeqavim. A well-preserved round pen was discovered west of Kh. Kesfa. A sheep pen was discerned in a cave, east of the main residential complex at Kh. Umm al-Hamam. This pattern is still practiced in the traditional Arab village (Ben-Ari 1989: 183). Tombs. The map in Fig. 7.7 shows that most tombs were located close to the Byzantine sites and may, therefore, be dated to this period (Table 7.3). An impressive burial complex was found west of H. Yeqavim, including the lid of a stone sarcophagus carved with a cross (Fig. 7.8). Several burial complexes are located near Kh. Kesfa, one of these adjacent to the settlement to the south and two larger ones to its west. Common tomb types were cist graves and arcosolia, both widespread in the Land of Israel during the Second Temple period (Tsafrir 1984: 144; Taxel 2010: 227f–270). An arcosolium burial cave was excavated east of Migdal Afeq (Table 7.3, No. 52), where human bones of male and female adults were discovered, not in articulation, probably due to ancient robbery. A small stone lid, a bronze pin and

Grain was grown in the deep soil of the wadis outlets and in small pocket valleys where alluvium had deposited. The soil in these areas is suitable for grain production and its depth enabled plowing with animals, in contrast to the rocky areas in the Samaria Hills which are more suitable for growing olives and grapes. Accordingly, historical sources, especially those from the end of the Byzantine period onward, hardly mention grain fields in the Samaria Hills but rather vine and olives plantations. A Crusader source from 1122 also noted that several settlements in the Samaria Foothills, such as Mirabel, Migdal Afeq and Rentie, supplied wheat to the Nablus area (Amar 2000). The series of threshing floors recorded east of Kh. Umm al-Hammam and the 188

Amit Shadman: Chapter 7, Settlement Patterns

Fig. 7.7 The distribution of graves in the study area.

189

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods Table 7.3. Tombs in the study area. No.

No. in Rosh Ha‘ayin map

1.

Size

Associated site

Notes

89

Qurnat al-Haramiya

Decorated arcolosium cave

2.

89

Qurnat al-Haramiya

Shaft tomb

3.

89

Qurnat al-Haramiya

Stepped burial

4.

89

Qurnat al-Haramiya

Niche for the rolling stone

5.

89

Qurnat al-Haramiya

6.

115

0.56 × 1.63 m

Kh. Kesfa

7.

115

0.58 × 1.34 m

Kh. Kesfa

8.

115

0.47 × 1.77 m

Kh. Kesfa

9.

115

0.54 × 1.78 m

Kh. Kesfa

10.

Kh. Kesfa

Arcolosium

11.

Kh. Kesfa

Arcolosium

12.

120

Kh. Kesfa

13.

120

Kh. Kesfa

14.

121

0.60 × 1.90 m

Kh. Kesfa

Arcolosium

15.

121

0.60 × 1.90 m

Kh. Kesfa

Arcolosium

16.

121

Kh. Kesfa

Cist grave

17.

121

Kh. Kesfa

Cist grave

18.

121

Kh. Kesfa

19.

128

Migdal Afek

20.

132

Migdal Afek

Arcolosium

21.

132

Migdal Afek

Cist grave

22.

132

Migdal Afek

Plastered grave

23.

132

Migdal Afek

Double arcolosia

24.

133

Migdal Afek

Burial caves on a Muslim cemetery (not excavated)

25.

134

2.95 × 2.95 m; 2 m long kokhim

Migdal Afek

Burial cave with 6 kokhim; fragments of ossuaries and Roman lamp

26.

134

Front room 2.6 × 4.8 m; opening 0.32 × 1.35 m

Migdal Afek

Tzuk 1993

27.

136

0.8 × 2.0 m

Migdal Afek

Cist grave

28.

153

H. Te’ena

Burial cave?

29.

158

Kh. Umm al-Hammam

Arcolosium

30.

158

Kh. Umm al-Hammam

Arcolosium

31.

159

Wadi al-Bureid (south)

Double arcolosia

32.

169

Migdal Afek

Arcolosium

33.

200

H. Yeqavim

Arcolosium converted into cistern

34.

200

H. Yeqavim

Arcolosium

35.

202

H. Yeqavim

Arcolosium

36.

202

H. Yeqavim

Arcolosium

37.

202

H. Yeqavim

Arcolosium

38.

205

0.6 × 2.0 m

H. Yeqavim

39.

205

0.6 × 2.0 m

H. Yeqavim

0.40 × 1.10 m

0.6 × 1.7 m

190

Amit Shadman: Chapter 7, Settlement Patterns Table 7.3. Tombs in the study area. No.

No. in Rosh Ha‘ayin map

Size

Associated site

40.

205

0.8 × 1.8 m

H. Yeqavim

41.

205

0.5 × 1.3 m

H. Yeqavim

42.

205

H. Yeqavim

43.

205

H. Yeqavim

44.

205

H. Yeqavim

45.

205

H. Yeqavim

46.

205

H. Yeqavim

47.

205

H. Yeqavim

48.

224

Kh. Tukkim

49.

224

Kh. Tukkim

50.

224

Kh. Tukkim

51.

253

Kh. Ta‘amur

Notes

3 burial caves west of the site

Stepped burial cave? Within the main complex

52.

0.6 × 2.0 m; 1 m deep

Cist grave

53.

2 m long; 0.6 m deep

Cist grave with 2 diseased

54.

2 m long; 0.6 m deep

Cist grave

55.

Cist grave north of H. Yeqavim

Fig. 7.8. Cist tombs west of Kh. Kesfa (E. Marco).

191

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods worn potsherds dated to the Byzantine period were also discovered near the burials.

Kesfa, converted, at a certain stage, into a quarry, there was no evidence pointing to a halt in wine production. Summary

The Early Islamic Period (Fig. 7.9)

This chapter presents the settlement patterns in the study area (Table 7.4) through its different periods of occupation. The western half of the area could be examined at high resolution since it was widely excavated. The eastern half of the area was surveyed both in the past and within the framework of the present study. Settlement patterns were analyzed for each period discussed in the work, with a brief reference to the Persian and Early Islamic periods (Figs. 7.10/1, 7.10/2).

The number of sites significantly decreased during this period to 10, half the number of sites in the previous period. These sites were mostly built on (ATOP?) large- and medium-sized sites such as H. Yeqavim, Kh. Umm al-Hammam, Migdal Afeq, Kh. Tukkim, H. Te’ena, and Kh. Umm al-Bureid. Architectural changes were often noticed within these settlements, suggesting they were carried out during the Early Islamic period. Except for the large winepress in Kh.

Table 7.4. Site distribution by period.1 Site

Persian period

Hellenistic period

Roman period

Byzantine period

Early Islamic period

Kh. Taha

+

+

-

-

-

Site 87 (Kochavi and Beit Arieh 1994)

+

+

-

-

-

Site 138 (Kochavi and Beit Arieh 1994)

+

+

-

-

-

Site 126 (Kochavi and Beit Arieh 1994)

-

-

-

+

-

Site 168 (Kochavi and Beit Arieh 1994)

-

-

-

+

-

Site 1, Qasr as-Sitt

+

+

-

+

-

Site 2, Wadi al-Bureid

+

+

-

-

-

Site 3, H. Te’ena

-

-

-

+

+

Site 4

+

+

-

-

-

Site 5

+

+

-

+

-

Site 6

-

-

-

+

-

Site 7, Kh. Umm al- Bureid

+*

+*

-

+

+

Site 8. Kh. Dayyar

-

-

-

+

-

Site 9, H. Yeqavim

-

-

+

+

+

Site 10

-

-

-

+

+

Site 11, Kh. Umm al-Hammam

-

-

-

+

+

Site 12

+

-

+

+

+

Site 13, Kh. Kesfa

-

-

-

+

+

Site 14, Migdal Afek

+

+

+

+

+

Site 15, Qurnat al-Haramiya

+

+

+

+

+

Site 16

-

+

-

+

-

Site 17

+

+

-

-

-

Site 18, Kh. Tukkim

-

-

-

+

+

Site 19, Kh. Ta‘amur

-

-

+

+

-

Site 20

+

-

-

+**

-

Site 21

-

-

-

+

-

Site 22, Khallat as-Shirij

+

+

-

-

-

1 

The first five sites were not included in Chapter 3.

192

Amit Shadman: Chapter 7, Settlement Patterns

Fig. 7.9. Settlement distribution in the Early Islamic period.

193

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods

Site distribution by period

Persian Period (1th–5th century) Hellenistic Period (3th–4th century BCE) Roman Period (1th century BCE–4th century CE) Byzantine Period (5th–7th century CE) Early Islamic Period (7th–8th century CE)

Early Islamic Period (7th–8th century CE)

Byzantine Period Roman Period (5th–7th century CE) (1th century BCE– 4th century CE)

Hellenistic Period (3th–4th century BCE)

Persian Period (1th–5th century)

Fig. 7.10. Site distribution by period: above by percentage, below by number of site.

The Hellenistic Period

The Roman Period

In this period, the settlements were usually built on Persian-period sites built, in turn, on Iron Age II farms. The Hellenistic occupation was characterized by farms with built-up areas measuring 0.5–3.0 dunams. The excavations revealed that their general plan had not significantly changed since their establishment in the Persian period, and only a few minor modifications were applied. Thus, the farms apparently operated for about 500 years, from the 8th century BCE until the end of the 3rd century BCE, without major changes. Each farm had an agricultural plot measuring about 400 dunams, most of it used for vineyards and cereals, with only a few used for olive groves for local consumption. The absence of tombs may be explained by the Roman use of earlier burial complexes, as discovered near Migdal Afeq. Hellenistic farms did not continue into the Roman period, except for the large non-farm sites of Migdal Afeq and Qurnat al-Haramiya.

Unlike the Hellenistic and Byzantine periods (see below), the rural open areas were not occupied in the Roman period. No Roman farmhouse was discovered in these areas, neither were larger settlements. This established the large settlement of Migdal Afeq in the western part of the area as the main site of this period. Its status is evidenced by its strategic location, the adjacent burial complexes, and the historical references by Josephus. Furthermore, based on the potsherds collected in various surveys, both Migdal Afeq and the agricultural areas to its east apparently functioned continuously throughout the Roman period (1st century BCE–4th century CE). The lack of rural sites (farms, estates, villages) during this period does not necessarily indicate a decline in population. It is possible that most of the inhabitants preferred to live in a large settlement and 194

Amit Shadman: Chapter 7, Settlement Patterns daily commute to the agricultural areas, as noticed in other parts of the Samaria Foothills (‘Ad 2000). Presumably, following the Great Revolt and the Bar Kokhba Revolt, the authorities prohibited settling in open areas, more difficult to supervise. The role of the fortress at Site No. 12 as an administrative center may be understood in this context. At least in the 1st century CE, the region was seemingly inhabited by a Jewish population, as evidenced by the archaeological finds and Josephus’ account.

around longitude 200 and were at a distance of about 1.0–1.5 km from each other. These large settlements had nearby satellite units, including farmhouses, estates, churches, and monasteries, sometimes interconnected by agricultural roads. The monastery at H. Te’ena is an important site in the region and was part of a chain of monasteries and churches present in the Samaria Foothills. It is not clear to which site this monastery was related, as it was located on the road between Migdal Afeq on the west and Deir Balut on the east. Road No. 3, which connected H. Te’ena to Deir Balut, suggests that the monastery maintained relations with ‘the eastern strip of settlements.’ However, the Christian population clearly dominated the area and maintained agricultural systems, as evidenced by rural roads, winepresses, and oil presses whose products (especially the oil) were marketed to other regions.

The Byzantine Period The residents of the Byzantine settlements were apparently responsible for creating today’s landscape. Four ‘rural towns’ were founded or expanded during this period: Kh. Kesfa, Kh. Umm al-Hammam, H. Yeqavim, and Kh. Ta‘amur, all continuing into the 7th century CE. The Byzantine settlements were located

195

CHAPTER 8

COMPARATIVE STUDY WITH THE AREA SOUTH OF NAḤAL SHILOH

Introduction

20 from the Byzantine period, and 10 from the Early Islamic period.1

The area south of Naḥal Shiloh shows archaeological and landscape characteristics similar to the study area, adjacent to it on the north. This area to the south, like the study area, underwent modern development during the past decades, resulting in archaeological surveys and salvage excavations that allowed its exploration at high resolution.

The area was first occupied in Iron Age II when seven farmhouses were founded. These farms were continuously active for about 500 years and were abandoned at the beginning of the Hellenistic period. The Roman period witnessed a dramatic occupational decline, with no real settlements except for the sites of Migdal Afeq, and Qurnat al-Haramiya to its west. A significant change apparently occurred in the 5th–6th centuries CE, when ‘rural towns’ were first established, based mainly on agriculture. Most Byzantine settlements are characterized by dense building with no pre-planned design and surrounded by agricultural systems. Several monasteries and churches were also discovered in the area, with their inhabitants engaging in agriculture. All these settlements were interconnected by a network of ten rural roads extending for a total length of approximately 30 km. Burial grounds were discerned solely in conjunction with the Byzantine sites.

The boundaries of this adjacent area were mainly determined by its topography, with its southern boundary at Naḥal Mazor, Route 444 passing at its foothill on the west, and H. Anusha and the ‘Green Line’ on the east. This area covers about 15 sq km, about half the size of the main study area. This comparative study is essentially an analysis of several aspects already examined in the main study area, in another part of the Samaria Foothills. It aims to draw a macro picture by examining the central settlements in the region and their history. This area has been investigated by various scholars whose studies have not been fully published. However, the excavations and surveys conducted at H. Zikchrin, H. Mazor, H. Anusha, H. Leved, and their surroundings, provide fairly good data enabling proper discussion. This data is chronologically presented from the Hellenistic to the Byzantine periods.

The Area Between Naḥal Shiloh and Naḥal Mazor East and west of the line separating between H. Zikhrin and H. Mazor are convenient alluvial areas suitable for agriculture, as evidenced by the thoroughly researched agricultural systems around H. Anusha and H. Leved (Safrai 1997a; 1997b; Sion, ‘Ad and Haiman 2007: 109– 159). The international road (‘Via Maris’) passed west of the H. Zikhrin–H. Mazor line, and it is not clear how these two sites were associated with the road. Presumably, various authorities throughout history had limited, to a certain extent, access to the road and its adjacent fields. In recent years, the area underwent extensive changes with the development of the town of ‘Elad and other major construction. The Gophna–

The Main Research Area—between Naḥal Rabah and Naḥal Shiloh Following are some key data elements from the main study area that are essential for the comparative discussion. The area extends over 30 sq km between Naḥal Rabah in the north and Naḥal Shiloh in the south, with its western boundary overlooking the Sharon Plain, and the ‘Green Line’ on its east. Twentyseven sites from the Persian to the Early Islamic periods were surveyed here: Three from the Persian period, 12 from the Hellenistic period, 5 from the Roman period,

1 

Several sites are counted more than once since they continued in the later periods.

197

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods Antipatris road passed to the north of this area, along the southern bank of Naḥal Shiloh (see Chapter 6).

operated here in conjunction with the Roman road mentioned above. The site reached its peak in the Byzantine period, when residential quarters, a church, a monastery, and bathhouses were built. The site flourished in the Early Islamic period, and subsisted, in fact, until the late Middle Ages.

The area was surveyed in the 19th century by Guérin (1984) and the PEF (Conder and Kitchener 1882), who were primarily interested in the ruined settlements. Finkelstein (1978a: 14–16, 1981) surveyed the area within the framework of the farmhouses study east of H. Mazor. He documented five farms dated to Iron Age II and the Persian period, and several dated even to the Hellenistic period. Finkelstein notes that, in several cases, the Gophna–Antipatris road ‘cuts’ some farms, concluding it was of a later date. Another survey was carried out called the ‘Map of Rosh Ha‘ayin’ (Kochavi and Beit Arieh 1994). Ten farmhouses were discovered (some of them also mentioned by Finkelstein), together with four other settlements: H. Zikhrin, H. Mazor, H. Leved, and H. Anusha. A large-scale study was carried out by ‘Ad (2000), who investigated settlement distribution in the Samaria Foothills in the Byzantine period. His study included an archaeological and historical overview of earlier periods, which provided the present study with useful data for spatial analysis.

Another ‘rural town,’ measuring about ten dunams, was in H. Anusha, located in the eastern part of the area. Surveys and salvage excavations (Sion, ‘Ad and Haiman 2007) indicated that the site was settled, in fact, in the Byzantine period, although a few potsherds from the Late Roman period were also found. The site was reoccupied probably during the Mamluk period. It seems, therefore, that H. Anusha was originally a Byzantine ‘rural town’, one of the settlements that extended along longitude 199–200. Unlike the neighboring site of H. Leved, the site was surprisingly not occupied in the Early Islamic period. In H. Leved, located 1.5 km northwest of H. Anusha, a Byzantine settlement (monastery?) was founded, occupied by a Christian population as evidenced by a cross carved in a lintel in the eastern part of the site (Sion, ‘Ad and Haiman 2007: 121). Two oil presses, a threshing floor (outside the settlement) and eight cisterns were also detailed (Safrai 1997a: 223). The site was occupied through the Early Islamic period—a common phenomenon in the region. The presence of rock-cut steps led several scholars to suggest that there was a miqveh (ritual bath) dated to the Second Temple period, but this is not evident.

Prior to constructing the town of ‘Elad, an extensive archaeological survey was conducted in an about 4 sq km area (Amit and Zilberbod 1996a: 65). Nine farmhouses, 26 field towers, six winepresses, eight burial caves, and 14 cisterns were documented during the survey. Several archaeological excavations were conducted in H. Mazor and its neighboring farmhouses following the survey (Amit 1996a: 57, 1996b: 58–61; Gudovitz 1996: 57–58; Amit and Zilberbod 1996b: 61–63; 1999: 55-56; Zilberbod and Amit 2001: 45–46; Zelinger and Amit 2001: 79–83; Taxel 2006). The results showed that the farmhouses, remote from each other, were apparently established in the Hellenistic period.

Two fortresses found in conjunction with the Roman road should also be noted, one at the mouth of Naḥal Shiloh, where the Gophna–Antipatris road joined the ‘Via Maris’ (Bahat and Bucheri 1963: 19–20; Kochavi and Beit Arieh 1994: Site No. 167), and the second controlling the road that descends into Naḥal Shiloh (Kochavi and Beit Arieh 1994: Site No. 248).

Most farmhouses beyond H. Mazor and its vicinity were excavated by R. Toueg, who reported (personal communication) that most were founded during the Persian period (6th century BCE) and lasted until the beginning of the Hellenistic period (4th century BCE). He also remarked that one farmhouse from the beginning of the Roman period was identified, a rare finding in the region.

In summary, this is a rural agricultural area typical of the Samaria Foothills. It is characterized by dozens of field towers, enclosed plots, agricultural roads, water reservoirs, quarries, winepresses, and cisterns. Two mausolea were also discovered in the region. One of them is southwest of H. Zikhrin, near the Gophna–Antipatris road, and the second, southwest of H. Mazor, is one of the most impressive monuments in Israel (Kaplan 1983; 1985). Burial caves and dozens of arcosolia tombs also were found, most of them in conjunction with the large settlements, such as the burial cave excavated southeast of H. Zikhrin, used in the Roman and Byzantine periods (Haddad 2007: 45– 57). Presumably, this cave was part of the cemeteries of H. Zikhrin (Fisher 1985: 119–120).

H. Zikhrin was thoroughly excavated by Fisher (1983, 1984, 1985, 1989, 1993) and later by Taxel (2005, 2010 2013). The excavations revealed a rural farmhouse dated to the 1st century BCE (Fischer 2008: 2083) probably occupied by a Jewish population (Taxel 2013: 150). It continued until the 1st century CE and was deserted as a result of the Great Revolt (Taxel 2005: 15). During the 2nd–4th centuries CE, a caravansary

198

Amit Shadman: Chapter 8, Comparative Study with the Area South of Naḥal Shiloh Settlement Patterns

seem an extensive settlement existed here during this period.

The Hellenistic Period. Settlement pattern during this period comprised farmhouses, some already established in Iron Age II, and a rural settlement (H. Mazor) which was actually a cluster of farms/estates whose area extent is not clear and may be defined as a ‘scattered village’. Excavations revealed several structures on the slopes of a vast hilly area typical of the region (Amit and Zilberbod 2001: 45–46). Unlike farms located in open areas (those found in Finkelstein’s studies), founded in Iron Age II, some farms/estates in H. Mazor’s area were established only during the Hellenistic period (e.g., Zelinger and Amit 2001: 79–83). It should also be noted that the excavated building 5H, defined by the excavators as a ‘mansion’, was dated to a later phase in the Hellenistic period (Amit and Zilberbod 1996b). Seleucid coins and stamped amphora handles were found in this building. Another Hellenistic farmhouse was discovered in Area S2-3, comprising a series of rooms surrounding a central courtyard (Amit and Zilberbod 1996b: 61–63). Among the finds from this building were Hasmonean coins (end of the 2nd century BCE), and the excavators attributed them to the period of John Hyrcanus.

Clear Christian symbols were discerned at H. Leved where a monastery or a medium-sized Christian settlement was located. It is worth noticing, however, that H. Leved was first settled during Iron Age II, possibly as a farmhouse (Finkelstein 1978a: 16, Kochavi and Beit Arieh 1994: Site No. 272). During the Byzantine period, the village master of H. Anusha may have lived here and owned the neighboring fields. The excavations at H. Mazor did not provide a clear understanding of its nature in the Byzantine period, and it should perhaps be evaluated by the agricultural installations in its vicinity, such as the elaborate winepress dated to this period (‘Ad 2008). This winepress, as well as the many other revealed in the area, attest to an extensive agricultural system in the area operating, inter alia, wineries. Operating such a winepress required a large and available workforce, which could be recruited in the vicinity of H. Mazor, perhaps explaining why elaborate installations such as winepresses and oil presses are usually absent in the open areas. Therefore, it would seem that the Byzantine village at H. Mazor controlled the agricultural areas located mainly east of the site and on the moderate slopes to its north and west.

Several farmhouses were excavated north and east of the H. Mazor, showing a slightly different picture. These farms operated mainly during the Persian period, some continuing into the Hellenistic period. Some of these farms were seemingly identified in Finkelstein’s study.

Discussion and Summary A comparison between the study area and the area south of Naḥal Shiloh revealed similarities as well as differences. The area north of Naḥal Shiloh is topographically characterized by alluvial lands on the west and rocky mountainous lands on the east. The area south of Naḥal Shiloh, on the other hand, is mainly composed of a vast plateau extending from the H. Zikhrin–H. Mazor line eastward to H. Anusha. The farmhouses, especially those recorded by Finkelstein (1978a, 1981), are generally located between longitudes 199–196, where lands are more suitable for agriculture.

The Roman Period. The Early and Late Roman periods are mainly represented by large sites, such as H. Zikhrin and H. Mazor. The first comprised a small settlement, maybe a farmhouse or a caravansary (Taxel 2005: 15). The second included several farmhouses or estates. The building (20 × 27 m) excavated in Area N (Amit 1996b: 58–61), composed of rooms surrounding a central courtyard, yielded 55 coins dated to the 4th century CE. An oil press that operated by the ‘beam and a screw’ method was excavated south of this building.

It should be noted that the settlement density was lesser in the eastern part of the area south of Naḥal Shiloh, despite the suitable lands. The reason for this phenomenon is not necessarily related to agriculture, but seemingly to the farmhouses both north and south of Naḥal Shiloh being associated with the large communities in the western part of the area, and not with those on the east. The farmhouses excavated north and south of Naḥal Shiloh were dated mainly to Iron Age II and the Persian period, with only a few dated to the Hellenistic period. However, it is important to mention again the farms excavated at H. Mazor, founded in the Hellenistic period and some continuing

The Byzantine Period. A significant increase in settlement distribution occurred during this period, as evidenced by the archaeological finds from excavations and surveys. H. Zikhrin was the most prominent site of this period, both from the architectural aspect and the size of its built-up area (c. 30 dunams). The excavation revealed a large rural village that played a major social, religious, and functional role. Another ‘rural town’ probably existed at H. Anusha, and although only a limited space was excavated, based on the size of its built-up area (c. 10 dunams) and the ceramic finds (Kochavi and Beit Arieh 1994: Site No. 280), it would

199

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods

Fig. 8.1. Schumacher’s map (1918).

up to the Hasmonean revolt. In any event, these farmhouses do not belong to the ‘classical’ farmhouses studied by Finkelstein (1978a, 1981), but were possibly part of a larger settlement (perhaps a village) founded at H. Mazor during the Hellenistic period, as opposed to the area north and south of Naḥal Shiloh.

‘margin dressing’ style were found (Amit and Zilberbod 1996b: 61–63; Area 1S). Hellenistic coins further contribute to the discussion of that period: Seleucid coins were found in the farmhouse at H. Mazor (Amit and Zilberbod 1996b), while Ptolemaic coins were found at Qurnat al-Haramiya and in the farmhouses in Sites Nos. 2 (Wadi al-Buried), 4 (Naḥal Shiloh) and 5 (Naḥal Shiloh [west]) north of Naḥal Shiloh. It is possible that these coins reflect the ‘Syrian Wars’ (see below).

Generally, the farmhouses built in the two areas were modest, with no wealth or luxurious elements, except in two cases. In Site No. 2 (Wadi alBuried), north of Naḥal Shiloh, a bath was found in Room VII of the ‘Administrative Building’. In H. Mazor, south of Naḥal Shiloh, remains of plaster molded in the

In the Roman period, a similar picture emerges between the two regions. Settlement during this period was evident mainly in large sites in the ‘western strip’, 200

Amit Shadman: Chapter 8, Comparative Study with the Area South of Naḥal Shiloh such as Qurnat al-Haramiya, Migdal Afeq, H. Zikhrin, and H. Mazor. Settlements are almost absent in open areas, and the inhabitants seemingly preferred to live next to the main roads (‘Ad 2000). The sites mentioned above are located near the ‘Via Maris’ and H. Zikhrin, along the road ascending to Gophna. A Roman fortress (Site No. 12 Naḥal Shiloh [southwest]), controlling this road and several rural roads, was discovered north of Naḥal Shiloh. This hilly strategic site was used for security purposes already before the Roman period.

4.

A rural road network, similar to the one north of Naḥal Shiloh, also existed to its south. Schumacher’s map (1918; Fig. 8.1) shows several roads linking H. Zikhrin, H. Mazor and Qula2 with their agricultural area to the east. According to this map, and assuming that these routes have not changed much through history, it would seem there was a clear link between the Antipatris–Gophna road and the rural roads in the region. Thus, a settlement pattern emerged in the Roman period of a few, preferred large centers, as opposed to the previous Hellenistic pattern of scattered farmhouses. The Roman farmers commuted to their fields daily, made possible by the considerable improvement in the road system. During the Byzantine period, an interesting picture emerged, evident in both compared areas. Analysis of the settlements shows that they were located in three strips:3 1.

2. 3.

The ‘western strip’ includes Qurnat al-Haramiya, Migdal Afeq, H. Zikhrin, and H. Mazor. These settlements, relatively large (but not towns), were first occupied in Iron Age I, and sometimes even earlier. In the Byzantine period, these settlements comprised residential and public buildings which spread over dozens of dunams, H. Zikhrin being a good example though it may not characterize all the settlements in this strip. Fig. 8.1. Comparison between the study areas (after Schumacher 1918). The ‘monasteries strip’ is located between longitudes of 199–198, and includes Kh. Dayyar, Kh. Umm el-Bureid, H. Te’ena, and H. Leved. H. Tukkim is also located in this strip, but bears no clear Christian monastic features. Each monastery is adjacent to a large settlement: Kh. Dayyar and Kh. Umm al-Bureid are 1.5–1.2 km northwest of Kh. Kesfa; H. Te’ena is one

2 

This site is not included in the comparative discussion. However, its size, location on a central crossroads and long historical sequence, points to its importance in the settlement system. 3  As stated in the Introduction, the study area was divided into four ‘settlements strips’. However, in the context of this discussion, only three strips are mentioned, the second one being, actually, this area.

201

kilometer west of Kh. Umm el-Hamam; H. Leved is one kilometer west of H. Anusha. The monasteries north of Naḥal Shiloh were founded during the Byzantine period and continued to exist until the beginning of the Early Islamic Period. Only H. Leved, south of Naḥal Shiloh, was probably founded on earlier remains, reaching its peak in the Byzantine period. The ‘rural-towns strip’ includes H. Kesfa, Kh. Umm al-Hammam, H. Yeqavim, Kh. Ta‘amur, H. Anusha, and also H. Al (located outside of the compared area). This group of settlements was probably built during the Byzantine period (5th– 7th centuries CE) and continued into the Early Islamic period. It is difficult to define whether these sites were occupied before the Byzantine period without excavating them, though most findings from surveys were from this period. Located south of Naḥal Shiloh, H. Anusha differs from the four settlements to its north, since the site was occupied in several other periods. The Gophna–Antipatris road that passed nearby likely affected the site’s settlement history. All the settlements in this strip were built between longitudes of 199–200 (see Chapter 10).

CHAPTER 9 THE ESTIMATED POPULATION IN THE REGION

Introduction

that more than five persons lived in one room and about 20–30 in an average countryside house in the Samaria Hills. Geva (2014: 132) distinguished between population density in urban and rural areas, the latter comprising most of the area of the present study, reaching the obvious conclusion that urban environments are likely to be denser. Similar studies have been carried out on traditional Arab society (Granowsky 1949; Finkelstein 1990; Grossman 2004). Thus, it would seem that, for reliably estimating the population in antiquity, additional factors must be taken into consideration, some related to the open areas in the vicinity of the settlements. The extent of these areas, especially the agricultural plots, correlate with the number of residents who cultivated them.

This chapter aims to carry out an initial study of demographic changes from the Hellenistic through the Byzantine periods in the study area. However, it should be pointed out that arriving at an accurate estimation is not possible, and that the figures quoted for any specific period are only conjectural. Being as this study, like others of ancient populations, relies on data that may not be accurate, the discussion is supplemented by studies of the traditional Arab village, especially involving their water consumption. Since archaeological data from ‘the farmhouses strip’ is more detailed, due to the various excavations conducted there, the discussion will focus mainly on the population of the Hellenistic farmhouses and the main Byzantine settlements, only briefly touching the subject of Roman settlements.

Many studies have been carried out on the ratio of agricultural plots per capita (Dar 1986c; Broshi 1986; Safrai and Fixler 1993). Granowsky’s study (1949: 159) on the traditional Arab village showed that substantial differences are sometimes present among villages. For example, a village with lands extending over 50,000 dunams and with 4,000 inhabitants compared to a village with 50,000 dunams or more but with only 200 residents. Presumably, such differences were also present in ancient societies. During the Roman period, and especially in the Byzantine period, enclosed plots were built and cultivated in the settlements’ hinterland, most times quite accurately measured. Other nonenclosed and unmeasured plots were cultivated too, evidenced in our region. Therefore, to calculate the agricultural plot accurately, both the enclosed plots and the rest of the area (such as the terraced slopes, wadis, and dams) should be taken into account.

Calculating ancient populations is an additional layer added to modern archaeological research. Many studies have been conducted to trace the number of people in a particular area (especially in urban archaeology) and determine its settlement density over the periods. The generally accepted formula for calculating population is by multiplying the settlement’s built-up area by its density factor (Geva 2014: 131). The built-up area includes the space within the boundaries of a wall (if any) and buildings in adjacent areas (even outside the wall). Public buildings, estimated to make up 25% of the entire built-up area, were not used as daily domestic structures and will be excluded. In rural villages or farmhouses, the main building complex (sometimes a single building) must be taken into account.

Studies on grain agriculture showed that each dunam provides about 75 kg of product (Geva 2007). Thus, to supply the annual consumption of a single individual who needs about 300 kg of wheat (Broshi 2001), an area measuring approximately four dunams was required. The suitability of the given area for growing grain must thus be taken into account. Naturally, this factor actually determined the crops raised by the farmer. Therefore, communities with lands not suitable for a particular crop probably traded

Unsurprisingly, scholars have reached different conclusions regarding the population density coefficient per built-up dunam, determined at 40–50 persons per dunam (Shiloh 1980: 30), 30 persons per dunam (Safrai 1997c: 278), and 25 persons per dunam for 8th-century BCE Jerusalem (Reich and Shukron 2003: 213). Dar (1982a: 147) wrote on the number of individuals per room within a building, arguing 203

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods with their neighbors. Also, produce from other crops, such as oil and wine, which were essential nutrients, were probably exchanged. In a study conducted at Ramat Beit Shemesh (Dagan 2010, 2011b), the carrying capacity of the area and its main crops during the Roman and Byzantine periods was examined in relation to the area’s population count (Bonfil and Hadas 2011).

were identified in the area through history, the population was estimated for each period separately. The Population Count in Hellenistic Period Farms Built-up Areas

Another way to estimate the size of the population is to examine the correlation between the available water sources and the number of the inhabitants. Past studies examined this issue in Shivta (Sobota) in the Negev (Tsuk 2003: 18–24), and H. Zikhrin (Tsuk 1994: 133–148), adjacent to our region. Tsuk (2011: 42–46) indicates that the average volume of an Iron Age II water cistern was about 25 cubic meters, enough to supply water to a family of five. Other cisterns of this period had a larger volume, like the 56-cubic-meters cistern in Tel Jezreel. The average volume of a water cistern during the Roman and Byzantine periods throughout Eretz Israel was about 40–50 cubic meters, similar to that in our region. It is generally accepted that the annual consumption of water during the Iron Age was about five cubic meters per capita and about seven cubic meters per capita in the Byzantine period (Tsuk 2011: 32). These amounts do not include the water consumed by the livestock, which varies according to the species (such as donkeys and camels) and their numbers (Evenari, Shanan and Tadmor 1971: 148–150). The present study addressed the data gathered in the traditional Arab villages, where the daily water consumption was 10 liters, having an annual consumption of 3.6 cubic meters (Brawer 1977: 382, note 15). Three main factors can help in reliably assessing the size of the population in the region: 1.

2.

3.

The Hellenistic period is manifest in farmhouses covering, each, 1–3 dunams. Also, larger settlements existed at Qurnat al-Haramiya (c. 10 dunams) and Migdal Afeq Tower (35 dunams), and in two other settlements of uncertain size. Accurate data on the built-up areas in our study area are available only for sites in ‘the farmhouses strip.’ Because of the rugged nature of the region, it is easy to discern the remains of walls of unexcavated buildings, helping to understand their plan and nature (Finkelstein 1997: 20–21). Almost half of the farmhouses’ built-up areas were occupied by yards or storage facilities (e.g., Sites No. 1, 2, 4, 5), meaning that no more than half of the built-up area was used for domestic purposes (seven dunams; see also Dar 1996: 151–157). Accordingly, and following a minimalist approach (15–25 individuals per built-up dunam), it would seem that less than 200 people lived in the farms that operated in the Hellenistic period. Taking into consideration several destroyed sites and the large settlements at Qurnat alHaramiya and Migdal Afeq, the population in the area during this period totaled a few hundred people. Water Cisterns A typical pattern of a single cistern located next to each Hellenistic farmhouse was discerned (Sites No. 2, 4–5). A similar situation was revealed in the Western Samaria Hills (Dar 1982a : 98). Most cisterns adjacent to the farmhouses were bell-shaped, with shafts of varying lengths in their upper part. It would seem the cisterns were used from the Iron Age II to the Hellenistic period when the farmhouses operated.

The settlement’s built-up area. The typical settlement in our region was not necessarily an enclosed community or a walled area, but rather composed of buildings scattered over large areas (farmhouses or ‘rural towns’). Therefore, it is necessary to distinguish between the density factor in farmhouses and that in the ‘rural towns’. Farmhouses being isolated scattered structures, the number of their residents did not exceed 15–25 individuals per built-up dunam. The larger ‘rural towns’ were characterized by rather densely-planned buildings with many rooms (e.g., Kh. Ta‘amur, Site No. 19) occupied by 25–35 residents per built-up dunam. The number of water cisterns and reservoirs in or near the settlement. The discussion of water storage capacity and its relation to the population count is based only on the cisterns within the settlements or in their immediate area. The extent of the agricultural plots in relation to the settlement. Since significant changes

Given the amount of water consumption mentioned above, it would seem that a farmhouse with 25 residents would have required more than a single water cistern, i.e., a single Hellenistic farmhouse needed two cisterns totaling 90 cubic meters. Thus, either a single farm had to be occupied by less than 25 residents, or other cisterns nearby could not be associated with a particular site or period. Another undetermined factor is how many times a single cistern was filled during winter, since multiple filling cycles in a rainy winter may jeopardize our conclusion. Based on the available data, the amount of water in a farmhouse extending over a dunam could sustain 15 residents. Accordingly, a single cistern could supply the required amount of 204

Amit Shadman: Chapter 9, The Estimated Population in the Region drinking water, since, otherwise, the question arises as to why no other water cistern was hewn. As for the other water consumers (mainly livestock), these were fed by cisterns in the pasture areas. For example, adjacent to cistern No. 106, located within the slot of a farmhouse1 southeast of H. Te’ena, was a rock-cut trough for watering animals.

the surveyed farmhouses totaled about 150–200 residents (Table 9.1). This figure can serve as a basis for calculating the population of the entire region during the Hellenistic period, or even a representative sample. The Population Count in the Roman Period Estimating the population count in the Roman period is not an easy task. Very few sites of this period were discovered in the study area, and they were frequently identified by only a handful of potsherds (except Migdal Afeq, see below) which do not necessarily point to an actual settlement at that location. Nevertheless, based on the few data brought below, a tentative assessment of the population count of the area in this period will be attempted.

The Agricultural Plots The farms dominated large areas in their immediate vicinity, especially in the alluvial areas east of Migdal Afeq. Each farm owned an agricultural plot of c. 400 dunams which bordered the adjacent farms. Based on the available data, each farm was occupied by 15–25 residents who needed about 100 agricultural dunams for domestic consumption (25 people per four dunams). This figure relates only to grain crops, the rest of the area being designated for vineyards, pasture, and other. A single farm apparently possessed a much larger area than required for its daily needs—16 dunams per resident. However, the 1931 survey conducted in the Arab villages showed differences between the various regions. For example, peasants in the Coastal Plain owned about 13 dunams, in the mountains about 20 dunams, and in Galilee 17 dunams (Granowsky 1949: 180). Therefore, it would seem that a slot of 16 dunams per person in the farms in our region is a reasonable number. It is also possible that significant portions of the agricultural yield of the area went to governmental taxing (Broshi 2001: 86–92).

Based on the archaeological finds, an extensive settlement seems to have existed in the Roman and Byzantine periods at Migdal Afeq, in the western part of the region (Site No. 14). It had a built-up area measuring c. 35 dunams, but only small parts of it were excavated. The large water reservoirs and the many cisterns found in the agricultural plot of the site further support the vast extent of the settlement. Agricultural areas apparently cultivated by the residents were located east and west of the site. Potsherds dating to the Roman period were found in these open areas, but no real settlements were discerned. It would seem that most residents lived in Migdal Afeq and only a very few lived in Qurnat alHaramiya to its northeast. Even the built-up area of Migdal Afeq was underestimated, being, in fact, the

Taking into account the built-up area, water cisterns, and agricultural plots, the population of

Table 9.1. Calculation of the farms’ population in the Hellenistic period. Site No.

Built-up area (dunams)

Cisterns

Agricultural plot (dunams)

Estimated population

1.5

2

400

15–25

2.

2

1

400

15–25

4.

1.5

-

400

15–25

5

2

1

400

15–25

16.

0.5

-

400

7

17.

1

-

400

13

20

1.5

1

400

15–25

22.

3

1 + underground reservoir

400

40

65/3 (Finkelstein 1978a)

1

-

400

13

Total 9

Total 14 dunams

Total 6 cisterns

Total slots: 4 sq. km.

150–200

1 

This farmhouse was poorly preserved and it is only briefly reported, see Chapter 3, site No. 20.

205

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods some general Byzantine-period standard pattern where two cisterns per one built-up dunam were planned.

largest and most important settlement in the region. Assuming that half of its area was used for residential quarters, the settlement was occupied by about 500 inhabitants. Josephus Flavius mentioned a few Jews who lived here (Wars II, 19, 1), but it is impossible to rely on his testimony for estimating the population at the site. He also points out that other villages existed in the vicinity of Migdal Afeq.

The settlements in our area were smaller than H. Zikhrin (except for Migdal Afeq), the number of water cisterns diminishing accordingly. In H. Yeqavim, for example, whose area equaled c. 15 dunams, 14 water cisterns were discovered. The total volume of the known cisterns was 630 cubic meters. An underground water reservoir of unknown volume was also recorded. Assuming half of the area was used for domestic purposes, it would seem that, during the Byzantine period, H. Yeqavim had been occupied by 262 inhabitants. Based on an annual consumption volume of 3.6 cubic meters of water per person, it emerges that the water in the cisterns did not fulfill the needs of the population for drinking water. Thus, additional cisterns remained probably unrevealed in the various surveys. In Kh. Kesfa, of a similar size, vast parts of its built-up area were occupied by several oil presses that operated there. Only seven or eight water cisterns were discerned within the boundaries of this settlement, obviously not sufficient for such a site, although it is clear there were other cisterns which have not yet been found. Still, three additional reservoirs (totaling a volume of c. 400 cubic meters) were discerned south and west of the settlement. Thus, the volume of these reservoirs and water cisterns totaled c. 700 cubic meters, similar to that in H. Yeqavim. It is worth noticing that in one of the rooms of the building (perhaps a courtyard) in Umm el-Bureid (Site No. 7, probably a monastery), a large pond (288 cubic meters) and a water cistern were found. The built-up area of the site measured about two dunams and was occupied by 20–30 people. Thus, the amount of water stored at the site was definitely much higher than it needed. A similar ratio of water quantity/population count was also observed in other Byzantine sites in our region.

In conclusion, the absence of practically all settlement types in the area during the Roman period apparently points to a decline in population. However, it would seem that this population preferred to live in the large settlements in ‘the eastern strip’ and ‘the western strip.’ Thus, most of the 500 people living in the study area in this period lived in Migdal Afeq or its immediate vicinity. It should be remembered that major historical events taking place in the Land of Israel during the Roman period (the Great Revolt and the Bar Kokhba Revolt) apparently left their impact also on study area’s population. The Population in ‘The Rural-towns Strip’ and The Monasteries at H. Te’ena, Kh. Umm Al-Bureid and Kh. Dayyar during the Byzantine Period Built-up Areas The built-up areas of the Byzantine settlements are estimated at c. 70 dunams (excluding Migdal Afeq and Qurnat al-Haramiya). However, only half of this area—about 35 dunams—was used for residential purposes, since considerable parts were occupied by many oil presses, winepresses, and public courtyards. Accordingly, the population in these settlements, during the Byzantine period (excluding Migdal Afeq), totaled c. 1100 people.2 Water Cisterns and Reservoirs

The Agricultural Plot

Population growth during the Byzantine-period required an appropriate water supply. Besides the numerous cisterns discovered in the settlements, large reservoirs were identified. In the large settlements, cisterns were discerned near almost every building or complex of rooms. H. Zikhrin, for example, mostly excavated (Fischer 1985), extended over c. 30 dunams and reached its peak in the Byzantine period. More than 50 cisterns, most dated to this period, were revealed in a survey conducted at the site (Tsuk 1994). Kh. al-Burak, in the Western Samaria Hills and also thoroughly studied (Dar 1982a: 82–131), covered an area of c. 30 dunams where 50–60 cisterns were traced with a volume totaling c. 3000 cubic meters. One may assume that both H. Zikhrin and Kh. al-Burak represent

In the Byzantine period, unlike other periods in the region, the area was maximally utilized for agriculture. Evidently, agricultural systems were systematically built on the hillsides, the wadis, and the alluvial valleys. The ratio between the agricultural plot and the size of the population during this period is presented below for each settlement. Based on the residential quarters of both Kh. Kesfa and Umm al-Bureid (Site No. 7), these sites were occupied by 290–300 inhabitants. Since the agricultural plot measured 2562 dunams, each resident received about 8.5 dunams. Assuming that each resident needed only about 4 dunams for growing wheat (see Introduction, above), the entire community produced twice its needs. The agricultural plot of Umm al-Hammam measured c. 2062 dunams, and the residents numbered

2 

The population density coefficient in the large settlements was higher: 35 individuals per built-up dunam.

206

Amit Shadman: Chapter 9, The Estimated Population in the Region about 240. These figures include an estate (Site No. 10), and a farm (Site No. 6). Thus, every resident owned c. 8.5 dunams. The settlement at H. Yeqavim had an agricultural plot of about 2,863 dunams, and a population totaling 275–300 inhabitants. Therefore, each resident owned an area of approximately 9.5 dunams. Both Kh. Ta‘amur and Kh. Tukkim to its west were the southernmost ‘rural towns’ in the area. The agricultural plots of these sites extended for about 2568 dunams, and the sites’ population totaled about 200–220 inhabitants, meaning that each owned about 11 dunams. Two sites defined as monasteries should be discussed in this regard. One is H. Te’ena, with an agricultural plot of approximately 1890 dunams and 60 inhabitants/monks (perhaps even less) owning, each, an area of about 30 dunams. Apparently, such an agricultural plot was far too large for a settlement that extended over an area not exceeding five dunams (see below). The second monastery was at Kh. Dayyar, whose agricultural plot extended over c. 2287 dunams. Based on its residential quarters, no more than 40 people lived there, i.e., approximately 57 dunams were allocated to each resident. It is clear that, also, in this case, no apparent correlation existed between the residential area and the agricultural plot.

correlation between the size of their population and the extent of the slots raises two main questions: Who cultivated the agricultural lands? And for whom were the crops destined? Since it is unlikely that a handful of people would build and maintain such an agricultural system, the answer may lie in hired laborers (Dan 2006). Dan argued that hiring workers was common both in cities and villages, and sometimes two nearby places shared such an arrangement. For example, the Talmud describes that people from the village of Beit Ma‘on, near Tiberias, used to work in the city, and vice versa. A 3rd-century story tells of a group of hired laborers (Fritin) in the village of Hittin (Bereshit Rabbah 62:16), who used to dine at the synagogue or in the nearby hostel. Thus, these hired laborers lived in the village and carried out temporary works. A similar reality, of workers hired by villages to carry out various tasks such as constructing public buildings, is known in Syria and Asia Minor (Dan 2006: 59). Thus, the monasteries may have been assisted by foreign workers (perhaps even monks who came for a limited time) in the day-to-day care of the fields and the construction of their dormitories. The crops from their fields were undoubtedly traded to the various nearby settlements and even outside the region.

The data presented above may shed light on the discussion on the region’s population count. In four ‘rural towns’ (except for Kh. Ta‘amur), with similarly sized agricultural plots, each resident owned 8.5–9.0 dunams. Since most open spaces in the region were utilized for agriculture, and most Byzantine settlements had adjacent enclosed plots built in a specific order (see Chapter 3), the plots may have been divided between the residents. This assumption may contribute to estimating the population count, but caution is called for since there were sites with no enclosed plots but with cultivated areas around them (e.g., Umm alHammam).

In summary, it can be cautiously said that about 1,100 people lived in the study area (excluding Migdal Afeq) in the Byzantine period, mainly in ‘rural towns’ and the monasteries (Table 9.2). Summary Studying the area through a high-resolution survey provided a solid basis for this discussion. Dividing the region into settlement strips facilitated, to a certain extent, the examination of the issue, and the excavations conducted in several farms also assisted in assessing their population. The population in the farms at the beginning of the Hellenistic period was about 200 people, with an additional yet somewhat vague number of people living in the large settlements to the

The situation in the monasteries at H. Te’ena and Kh. Dayyar was totally different, and the lack of

Table 9.2. Population count in ‘rural towns’ and monasteries. Site

Built-up area (dunams)

Cisterns/reservoirs

Agricultural plot

Estimated population

Kh. Kesfa and Umm al-Bureid

17

8/4

2562

c. 290

Kh. Umm al-Hammam and Sites 6, 10

15

16/2

2062

c. 240

H. Yeqavim and Sites 12, 21

16

14/1

2863

c. 200

Kh. Ta‘amur and Kh. Tukkim

13

12/2

2568

c. 200

H. Te’ena

5

3/1

1890

c. 60

Kh. Dayyar

2.5

1

2287

c. 30

Total

68.5

50/10

14000

c. 1100

207

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods west. In most excavated farmhouses, one water cistern was found, indicating the water consumption volume of their residents.

Afeq and Deir Balut) reached c. 1,100 persons. An individual calculation for each site yielded similar data: Every ‘rural town’ had an agricultural plot of c. 2000 dunams, and each resident owned over eight dunams. Even if this calculation were inaccurate, it is still clear that some rule set the ratio between the number of persons who lived in the ‘rural towns’ and the area at their disposal. It is also plausible that the boundaries of settlements were determined by the official authorities (see Chapter 10).

In the absence of Roman settlements, we know but little about the population during this period. We have relied on historical sources, especially Josephus Flavius who informed of a large Jewish settlement at Migdal Afeq during the Great Revolt. Burial caves excavated in the vicinity of this site (Tsuk 1993) attest to the presence of such a community. In any case, based on the extent of the settlement, it is estimated that about 500 people lived there and cultivated the areas to its east.

However, caution should be taken when estimating the population count based on the division of the enclosed plots, as there were sites with no such plots (e.g., Umm al-Hammam). In contrast to the ‘rural towns’, occupied by the few hundreds of persons working their fields, the situation in the monasteries was utterly different. Each monastery owned a large area whose carrying capacity was far higher than that required for meeting local needs. Cultivating these large areas could have been carried out by hired laborers, a practice already customary in the Roman period and perhaps even earlier.

During the Byzantine period, a dramatic increase in population is seen after relatively large settlements were established in the region—’rural towns’ and monasteries—whose residents cultivated the land and processed its produce. Based on the sizes of the built-up areas, the capacity for storing water, and the agricultural plots, it would seem that the population in the region (excluding Migdal

208

CHAPTER 10

HISTORICAL SOURCES AND ARCHAEOLOGICAL FINDS

The Iron Age II (8th–6th centuries BCE)

of Qurnat al-Haramiya, shedding light on the origins of the founders of the farms. It would seem that the ‘four-room house’ was widespread in this area during the Iron Age. Thus, if the builders of the farms came from the surrounding settlements, including Qurnat alHaramiya, they probably built their new buildings in a familiar style, i.e., ‘the four-room house.’ However, by the end of the 8th century BCE, in our region, farmhouses were clearly built according to a new, never used before plan. It may be thus concluded that the founders of the farms originated elsewhere, although we do not know precisely where. Unfortunately, no tombs from this period, which could have shed light on the inhabitants’ identity and their origins, were found in the vicinity of the farms (Faust 2011: 13–32).

Contrary to the generally accepted notion, the Samaria Foothills, which include the ‘farm strip,’ is not a harsh habitat region but rather a relatively comfortable area both for residential and agricultural activities (see Chapter 1). Based on the pottery recovered during the excavations in the farmhouses and on Finkelstein’s studies (1978a, 1981), the region was apparently first occupied in the late 8th century BCE. Major events took place during this period, resulting in political changes throughout the entire region. While in the Kingdom of Israel, the reign of the Jehu dynasty, which ruled for a hundred years (Aharoni 1974: 92), was coming to its end, the Assyrian empire was gaining power after Tiglat Pileser III annexed vast areas of Syria and the Land of Israel and defined them as Assyrian provinces (Aharoni 1974: Map No. 146). The Samaria-Foothills location, between the Coastal Plain and the mountains, made scholars draw its boundaries inconsistently. For example, Aharoni (1974: 95, map 148) drew the borderline between the Kingdom of Israel and the province of Dor, during the time of Tiglat-Pileser III (732 BCE), near Aphek’s region (see also Na’aman 2009: 310–317).

The pattern of small farms was, therefore, essential for maintaining agriculture. Although the farms were remote from the neighboring large settlements, their inhabitants continued to cultivate the same areas that had been cultivated long ago, but living close to the fields significantly increased productivity and reduced the daily transporting time to them. Moreover, during this period, known as Pax Assyriaca, the entire Levant flourished under the Assyrian regime, economically and commercially (Lipschits 2004: 28). These circumstances seem to have opened new perspectives for a significant increase in the trade of agricultural products throughout the various regions. These products were brought from the farmers’ fields, next to the small farmhouses, to the markets in the region’s large settlements.

The Assyrian conquests caused significant demographic and ethnic changes in the Land of Israel (Na’aman 1990: 43–62), possibly affecting also the Samaria Foothills. The Assyrians used to resettle deportees according to their needs, including for agricultural works, and the Samaria Hills are mentioned in this regard in Assyrian documents (Parpola 1997; see also Na’aman and Zadok 2000: 159–188). Thus, settling deportees could have been the reason for establishing the farmhouses in our region. Unfortunately, small finds which could shed light on the Assyrian connection have not yet been discovered, but the building plans of the Persian period may evoke it (see the Persian period, below).

The Babylonian and Persian Periods (6th–4th centuries BCE) Except for one building1 in Site No. 2, showing some similarity to the Assyrian-Babylonian style, no other archaeological remains could be associated with this culture. The Babylonians, who apparently inherited without significant changes the Assyrian administrative system (Aharoni 1987: 313), ruled the Land of Israel for

The farmhouses showed no similarity to the typical ‘four-room house’ associated with the Israelites during the Iron Age (Faust 2005: 237–255; Faust 2006). Many such buildings were found in the nearby site

1 

The building was founded in the 5th or 4th centuries BCE, much later than the Babylonian period.

209

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods 70 years. It is generally accepted that this short period (609–539 BCE) witnessed a significant economic and political decline (Stern 2001: 307–311; Lipschits 2004: 218). However, unlike the areas of Jerusalem, the Judean desert, and the Jordan Valley (Lipschits 2004: 296), the destruction caused by the Babylonians to the large cities was not evidenced in our rural region. It seems, therefore, that no significant changes occurred in our region during the transition between Iron Age II and the Persian period.

to ongoing confrontations between them—the ‘Syrian wars’—in the 3rd and 2nd centuries BCE. The Land of Israel remained under Ptolemy (Stern 1981: 46) until 200 BCE when Antiochus III established in the region a Seleucid regime that lasted about a hundred years (200–104 BCE; Avi-Yona 1984: 30). Avi-Yona (1984: 128–129) argued that the study area was associated with the territory of Antipatris, known in the Hellenistic period as Pegai (‘the springs’— Frankel and Kochavi 2000: 12). In contrast to the meager Persian finds found at Tel Aphek-Antipatris, the excavations revealed remains of a 3rd century BCE preplanned small city, built according to the hippodamic pattern (Kochavi 1989b: 93).

The Persians, like their Assyrian and Babylonian predecessors, apparently maintained the 8th–6th centuries BCE prevailing administrative system (Avi-Yona 1984: 17); Stern 1982: vii; Zertal 1999: 75). Thus, our region remained within the boundaries of the Samaria province (Finkelstein 1981: 347), whose territory remained stable throughout the Persian period (Kallai 1960: 88–89). Based on the excavation finds, the number of settlements continued to increase during the Persian period, and the area reached a clear peak. Considering this thriving, it is surprising how poorly settled Aphek-Antipatris was, with only a single residential structure having been discovered, apparently a farmhouse (Kochavi 1989: 93; Beck and Kochavi 1993: 70, Area F). Perhaps the Persian settlement in place was covered or destroyed by the Hellenistic city of Pegai or the whole area was settled during the period only with small farms (see below).

The high status of Pegai is evidenced in the mid-3rd century BCE letters by Zeno, according to which the city of Pegai was on the main road (Tcherikover 1961: 39). Presumably, Pegai was a main urban center in the Hellenistic period which controlled several smaller settlements in the area (Roll and Ayalon 1989: 128–129), including Migdal Afeq, Qurnat al-Haramiya, and the farms to their east and southeast. The finds revealed in the excavations, and more particularly the coins, indicate the farms were occupied mainly during the 4th and 3rd centuries BCE. The Ptolemaic regime was well represented by coins of Ptolemy I (310 BCE), Ptolemy II (285 BCE) and Ptolemy III (246 BCE; AvnerLevy and Torge 1997; 2018). Additional Ptolemaic coins were discovered in Sites No. 2, 4, and 5, and in the open areas around Qurnat al-Haramiya (Haddad 2011: 57).

Excavations in Site No. 2 revealed a building completely different from the other farmhouses (Shadman 2014; Tendler and Shadman 2015: 193–207; Haddad et al. 2015: 50–67). Unlike most farms in the area, it was founded during the Persian period and was carefully constructed, comprising two courtyards, one of these connected to the storage area to its south. Its plan shows an Assyrian-Babylonian-style influence, later adopted by the Persians (Haddad et al. 2015), as seen in the buildings exposed in the Ramat Beit Shemesh area (Kogan-Zehavi 2014: 120–133). The outstanding features of the building, contrasting with the rest of the buildings in the area, and its location on a leveled plateau adjacent to Road No. 2, suggest that it was an administrative center that ruled over the neighboring farms, while its large courtyard may have served as a marketplace.

Thus, based on the numismatic and ceramic finds, it would seem hardly any farms existed in the 2nd century BCE. The large settlement at Qurnat alHaramiya was also mainly active in the 3rd century BCE, with a meager presence in the 2nd century BCE. It is difficult to determine why activity in the farms ceased in the 2nd century BCE (Faust 2003: 37–53), but it may have resulted from the two following circumstances: 1.

The Hellenistic Period (4th–3rd centuries BCE) Following his victory over the Persians, and after being in reign for 13 years (336–323 BCE), Alexander the Great ruled over the entire East. The internal struggles—‘the Diadochi wars’—that followed his death significantly affected the entire region of the Land of Israel (Thompson 2005: 105–120). The battle at Triparadisus resulted in Ptolemy and Seleucus (Tal 2006: 5) sharing their rule over the Near East, leading

2.

210

Insecure conditions following the Seleucid takeover (200 BCE). Although the Samaria Foothills was a remote region, it was adjacent to the main road, and thus subjected to political and military changes in the area (Rappaport, 2013: 79). These circumstances have been also illustrated in the ‘Heftziba Inscription’, which refers to the brutality of the army of Antiochus III and its governor who settled in the villages and deported their inhabitants (Landau 1966: 54–70; Stern 1981: 65). The other reason could be an economic crash or a severe drought.

Amit Shadman: Chapter 10, Historical Sources and Archaeological Finds It seems, therefore, that at the end of the 3rd century BCE, the inhabitants abandoned their farms, although they apparently intended to return. It is unlikely the new Seleucid regime expelled the inhabitants, as they were taxpayers and productive farmers who could have benefitted the government.

(Wars I, 4, 7; see also Site No. 14) and entrenched in Antipatris and the hills in its vicinity. He probably tried to take advantage of the region’s natural conditions to overcome his enemy (Kaplan 1967: 201–205). The administrative dispositions in the region during the Hasmonean period are not evident. However, based on other regions in the Land of Israel, changes in land ownership seem to have taken place among the rural population, with peasants not being expelled but, instead, becoming tenant farmers (Rappaport 2013: 267). Faust and Erlich (2008: 5–32) argued that the non-Jewish rural population also suffered from the Hasmonean struggles. Unlike the scant evidence of the Hasmonean period in the study area, other parts of the Samaria Foothills yielded extensive evidence. Following the Hasmonean takeover of Pegai in the middle of the 2nd century BCE, the entire region was seemingly administered as one unit.

Contrary to the farms in our region, Seleucid coins were found in communities to its south, at Mazor (which was a small village or a cluster of farms; Amit and Zilberbod 1996b) and Kh. Burnat (Bijovsky 2012: 147–155). These may point to settlement continuity, which characterizes most large sites on ‘the western strip.’ However, even in the farms south of Naḥal Shiloh, not associated with the settlement of Mazor, no continuity was observed. In conclusion, by the end of the 3rd century and beginning of the 2nd century BCE, the rural settlements south of Migdal Afeq seemingly continued to exist, extending in the following Hasmonaean period (see below). These settlements, including Mazor and Kh. Burnat, are located above the alluvial Coastal Plain.

Archaeological finds associated with this period were discovered mainly south of the study area. A building found in Area S2-3 at Mazor yielded coins dating from the end of the 2nd century BCE, and the excavators suggested this may attest to Yohanan Hyrcanus’ activity in the region (Amit and Zilberbod 1996b: 63). Also, they identified evidence of destruction, apparently caused by the Hasmoneans. Excavations at Kh. Burnat, several kilometers south of the Mazor, yielded impressive finds from the HellenisticHasmonean and Roman periods (Amit, Torge and Gendelman 2008: 96–107). Initially, this settlement included several buildings of an earlier Hellenistic farm, but in the 2nd century BCE, when the Hasmoneans took over the district of Lod, it was expanded and became a Jewish village (Rappaport 1995: 61). Several hundred meters south of Kh. Burnat, another Hasmonean settlement was discerned, dating from the middle of the 2nd century BCE when Alexander Balas handed over the area to Jonathan the Hasmonean (Torge 2012b: 64–65).

During the Ptolemaic era, the farms (in the east) coexisted with the large settlements (in the west), but at the end of the 3rd century and the beginning of the 2nd century BCE, the activity in the farms ceased, while in the rural settlements, in the western part of the region, it increased. The numismatic finds from the farms (although few) were clearly dated to the Ptolemaic era. In contrast, the number of Seleucid coins found in the large settlements above the alluvial plain (e.g., Kh. Burnat and Shoham) was remarkably higher than the number of Ptolemaic coins (Zelinger 2009: 150, Table 2), showing that these large settlements, as opposed to farms, continued to exist after the Ptolemaic period. The Hasmonean Period (2nd–1st centuries BCE) The primary historical source for this period is the writings of Josephus Flavius, who described that following a dispute between John Hyrcanus and Antiochus VII, the former sought to take control of Jaffa, Gezer, and Pegai (Antiquities XIII, 9, 2). In 132 BCE, he captured some areas and cities along the coastline, among them the city of Apollonia (Avi-Yonah and Safrai 1966: 37; Avi-Yonah 1984: 43). The numismatic evidence from this city supported this event (Roll and Ayalon 1989: 129). Hyrcanus extended his conquests to include Pegai, which then apparently shared an administrative boundary with Apollonia (Avi-Yonah 1966: 146; Rappaport 1981: 194, 221). Evidence for a Hasmonean presence in the study area was found only in the vicinity of Migdal Afeq, east of Pegai. Later on, Alexander Jannaeus rebelled against Antiochus XII

Another farm dated to the 2nd century BCE was exposed in a large-scale excavation along Route 444 (Dahari and ‘Ad 1999: 56–59). This farm was overlaid by a Hasmonean fortress that included a wall, underground systems, a water drainage system, and coins dated to Alexander Jannaeus. Based on the results from several excavations carried out south of the study area (Zelinger 2009), and mainly on the finds from the excavations at Mazor, where evidence of destruction was noted, it would seem that the Hasmonean takeover of this region was achieved through bitter struggles. Nevertheless, in other villages, such as Kh. Burnat, the Hasmonean destruction was not always evident. It should be recalled that the farms in the study area, north of Naḥal Shiloh, were not destroyed. 211

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods In contrast, the farms near Mazor (Fig. 10.1—red dots), founded in the Early Hellenistic period (probably by a Pagan population), were destroyed by the Hasmoneans. Thus, the following may be concluded: 1.

2.

practice for an extended period (2003: 33). Based on their study, the relationship between this burial custom and the Jewish population during the Second Temple period cannot be ignored. The association of burials in kokhim caves with a Jewish population may point to this ethnic group in our region (Zelinger 2009: 132). A kokhim cave dated to the end of the Second Temple was excavated north of Midgal Afeq (Tsuk 1993: 67–76). The miqva’ot, appearing in the late 2nd century BCE—the Hasmonean period (Adler 2011: 61)—were discovered mainly in Shoham and Kh. Burnat (Torge 2012: 2).

The farms north and south of Naḥal Shiloh and east of ‘the western strip’ were abandoned before the Hasmonean period, probably at the end of the 3rd century BCE, for if they had been inhabited during this period, their destruction would be evident. The farms in Mazor and its vicinity, founded in the Hellenistic period, continued to function at least until the Hasmonean conquest.

Study of the region between Antipatris on the north and Tel Ḥadid on the south revealed that, in the Hasmonean period, the population concentrated south of Naḥal Shiloh. However, the archaeological evidence from Migdal Afeq showed that this site and its surroundings were probably inhabited by Jews in the Hasmonean period. A Jewish occupation is evident here also a few decades later, in the Roman period, based on historical sources and archaeological finds.

In any case, it is most likely that Jewish rural occupation expanded, during the Hasmonean period, into ‘the western strip’, an expansion at least partly peaceful well illustrated in the excavations at Kh. Burnat (Amit, Torge and Gendelman 2008; Torge 2012b) and the site on Route 444 (Dahari and ‘Ad 1999). Moreover, many parts of the area may have been inhabited by Jews already during the Seleucid regime, and by a nonJewish population that accepted (or did not oppose) the new Hasmonean regime.

The Roman Period (1st century BCE–4th century CE) The study area is located in this period between two important cities: Lod (Diospolis), c. 18 km to its southwest, and Antipatris, three kilometers to its northwest. The territory of Lod bordered with Antipatris on the north and with Timna on the east (Avi-Yonah 1984: 106; Safrai 1980b: 199, 201–202). The territory of Antipatris bordered with Caesarea in the north, and with Timna in the east (Safrai 1980b: 105–106). The communities of Qalqilea, Kfar Saba, Migdal Afeq, Beni Brak, and Rantya were associated with Antipatris, while Ono and Yahud were associated with Lod (Avi-Yonah 1984: 128). Naḥal Shiloh served as a clear natural buffer between the territories of Antipatris and Lod, and H. Zikhrin, located south of Naḥal Shiloh, was apparently attributed to Lod (Taxel 2005: 9).

The Identity of the Population in the Hellenistic/ Hasmonean Period The many events that took place between the conquest of the Land of Israel by Alexander the Great and later by Rome (63–37 BCE; 37 BCE—Herod’s coming to power) shaped the entire region and must have caused demographic and ethnic changes, yet it would seem that, most of the time, a mixed population lived in our region. Jewish population lived in Lod and its vicinity already in the Persian period (Safrai 1980b: 70) and likely continued to exist also in the Hellenistic period. It is also possible that, following the conquests of Alexander the Great, the regime gave veterans lands that were, in fact, cultivated by local tenant farmers (Stern 1981: 72). The newly established cities in the Land of Israel also naturally contributed to the population’s increase. Consequently, an ethnic mixture was created, composed of Greek-Macedonians and local people.

Despite the importance of Antipatris in the Roman period, its status was somewhat unclear (Frankel and Kochavi 2000: 21–22). Josephus Flavius reported that Herod founded a new city—Antipatris, named after his father (Wars I, 21)—after the city of Caesarea had been dedicated to Caesar Augustus in 9 BCE. The New Testament mentions a campaign started by Claudius Lysias to Caesarea (58 CE) during which Paul was secretly sent to Antipatris (Acts 23: 31–32). Later on, in 66 CE, Josephus reported on Gallus’ campaign against the Jews (Wars II, 19: 1, 9), and mentioned the Jewish settlement of Migdal Afeq. During the Bar Kokhba Revolt (132 CE), Antipatris is mentioned, mainly in the rabbinical sources, as a market city on the border of Judea (Bava Metzia 11: 7, 11; Frankel and Kochavi 2000: 19–20). Eusebius, in the 4th century CE, mentioned the area several times and emphasized its road system (Safrai 1980b: 195–219).

Identification of the population in the Hasmonean period is based on historical sources and archaeological finds showing clear evidence enabling such an identification. Two main features contribute to identifying a Jewish population: Burial caves with kukhim, and ritual baths (miqva’ot). Kloner and Zissu argued that, according to rabbinical sources, burying in kokhim was not a mandatory custom but extensively addressed the subject. They attributed the beginning of this custom, in the vicinity of Jerusalem, to the 2nd century BCE, emphasizing it remained a common 212

Amit Shadman: Chapter 10, Historical Sources and Archaeological Finds

Fig. 10.1. ‘The farm strip’. Black dots: Farms excavated by the author or surveyed by Finkelstein (1978a, 1981). Red dots: Farms established in the Hellenistic period.

213

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods The archaeological evidence is somewhat confusing for the Roman period, with a significant decline in the number of sites and the total abandonment of the farms. Except for the occupied sites of Migdal Afeq and Qurnat al-Haramiya, only a few potsherds were discovered at Kh. Ta‘amur, H. Yeqavim, and Site No. 12. Abandonment of open agricultural areas raises the question of why did the inhabitants prefer, in the Roman period (for over 400 years), to gather in large settlements, leaving the farms unoccupied. This situation could be attributed to the security conditions prevailing at the time, but it seems odd that this area remained neglected for hundreds of years. Moreover, the Roman period, including the Pax Romana period, was, to some extent, a peaceful era (Levine 1984: 37) and is therefore inconsistent with harsh security conditions, except for activities carried out by bandits and lawbreakers.

the Jewish occupation had declined at an earlier time. The pressure applied by Rome against the Jews already before the Bar-Kokhba Revolt (Dar 2015: 111, 117) also contributed to the abandonment of the area by the Jewish community. During the Roman period, the occupational vacuum east of ‘the western strip’ was also observed in adjacent areas, mostly surveyed but not excavated. Between longitudes 197 and 200, in the southern part of the Lod map, the survey of the area showed that it was almost vacant in this period (Gophna and Beit Arieh 1997: 107, map 4). This area was settled again in the Byzantine period with the sites of Kh. al-Beida and Kh. Ba‘ana (ibid. 108, map 5). Only the area next to the Gophna–Antipatris main road, especially in the vicinity of H. Anusha H. Leved, was also occupied in the Roman period. This situation is consistent with the clear preference for settling near the main roads in this period.

‘Ad (2000) rightly noted that, in this period, people preferred, at least in our region, to live in settlements located alongside the main roads, like Migdal Afeq and H. Zikhrin. However, even this distinction does not explain the preference mentioned above. It would seem, therefore, that between the end of the 3rd century BCE, when the farms were abandoned, and the 4th century CE, the authorities must have prohibited settling in the open areas and occupying the farms, or that at least they had no interest in such settlements.

In conclusion, it seems that the ‘settlement vacuum’ in the open areas during the Roman period resulted from three main factors: 1. A governmental policy which aimed to avoid problems that might emerge from a lack of control over open rural areas. If so, the Jews and the Samaritans seem to have disturbed the Roman rule in the area, both after or before the Bar Kokhba Revolt (see below). Accordingly, Dar suggested that the fortress at Site No. 12 was meant to prevent occupation in our region. 2. Cultural preference, with cities playing an important cultural and commercial role in the Roman imperial realm (Faust and Safrai 2015: 172–173). Past studies have even suggested that founding Roman cities was a means by which the presence of a new regime—of the conquerors— was reported (Jones 1971: 60). Recent studies suggest that establishing Roman cities on major roads contributed to a more efficient control over the various regions of the empire and tax collection, especially in the rural areas (Bar 2008: 52–53). Harmonious relations between the urban centers and the rural periphery seem to have been maintained (Cameron 1993: 153). 3. Agriculture. Presumably, the study area was used for a long period as the agricultural hinterland of the large communities of Migdal Afeq and Deir Balut. Cultivating the lands, carried out by the small farms until the Hellenistic period, ceased in the Roman period. Considering the archaeological and historical evidence, the site of Migdal Afeq appears to have functioned as a central settlement in the western part of the study area. The farmers commuted to the field daily—a walking distance of only 2.3 km. It is also clear that Migdal Afeq

Officials and Jewish landlords who lived in Western Samaria during the reign of Herod, and apparently even during the preceding Hasmonean period, are known from historical sources (Dar 1982a: 73–75; Raviv 2013: 129–132). Thus, presumably, landlords who controlled large areas in their vicinity lived in Migdal Afeq and even at Deir Balut and Rafat. It should also be mentioned that no real historical evidence confirmed the presence of Roman veterans in our region, contrary to other regions in the land of Israel (Wars VII, 6, 6; Bull 1966, Richardson 1976: 47–48; Butcher 2003: 399–421; Klein 2011: 315, 326). In any case, it seems that the government’s policy continued to prohibit occupying open areas, resulting in settlement concentrations mainly on the western and eastern stripes of the study area. The Jewish settlement of Migdal Afeq apparently survived Gallus’ 66-CE campaign. While a distinct destruction layer associated with the Great Revolt was found in the neighboring Antipatris (Kochavi 1989: 101–102), no such evidence was present in Migdal Afeq. However, although there is no direct evidence regarding this site and its vicinity after 70 CE, it clearly underwent some trauma. The absence, in our region, of the hiding systems attributed to the Bar Kokhba Revolt (see below) and other Jewish features may suggest that 214

Amit Shadman: Chapter 10, Historical Sources and Archaeological Finds and the agricultural area around it maintained close relations with the city of Antipatris and its markets.

was severely damaged by this earthquake, including the cities of Lod and Gophna (Brock 1977).

The Bar Kochba Revolt

Jerome, in his description of the journey of the pilgrim Paula, mentioned that Antipatris had been semi-destroyed (Limor 1998: 133, 141), but that it nevertheless still maintained its status among the settlements in the region. For example, the name of Bishop Polychronius of Antipatris was mentioned in the lists of the 5th century CE ecclesiastical conferences (Dauphin 2000: 56–58) showing the ethnic-religious composition of the local population (see below). The List of Duxperius, edited in the 12th century, describes the 5th century CE Christian administrative regions, including several settlements in the Samaria Foothills and Antipatris, which then belonged to the territory of Jaffa (Safrai and Dar 1997b: 92).2 It seems, therefore, that during the 5th–7th centuries CE, Antipatris served as an administrative and religious center for the rural settlements in the area, presumably on behalf of the government and the Church (Kochavi 1989: 121). Despite the damage caused to the city in 363 CE, there was no significant demographic change in the region. Not long after the earthquake, there was a significant increase in the number of settlements in the area (including the study area), beginning already early in the 5th century CE and continuing in the 6th century CE. However, Taxel (2005) argued that most Byzantine building activity in H. Zikhrin should be dated to the 6th century and not earlier. This settlement expansion took place throughout the Land of Israel and reached its peak in Emperor Justinian’s time (357–527 CE; Tsafrir 1996: 274).

Only meager evidence from the Bar Kokhba Revolt period was found in the study area: a 1st–3rd-centuries CE kokhim burial cave north of Migdal Afeq (Tsuk 1993: 67–76). The cave was used by the Jewish residents at least until the Great Revolt, after which, according to Josephus, the site was destroyed. However, it is not clear whether after this destruction the Jewish settlement was renewed or not, for later structures were built over the Roman settlement, pointing to such a renewal (see below). Unlike the lack of evidence regarding the Bar Kokhba Revolt period in the study area (excluding Migdal Afeq), it was explicitly discerned in many other parts of the Samaria Foothills and south of it. Many studies conducted in the vicinity of Shoham, for example, revealed that the predominantly Jewish Hasmonean-period settlements continued through the Early Roman period (1st century BCE–1st century CE) and up to the Bar Kokhba Revolt (Torge 2012b; Amit, Torge and Gendelman 2008; Zelinger 2009: 56–77). Presumably, several Jewish settlements in this area took part in the Great Revolt but were apparently not seriously affected, continuing for another 60 years. Thus, it seems that Antipatris and our study area marked the northwestern end of the Jewish occupation in Judea until the Bar Kokhba Revolt (Eshel and Zissu 2015: 11, Figs. 1, 17, 67, 110; Zissu 2001: 271–301). Hiding systems dated to the Bar Kochba Revolt were found in Shoham, south of our region, and in Elqana, to the east, but not in the study area.

Examining the survey maps conducted in the study area and its vicinity reveals a similar pattern in the Samaria Foothills and the adjacent areas to the east. The surveyors of the map of Rosh Ha‘ayin reported over 70 sites from the Byzantine period (Kochavi and Beit Arieh 1994: 10). In the map of Lod, adjacent on the south, over 100 Byzantine sites were recorded (Gophna and Beit-Arieh 1997: 11). ‘Ad’s study (‘Ad 2000) also demonstrated the vast expansion of sites in the Samaria Foothills during the Byzantine period—43 sites spread over a strip of 20 × 2.5 km, 25 of these established in locations never occupied before. His study also revealed that in the large settlements, especially in those located above the alluvial plain, the occupation continued from the Roman to the Byzantine periods. However, as with the study area, the small satellite sites (including monasteries and farmhouses), adjacent to the large ones, were first founded in the Byzantine period on virgin soil (e.g., H. Te’ena and Kh. Dayyer). It is also clear that Antipatris, in the north, and Lod, in the south, were the only urban settlements in the study area and its vicinity.

The archaeological and historical evidence illustrates the clear identity of the population during the Roman period, up to the Bar Kokhba Revolt. No significant ethnic changes occurred in the region following the end of the Hasmonean rule and the inception of the Roman regime. The Jewish population was present in ‘the western strip’ until the end of the 2nd century CE (Safrai and Dar, 1997b: 88), with gentile population presumably settling there following the Bar Kokhba Revolt and its harsh consequences. The Byzantine Period (4th–7th centuries CE) In 409 CE, when a new administrative division was established in the Land of Israel (Taxel 2005: 9), the study area became part of Palaestina Prima (Tsafrir, Di Segni and Green 1994: 17, figure 4). The city of Antipatris, which reached its peak in the 3rd–4th centuries CE, was destroyed in the 363-CE earthquake (Kochavi 1989: 116; Frankel and Kochavi 2000) but was eventually wholly restored. Historical sources state that the whole area

2 

215

The list itself was edited in the 12th century.

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods In the 5th century CE, large villages—‘rural towns’— began to appear, possibly providing a worthy substitute for dwelling in the large cities. The large Byzantine settlements were built along longitude 199–200 and at a distance of 1.0–1.5 km from one another. This pattern also continued south of the study area. Establishing ‘rural towns’ was a new settlement pattern, not previously discerned in the area between Migdal Afeq (on the west) and Deir Balut and Rafat (on the east). The buildings and rooms in these villages were densely constructed following no specific plan (see also Hirschfeld 1997a: 33–71), although a few outlines of streets or alleys were often noticed. Lack of planning in Byzantine villages is a widespread phenomenon both in the Land of Israel and elsewhere (Hirschfeld 1997b: 118–119).

extent, with the Samaritan rebellions that broke out at the time of Anastasius (498 CE). The settlements continued to exist, and perhaps to expand, during the reign of Justinian (527–565 CE), and it is possible that these reflect a second occupational stage within the Byzantine period (Safrai 1995b). Additional settlements and monasteries were discovered east of the study area, e.g., Deir Kala‘a and Deir Sam‘an (Dar 1986; Finkelstein and Lederman 1997: 159–170; Yitach and Baruch 2001: 159–170; Hirschfeld 2002). There are actually more north-south strips in the Western Samaria Hills: ‘the rural-towns strip’ and ‘the eastern strip’ are 2 km apart, as are ‘the eastern strip’ and the strip further east. Similarly, the monasteries’ line is about 1.0–1.5 km west of ‘the ruraltowns strip’, and 2.0–2.5 km west of the monasteries’ line are the large settlements above the alluvial plain where many Christian features were identified. Thus, during the Byzantine period, our region was seemingly extensively developed for several political, economic and security reasons (See also Yitach and Baruch 2001: 168).

Scholars have assumed that one reason leading to founding ‘rural towns’ at the beginning of the 5th century CE was probably the 363-CE earthquake (Fischer 1985: 121). However, this was certainly not the only reason. The vast number of villages, monasteries, and farms built during this period in our region indicates that their inhabitants came not from a single city and that migration was probably a long process. Also, besides the cities’ destruction at the end of the 4th century CE, there was a significant increase in settlement distribution and population size. The significant improvement in agricultural technologies (Feliks 1982: 419–441), increasing exports and improving trade (Dan 2006: 58–73), also affected the demographic growth (see Broshi 1982: 442–455).

Although non-Christian populations apparently lived in adjacent areas, the authorities, for some reason, seem to have kept the study area and its surroundings (mainly to the south and north) clear from non-Christian populations. The presence of monasteries and Christian settlements along the main roads, including those used by pilgrims on their way to Jerusalem, was apparently of great importance, as they probably functioned as road stations for the pilgrims.

What was the reason for this significant growth in settlements? Had the governmental policy which prevailed in the Roman period been changed? Did this relate to the identity of the population of the region? Despite the limited area of the present study, the evidence is clear: The number of settlements increased from five in the Roman period to 20 in the Byzantine period. Magen (2002b: 245–271) mentioned the relation between the Christian and the Samaritan populations and their distribution pattern. He argued that the modern Cross-Samaria Highway behaved as a geographic dividing line, south of which monasteries and churches abounded (see also Di Segni 2001: 36). This occupation aimed to prevent the Samaritans from spreading southward (Magen 2002b: 270). The restrictions applied on the Samaritan population by the Byzantine-Christian regime caused it to seek living quarters elsewhere (see below).

Regional Hierarchy and Land Management The study area, although associated with Antipatris, maintained an internal hierarchy that comprised large settlements, monasteries, estates, and farmhouses. According to Hirschfeld (1997b: 119), the lack of planning, which characterizes the Byzantine village, points to an organic development accompanied by a high level of autonomy in managing the village’s life. It means that the local Byzantine government, and probably also the earlier one, did not interfere in the village’s daily life and its internal organization. However, although the head of the village apparently handled the village’s affairs (see Taxel 2005 regarding H. Zikhrin), the location of the settlements in the countryside and their size seem to have been subjected to the policy dictated and controlled by the authorities. This policy is evidenced in the similar and somewhat standard extent of the settlements in the ‘rural towns strip’, c. 10–15 dunams, and their associated agricultural plots. The settlements’ density, usually ‘enclosed villages’, shows that their expansion was restricted. Presumably, the limited vacant residential and agricultural areas required from the Byzantine authorities to control

No finds pointing to a Samaritan population in the study area were discovered. It is also possible that ‘the rural towns strip’ and the monasteries to its west were devised to prevent the infiltration of undesirable populations. The period when these settlements were founded—the 5th century CE—overlaps, to a certain 216

Amit Shadman: Chapter 10, Historical Sources and Archaeological Finds land distribution. The lands were generally managed or owned by various statutory landlords (Safrai 1998: 37–38), as follows:

delimit them. This monastery was actually composed of two ‘circles’: the inner circle included the hill on which the monastery was located, encircled by a double wall. The outer circle included the agricultural plot, marked by large stones on its east, north, and west.

1. Imperial lands owned by the Emperor. This status is evidenced in several historical sources (Sion 2004: 191). Emperors did not refrain from confiscating private lands and annexing them for their benefit. Sometimes, the emperor’s lands were unlimitedly leased, but the lessee had to pay taxes on the lands (Dan 1976: 215, 452). 2. Church lands. The Church owned many lands during this period, as evidenced in other regions such as Emmaus, Jerusalem, and the Judean Desert (Dan 1976: 183–188). 3. Non-farmer landlords. At the end of the 4th century CE, ‘The Tenant Farmer Law’ (colonus) was implemented in the Land of Israel and had a long-lasting effect on the land ownership system (Dan 1990b: 188–190; Sion 2004: 189). The status of tenant farmers caused problems between the landlords, who gained great power, and the tenants forced to remain confined to their land. The landowners apparently lived in the city or a mansion near their lands (Safrai 1997b: 47). Despite this apparent oppressing reality, there were quite a few cases which showed that the tenants often had large areas at their disposal and that many tenants were actually quite free (Dan 2006: 66–70). 5 Lands owned by farmers. It is difficult to identify this status, but it would seem these were farmers who cultivated their fields, either through a lease or as tenant farmers. Unfortunately, there are no written sources for our region, but possibly the Nessana papyri can shed light on daily life and land ownership in the late Byzantine period (Kraemer 1958; Stroumsa 2008), including on our study area and other regions (Safrai 1997b: 41–43).

The large villages (Kh. Kesfa, Kh. Umm al-Hammam, H. Yeqavim, and Kh. Ta‘amur) were apparently owned by private landlords. The head of the village and (or) someone on his behalf lived outside the community but close to it. In conclusion, the settlements were essentially agricultural villages whose lands were cultivated by their residents. The number of installations found in the area (mainly oil and wine presses), along with the high density of agricultural systems, indicates that the inhabitants produced surplus crops that enabled commerce with the neighboring regions. The Byzantine period was characterized by a road network, mainly with roads extending from the Samaria Hills westward to the Sharon Coastal Plain. The port city of Apollonia was one of the most important cities in the Coastal Plain, as it was the main and only port in the southern Sharon area (Roll and Ayalon 1989: 27). Since the settlements in the Sharon area and the Samaria Hills were apparently auto-sufficient, olive oil and wine surpluses were exported overseas via the Apollonia port. The Identity of the Population in the Byzantine Period The Jewish-population disappearance from large parts of the region following the Bar-Kokhba Revolt was evident hundreds of years later. Based on archaeological finds, it would seem that starting in the early 5th century CE, the area was occupied by a distinct Christian population, as evidenced by the carved crosses on potsherds and sarcophagi covers. The excavations at H. Te’ena and H. Zikhrin, also support this notion (Fischer 1985, Texel 2005, Taxel 2013). However, the subterranean spaces in Kh. Umm al-Hammam and H. Yeqavim may have been ritual baths (miqva’ot), pointing to the presence of a Jewish population.

In light of the above, it is suggested that land management in our region included two main types of ownership: church lands and non-farmer landowners. The monasteries at Kh. Dayyar, Kh. Umm al-Bureid and H. Te’ena were seemingly owned by the Church, apparently confirmed by the permanence of a bishop in nearby Antipatris and by the inscription found at H. Te’ena which relates the assistance provided by a priest named Theodosius, under whose hospices the monastery was built. Generally, monasteries received financial donations from visitors and high-status urban residents, but not from the surrounding villages (Di Segni 2001: 35–36). This financial-source reality was possibly the reason behind the distribution of the monastery’s lands at H. Te’ena and the monks’ desire to

The Samaritan population was discerned in other parts of the Samaria Foothills, e.g., Ẓur Natan (Ayalon 2002: 273–288), the Southern Sharon, and the Coastal Plain (Ayalon 1989: 273–278; Dar 2002: 444–453; Tal and Taxel 2015), but not in the study area. Only two finds in the region were identified as Samaritan: a sarcophagus, discovered in Antipatris (Eitan 1967: 114–118; Barkai 2002: 316; see also Magen 2002a: 238–239) and another one found at H. Zikhrin (Taxel 2005: 179). 217

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods The Early Islamic Period (7th–9th centuries CE)

Afeq, H. Zikhrin, Mazor continued to exist. The same pattern was also obvious in the Lod map (Gophna and Beit Arieh 1997).

This discussion focuses mainly on the period of the Umayyad and Abbasid dynasties. Following the Islamic conquest, the Palaestina Prima province was renamed Jund Filastin, and its capital was moved from Caesarea to Lod, and later to Ramla (Gil 1981: 50). Ramla was founded in the early 8th century CE by Suleiman, the son of Abd al-Malik, and was the first Islamic city in the country. According to the new administrative division, the southwestern Samaria Hills were administered from Antipatris (Taxel 2005: 11), now called Abu Futrus (Kochavi 1989: 121). The 750-CE massacre at Antipatris, during which eighty Umayyad individuals were murdered, actually marked the transition between the Umayyad and the Abbasid dynasties (Gil 1983: 73; Frankel and Kochavi 2000: 23).

The question of how the Islamic conquest affected agriculture in general, and the wine industry in particular, has been a central issue in archaeological research. Compared to the enormous production capacity of the Byzantine-period wine industry (Ayalon 1997b: 163; Ayalon, Frankel and Kloner 2012), the archaeological finds from the Early Islamic period show that it gradually ceased (Ayalon 1997b: 149–166), or was, at least, significantly reduced. Various reasons have been suggested for this break: Abandonment of settlements in the 8th–9th centuries CE (Kedar 1985: 1–15), the decline of commercial relations due to the struggles between Byzantines and Muslims, or climate changes (Isser and Govrin 1992: 67–83). It should be noted that most installations in our rural mountainous region are hewn in bedrock, making it difficult to determine their abandonment date. Nevertheless, in several cases, changes have been observed in the oil presses located within the settlements, and in some adjacent winepresses. For example, during the 8th century CE—the transition period between the Umayyad and Abbasid dynasties—the large winepresses at Kh. Kesfa and H. Te’ena were converted into quarries. The winepresses’ termination may have resulted from the Islamic attitude toward wine consumption. However, if this were the case, this would have been expected to occur after the 7th century CE, whereas Kh. Kesfa and H. Te’ena were occupied to a large extent by Christians during this period. It is worth noticing that the Koran does not favor drinking wine, but it does not explicitly prohibit it (Wensinck 1987: 894–897). However, Theophanes’ Chronicle (Turtledove 1982: 91) mentioned that Caliph Omer II (717 CE), unequivocally claimed that drinking wine is forbidden.

There was a sharp decline in the settlement density in the study area, and only ten of the 20 Byzantine settlements subsisted during the 7th–8th centuries CE. This decline was also evident in the small number of Islamic potsherds found in those sites which had also been occupied during the Byzantine period. The main occupation area was in the ‘rural towns strip’ (except Kh. Ta‘amur) and the monasteries (except Kh. Dayyar). ‘The farmhouses strip,’ which was not settled in the Byzantine period for residential purposes, remained vacant during the Early Islamic period, while all the other sites were either founded atop earlier Byzantine sites or maintained them. Settlement decline did not result from any specific event. Some scholars claimed it occurred, in various geographic areas, as early as in the Byzantine period (mainly in the 6th and 7th centuries CE) as the result of earthquakes and epidemics (Taxel 2010: 13– 15; Faust and Safrai 2015: 263–267), or of the PersianSassanid conquest (Avni 2010: 35–48) which lasted 14 years (614–628 CE) and likely affected both the Byzantine rule and the region in general. However, such a short occupation could not have had an impact on material culture (Russell 2001: 43). Whatever the case, nothing in the study area may support or disprove any of these explanations.

As olive-oil consumption was not prohibited according to Islamic law, this issue should be examined from other angles. In Kh. Kesfa, fragments of oil presses were found in secondary use (Magen 2008b: 275–278), but these changes cannot be dated and may have occurred after the Early Islamic period. In other cases, e.g., Kh. al-Birah, oil presses were built within churches, sometime between the 7th and 9th centuries CE. The excavators of Kh. al-Birah believed that the builders of the olive press, wanting to carry out an ‘anti-Christian’ act, selected the church for this purpose. The oil press at H. Te’ena was modified already during the Byzantine period and operated side-by-side with the monastery, as reflected in the general plan of the site. At H. Tinshemet, 9 km south of the study area, an oil press was built as an integral part of the church complex (Dahari 2012). Thus, the Islamic oil presses may have been usually located adjacent to churches to take advantage of the roofed buildings already there.

In other neighboring regions, many settlements from the Byzantine period seemingly continued into the Early Islamic period (e.g., Kh. alBireh—Safrai and Dar 1997b; H. Hermeshit—Yron-Lubin 1997: 68–74; Greenhut 1998; H. Leved—Sion and others 2007; see also Avni 2009: 29–42). However, examining the ‘survey maps’ conducted in the Samaria Foothills, a similar settlement pattern emerges, especially in the large sites. In the Rosh Ha‘ayin map, the absence of Early Islamic finds in the open areas and the agricultural installations is evident (see below), although main sites, such as Kh. Umm el-Hammam, H. Yeqavim, Migdal 218

Amit Shadman: Chapter 10, Historical Sources and Archaeological Finds The Identity of the Population in the Early Islamic Period

To sum up, it is quite clear that the effects of the Islamic conquest in its early stages (until the 7th century CE) are not evidenced in our region. Even in case of rapid change, this cannot be discerned, since the Byzantine material culture continued to prevail (Taxel 2005: 11). Thus, considering the archaeological finds in the study area, it can cautiously be suggested that the change in the settlement pattern occurred only in the of the 8th century CE. This conclusion is founded on the number of potsherds collected at various sites and changes made to the agricultural installations. The large sites, inhabited also during the Early Islamic period, yielded many potsherds from the 5th–7th centuries, and only a very limited number from the 8th century CE. Thus, based on the pottery and numismatic findings, it is assumed that the area continued to be administered according to the Byzantine system at least until the end of the 7th century CE, as the political and religious changes had not yet permeated the region (Prawer 1981).

On the eve of the Islamic conquest, the study area was clearly occupied by a Christian population living mainly in large villages and monasteries and cultivating the fields in the vicinity. Christianity continued to prevail in our area at least until the end of the 7th century and perhaps even the early 8th century CE. Archaeological finds show that no major events occurred in the area during the Islamic conquest. The monastery at H. Te’ena, for example, was probably abandoned at the beginning of the 8th century CE and was not destroyed or damaged for religious or political reasons. Military personnel or people associated with the Umayyad regime, who took over large areas throughout the country (Gil 1981: 152), had a clear interest in maintaining the region’s nature since its Christian inhabitants were skilled farmers. However, no evidence was found for the presence of a Samaritan population. An important source for understanding this population during the Early Islamic period, and perhaps even earlier, is the chronicle of Abu al-Fatah (Levy-Rubin 2002: 562–586). According to this source, Samaritans dwelled in the Coastal Plain during the Umayyad period, where they apparently had migrated after their rebellions against the Byzantines. Following the Islamic Umayyad conquest, the Samaritans fled to Byzantium rather than to their homeland in the Samaria Hills, a fact which also confirms that our region was devoid of this population. Similarly, there was no evidence of a Jewish presence in our region during this period, and it would seem that Jews did not inhabit the area after the Bar Kokhba Revolt.

The Islamic conquest first affected the social elite and only later on the low classes. Moreover, it was in the Islamic government’s interest to allow farmers in rural areas to maintain their agricultural activities (Amar 1990: 51; 1997: 171). The Umayyad dynasty (660–750 CE), for example, even made efforts to preserve the life patterns that prevailed in the Byzantine period (Kennedy and Andrews 1985: 141–183; Kennedy 2006). Therefore, it is plausible that the vast wine industry, still being operated by Christians, continued at least until the end of the 7th century CE, and gradually declined in the 8th century CE, while the oil presses continued operating during this century, until it, too, ceased in the 9th century CE. The transition between the two periods seems to have been a gradual process, and no major architectural changes indicating the presence of a new social/ethnic entity have been discerned in the large settlements.

219

CONCLUSIONS

This study is the result of extensive archaeological excavations carried out in recent years in the Samaria Foothills between Naḥal Rabah and Naḥal Shiloh. Excavating many sites, including their agricultural systems, produced a broad database which facilitated making a synthesis of the information on rural settlement in the region from the Hellenistic through the Byzantine/Early Islamic periods. The relatively small size of the area under investigation, about 30 sq km, enabled fairly high-resolution research.

these crops could have been grown in almost any area, either rocky or alluvial. In contrast, growing grain, also evidenced in our region, requires the deeper soil of the wadis where dams were built to create small fields convenient for cultivating, and the alluvial areas situated in the western part of the area. Examining 20 sq km, which make up about two-thirds of the study area, revealed that approximately 4290 dunams were used for olive groves, about 5830 dunams for vineyards, and about 2820 dunams for grain. The rest of the area was not apt for cultivation, but the steep rocky slopes could be used for cattle grazing. Raising cattle was an important economic sector in local agriculture, as reflected notably in the farmhouses where half of their build-up areas were usually used for cattle sheds.

Geographically, the study area is in the Samaria Foothills, a region characterized by a rocky landscape with relatively few alluvial areas. Through the ages, the inhabitants were able to exploit these areas by building dams and farming terraces, but for settling the region, the settlers in antiquity had to resolve the issue of the lack of stable water sources.

A major problem in studying open areas is the issue of dating the various features. When it comes to a large area, rather than a limited multi-strata site, building a horizontal stratigraphy is needed for associating the scattered agricultural systems with the different communities. Sometimes it is difficult, and even impossible, to determine the date of these agricultural systems, as their excavation yielded only worn potsherds dating from many periods. For this purpose, in addition to the many excavation results, we examined the settlements themselves and the road network leading to them, as well as the results of OSL tests taken from several farming terraces. It is worth noticing that OSL tests taken from farming terraces unrelated to any specific settlement were found to be less helpful. It became clear that, although the area was first occupied as early as in Iron Age II, the farmhouses were established only during the Hellenistic period. The road network was apparently first established only in the Late Roman period, operating mainly during the Byzantine period, as clearly evidenced by the fact that no Iron Age II–Hellenistic period farmhouses were approached by rural roads. The rural landscape comprised, of course, other installations, with field towers and winepresses being the most common. Most field towers yielded no potsherds, and in the other few, the potsherds found on their floors dated from the Byzantine period. Nevertheless, it is assumed that the towers operated already in earlier periods but ceased

The area had been already studied in the 19th century by researchers concerned mainly with medium- to large-sized sites who hardly documented small structures such as farmhouses. The area was surveyed by Finkelstein in the early 1970s (1978a, 1981), who reviewed the agricultural farmhouses dating from Iron Age II through the Hellenistic period. The Rosh Ha‘ayin Map survey conducted in the early 1980s (Kochavi and Beit Arieh 1994) was of great help for carrying out the present study. In the last two decades, with the development of the town of Rosh Ha‘ayin, additional extensive surveys and salvage excavations have been conducted in the area, some by the author and published here for the first time. Understanding the area required a multidisciplinary approach combining surveys, excavations and extra-archaeological sciences (e.g., OSL), all included in the ‘landscape archeology’ approach. Agricultural fields, adjacent to the settlements, were examined using this approach, and the findings showed that most of the area was utilized for crops characteristic of the central-mountainous landscape of the Land of Israel: vineyards and olive groves. The description by the Roman agronomist Columella in his work de re rustica neatly emphasizes 221

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods to operate after the Byzantine period. Winepresses were also studied in relation to the local communities, and it became clear that most presses in the vicinity of Iron Age–Hellenistic period farmhouses were of the irregular-shaped simple type. It was also evident that winepresses in the agricultural plots of Byzantine sites were plastered, larger, more complex, and with mosaic floors.

emphasizes the former’s importance. Considering the limited number of settlements in the study area during the Roman period, one may question the designation of the area in this period. The answer is apparently in the numerous agricultural installations excavated there in recent years. Although dating the various installations is somewhat problematic, the ceramic finds seem to point to periods when the land was being cultivated. Accordingly, in the Roman period, the study area was mainly used for agriculture, with the large settlements adjacent to the main roads, not the farmhouses located amid the area, being preferred for residence. The farmers apparently commuted daily to their fields, thanks to the developed road network.

Archaeological surveys were one method used in the current research, and one of our goals was to examine the reliability of the results of previous surveys carried out in the study area. The crucial conclusion that emerges from the many excavations conducted by us, especially in the ‘farmhouse strip,’ is that the surveys are, in fact, quite reliable. Finkelstein’s surveys results (1978a, 1981) and the dates he proposed are remarkably consistent with the chronological range emerging from our excavations. However, it should be noted that Finkelstein’s historical analysis of the reason for the farmhouses’ termination reaches somewhat different conclusions than ours. While he argued that the farms were abandoned because of the Hasmonean revolt (Finkelstein 1978b: 72), it became clear in our excavations that the farmhouses did not continue into the 2nd century BCE, as also proved by the numismatic finds.

In conclusion, until the Byzantine period, the study area seemingly functioned as an agricultural zone, probably owned by the residents of Migdal Afeq and Deir Balut. It is not inconceivable that, as a result of the Great Revolt and the Bar Kokhba Revolt, the Roman regime opted for keeping the area, with its important roads, uninhabited. Based on archaeological finds and historical sources, particularly Josephus Flavius’ writings (Wars II, 19, 1), it appears that, following the Hasmonean takeover of the area, it was occupied by a Jewish population until the Bar Kokhba Revolt.

Twenty-two sites dated from the Hellenistic through the Byzantine/Early Islamic periods were selected for the present study. The analysis of these sites shows that each period has a distinctive settlement pattern. The Hellenistic period is evidenced in small agricultural farms (c. 2 dunams) first occupied in Iron Age II. The number of sites in the Hellenistic period was almost identical to that in the Persian period (12 vs. 13 sites respectively). These farmhouses, each hosting probably about 15–25 residents and owning approximately 400 dunams of agricultural fields, were apparently permanent structures rather than seasonal settlements and were usually built on hillsides or a leveled plateau. The crops in this period were vine, olive, and grain. At the end of the 3rd century BCE, the farms were abandoned, and it is possible that the exchange of power between the Ptolemies and the Seleucids had also affected the study area. However, the abandonment of these settlements may have been the result of an ecological catastrophe or a severe drought.

In the 5th century CE, an occupational new wave in the area founded ‘rural towns,’ monasteries, churches, farmhouses, and one estate. Twenty Byzantine sites were recorded, most established during this period. A rough estimation points to each ‘rural town’ being occupied by no more than 300 inhabitants. The Byzantine settlements of Kh. Kesfa, Kh. Umm al-Hammam, H. Yeqavim, and Kh. Ta‘amur are located around longitude 200 and are distanced 1.0–1.5 km from each other. It is possible that this settlement strip played a strategic, significant role in the Byzantine settlement system. The Byzantine settlement pattern is characterized by large settlements with smaller neighboring sites. Thus, the Byzantine settlement pattern was seemingly created under a pre-planned governmental policy that dictated the location of the settlements within the region and their agricultural plots. This Byzantine occupation maintained this pattern at least until the beginning of the 8th century CE, and it would seem that the Muslim conquest of the middle of the 7th century CE had only a limited impact on the area. The Christian-Byzantine character that prevailed for hundreds of years continued in the next two generations, before the Muslim population significantly took over.

During the Roman period, settlement distribution decreased to only five sites in the area, with Migdal Afeq Tower and Deir Balut, on the western and eastern ends of the region respectively, representing ‘real’ villages. The main road connecting Deir Balut and Shechem-Neapolis (Safrai 1986b: 109)

222

BIBLIOGRAPHY

ʿAd U. 2000. Landscape and Pattern in the Lowland of Samaria in the Byzantine Period. M.A. thesis, The Hebrew University of Jerusalem (Hebrew).

Amar Z. 1997. The Relation of Man to the Flora and Agriculture of the Land of Israel in Medieval Times. In Y. Friedman, Z. Safrai and J. Schwartz (eds.). Hikrei Eretz. Studies in the History of the Land of Israel Dedicated to Prof. Yehuda Feliks. Ramat Gan. Pp. 167– 178 (Hebrew).

ʿAd U. 2007. Rosh Ha-ʿAyin, Miẓpe Afeq. ESI 119. ʿAd U. 2008. Elʿad (Mazor). ESI 120.

Amar Z. 2000. Agricultural Products in the Land of Israel in the Middle Ages. Jerusalem (Hebrew).

ʿAd U. 2011. Modiʿin, Khirbat Abu Fureij (Northwest). ESI 123.

Amit D. 1996a. Mazor, 1993–1996. ESI 19: 57.

ʿAd U. 2016. The Agricultural Landscape of the “Zippor Compound” in Modiʿin. ʿAtiqot 84: 25*–60* (Hebrew; English summary, pp. 123–124).

Amit D. 1996b. Mazor, 1993. ESI 18: 58–61. Amit D. and ʿAd U. 1999. Mazor (Elʿad), Site 1. ESI 109: 61*–62*.

Adler Y. 2011. The Archaeology of Purity: Archaeological Evidence for the Observance of Ritual Purity in Ereẓ-Israel from the Hasmonean Period until the End of the Talmudic Era (164 BCE–400 CE). Ph.D. dissertation, Bar-Ilan University. Ramat Gan (Hebrew).

Amit D., Torge H. and Gendelman P. 2008. Ḥorvat Burnat, a Jewish Village in the Lod Shephelah during the Hellenistic and Roman Periods. Qadmoniot 136: 96–107 (Hebrew).

Aharoni Y. 1962. Excavations at Ramat Rahel. Rome

Amit D. and Zilberbod I. 1996a. ESI 18:65.

Aharoni Y. 1974. Carta’s Atlas of the Bible. Jerusalem (Hebrew).

Amit D. and Zilberbod I. 1996b. Mazor, 1994–1995. ESI 18: 61–63.

Aharoni Y. 1987. Eretz Israel in Biblical Times. A Geographical History. Jerusalem (Hebrew).

Amit D. and Zilberbod I. 1999. Mazor (Elʿad), Areas H8 and S6. ESI 109: 55*–56*.

Ahlstorm G.W. 1978. Wine Presses and Cup-Marks of the Jenin-Megiddo Survey. BASOR 231: 19–49.

Amitzur H. 2006. Estates and Villas in Roman-Byzantine Palestine. Ph.D. dissertation, Bar-Ilan University. Ramat-Gan (Hebrew).

Albright W.F. 1923. The Site of Aphek in Sharon. JPOS 3:50–53.

Amiran D. 1990. The Value of Stability in the Southern Hebron Hills and the Impact of Geographic Conditions on Worked Land. Merhavim 3: 17–45 (Hebrew).

Alexandre Y. 2006. Nahal Tut (Site VIII): A Fortified Storage Depot from the Late Fourth Century BCE. ‘Atiqot 52: 131–189. Amar Z. 1990. Topics on Settlement, Agriculture and the Vegetation of Eretz Israel during the Medieval Period. M.A. thesis, Bar-Ilan University. Ramat-Gan (Hebrew).

Ashbel D. 1968. Climate of Israel: The Coastal Plain. Jerusalem (Hebrew).

223

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods Aviam M. 1994. The Christian Settlement in the Western Galilee during the Byzantine Period. M.A. thesis, The Hebrew University of Jerusalem (Hebrew).

Hirschfeld, J. Patrich and R. Talgam (eds.). Man Near a Roman Arch: Studies Presented to Prof. Yoram Tsafrir. Jerusalem. Pp. 29–42 (Hebrew).

Aviam M. 2003. Recent Excavations and Survey of Churches and Monasteries in Western Galilee. In: C. Claudio Bottini, L. Di Segni and L.D. Chrupcata (eds.). One Land – Many Cultures. Jerusalem. Pp. 41–60.

Avni G. 2010. The Persian Conquest of Jerusalem (614 C.E.): An Archaeological Assessment. BASOR 357: 35–48. Avshalom-Gorni D., Getzov N. and Frankel R. 2008. A Complex Winepress from Mishmar Ha-ʿEmeq: Evidence for the Peak in the Development of the Wine Industry in Eretz Israel in Antiquity. ʿAtiqot 58: 49–66 (Hebrew; English summary, pp. 65*–67*).

Avital A. 2014. Representation of Crops and Agricultural Tools in Late Roman and Byzantine Mosaics of the Land of Israel. Ph.D. dissertation, Bar-Ilan University. Ramat-Gan (Hebrew). Avitsur S. 1957. The Yarkon: The River and Its Environment. Tel Aviv (Hebrew).

Ayalon E. 1987. A Byzantine Road and Other Discoveries in the Vicinity of Tell Qasile. Israel – People and Land 4: 9–33 (Hebrew).

Avitsur S. 1968. From Saddle to Wheel: Transportation in Eretz Israel before the Rise of the Automobile. Tel Aviv (Hebrew).

Ayalon E. 1989. The Material Culture of the Samaritans. In I. Roll and E. Ayalon (eds.). Apollonia and Southern Sharon: Model of a Coastal City and Its Hinterland. Tel Aviv. Pp. 273–278 (Hebrew).

Avitsur S. 1976. Man and His Labor: Historical Atlas of Tools and Workshops in the Holy Land. Jerusalem (Hebrew).

Ayalon E. 1997a. Ecological Aspects of Villages in the Southern Sharon Region During the RomanByzantine Period. In S. Dar and Z. Safrai (eds.). The Village in Ancient Israel. Tel Aviv. Pp. 209–227 (Hebrew).

Avitsur S. 1984. The Threshing Floor (Eretz Israel Museum Publications, Man and His Labor Series). Tel Aviv (Hebrew). Avi-Yonah M. 1954. Roads. In E.L. Sukenik (ed.). Encyclopaedia Biblica: thesaurus rerum biblicarum alphabetico ordine digesus II. Jerusalem. Pp. 711–715 (Hebrew).

Ayalon E. 1997b. The End of the Ancient Wine Production in the Central Coastal Plain. In Y. Friedman, Z. Safrai and J. Schwartz (eds.). Hikrei Eretz. Studies in the History of the Land of Israel Dedicated to Prof. Yehuda Feliks. Ramat Gan. Pp. 149–166 (Hebrew).

Avi-Yonah M. 1959. Roman Inscriptions from Maʿagan Mikhael. Bulletin of the Jewish Palestine Exploration Society 24: 36–41 (Hebrew).

Ayalon E. 2002. Horbat Migdal (Tsur Natan) – An Ancient Samaritan Village. In E. Stern and H. Eshel (eds.). The Samaritans. Jerusalem. Pp. 235–251 (Hebrew).

Avi-Yonah M. 1966. The Holy Land from the Persian to the Arab Conquest (536 B.C. to A.D. 640): A Historical Geography. Grand Rapids, MI.

Ayalon E. 2012a. Images from the Land of the Bible: People, Life and Landscapes, 1898–1946. Tel Aviv (Hebrew).

Avi-Yonah M. 1984. Historical Geography of Eretz Israel from the Return to Zion until the Arab Conquest. Jerusalem (Hebrew).

Ayalon E. 2012b. Agricultural and Industrial Installations and Burial Caves from the Roman and Byzantine Periods at Zur Yigʾal. ʿAtiqot 70: 123–144 (Hebrew; English summary, pp. 90*–91*).

Avi-Yonah M. and Safrai S. 1966. Carta’s Atlas of the Period of the Second Temple, the Mishnah and the Talmud. Jerusalem (Hebrew). Avner-Levy R. and Torge H. 1997. Rosh Ha-ʿAyin. ESI 19: 40*.

Ayalon E., Frankel R. and Kloner A. 2012. Wine Production  in the Land of Israel in Antiquity: Advancement in Research or a Completely New Approach? Cathedra 145: 15–36 (Hebrew).

Avner-Levy R. and Torge H. 2018. Rosh Ha-ʿAyin, Qurnat Haramiya. ESI 130.

Bahat D. and Bucheri M. 1963. Migdal Tzedek – A Roman Tower. ESI 8: 19 (Hebrew).

Avni G. 2009. The Invisible Conquest: The Negev in Transition between Byzantium and Islam in the Light of Archaeological Research. In L. Di Segni, Y.

Bagatti B. 1983. Antichi villaggi cristiani della Giudea e del Neghev (SBF 24). Jerusalem.

224

Amit Shadman: Bibliography Bagatti B. 2002. Ancient Christian Villages of Samaria (SBF 39). Jerusalem.

R. Zeevi (eds.). Judea and Samaria: Studies in Settlement Geography I. Jerusalem. Pp. 166–185 (Hebrew).

Banning E.B. 1996. Highlands and Lowlands: Problems and Survey Frameworks for Rural Archeology in the Near East. BASOR 301: 25–47.

Bennett W.J. and Blakely J.A. 1989. The Coarse Ware. In W.J. Bennett and J.A. Blakely (eds.). Tel el-Hesi: The Persian Period (Stratum V) (The Joint Archaeological expedition to Tell el-Hesi 3). Winona Lake, IN. Pp. 139–230.

Banning E.B. 2002. Archaeological Survey. New York. Bar D. 2001. Changes in the Rural Population of Palestine in the Late Roman and Byzantine Periods. Ph.D. dissertation, The Hebrew University of Jerusalem (Hebrew). Bar D. 2008. “Fill the Earth”: Settlement in Palestine during the Late Roman and Byzantine Periods 135–640 CE. Jerusalem (Hebrew).

Ben-Tor A. 1980. Yokneʿam Regional Project Looks Beyond the Tell. BAR 6 (2): 30–44. Benvenisti M. 1970. The Crusaders in The Holy Land. Jerusalem. Benvenisti M. 1984. Cities and Sites in Eretz Israel in the Crusader Period. Kardom 35–36: 130–132 (Hebrew).

Barkai R. 2002. Samaritan Sarcophagi. In E. Stern and H. Eshel (eds.). The Samaritans. Jerusalem. Pp. 310–338 (Hebrew).

Ben-Yosef D. 2007. The Jordan Valley during the Iron Age I: Aspects of Its History and the Archaeological Evidence for Its Settlement. Ph.D. dissertation, Haifa University (Hebrew).

Baruch Y. 2006. Buildings of the Persian, Hellenistic and Early Roman Periods at Khirbat Kabar, in the Northern Hebron Hills. ʿAtiqot 52: 49–71 (Hebrew; English summary, pp. 205–207).

Ben-Zvi Y. 1990. Water in the Sharon. In A. Degani, D. Grossman and S. Shmueli (eds.). Hasharon Between Yarkon and Carmel. Tel Aviv. Pp. 103–108 (Hebrew).

Baruch Y. 2009. Horbat Susya and Rujum el-Hamiri as a Case Study for the Development of the Village and the Rural Settlement in the Southern Hebron Hills from the Early Roman to the Early Muslim Periods. Ph.D. dissertation, The Hebrew University of Jerusalem (Hebrew).

Berlin A. 2012. The Pottery of Strata 8–7 (The Hellenistic Period). In A. De-Droot and H. Bernick-Greenberg (eds.). Excavations at the City of David 1978–1985 Directed by Yigal Shiloh. Vol VIIB: Area E: The Finds (Qedem 54). Jerusalem. Pp. 5–29.

Beck P. and Kochavi M. 1993. Aphek (In the Sharon). In E. Stern (ed.). NEAEHL 1. Pp. 62–72.

Beyer G. 1951. Die Kreuzfahrergebiete Südwespalästinas. ZDPV 68: 148–282.

Ben-Ari Y. 1989. The Shephard and His Flock in Eretz-Israel during the Time of the Mishnah and the Talmud. M.A. thesis, Bar-Ilan University. Ramat-Gan (Hebrew).

Bijovsky G. 2012. The Coins from Khirbat Burnat (Southwest). ʿAtiqot 69: 147–155.

Ben-Arieh S. 2000. Salvage Excavations near the Holyland Hotel, Jerusalem. ‘Atiqot 40: 1–24.

Blakely J. A. Brinkmann R. and Vitaliano C. J. 1992. Roman Mortaria and Basins from a Sequence at Caesarea: Frabrics and Sources in Vann R. L. (ed.) Caesarea Papers: Straton’s Tower and Byzantine Caesarea. Journal of Roman Archaeology. Supplementary Series 5.

Beit-Arieh Y. and Ayalon E. 2012. Archaeological Survey of Israel. Map of Kefar Saba (77). Jerusalem (Hebrew). Ben-David H. 1989. The Olive and Olive Oil Production in the Golan Heights during the Time of the Mishnah and the Talmud. M.A. thesis, Bar-Ilan University. Ramat-Gan (Hebrew).

Bodenheimer S. 1956. Animal Life in Biblical Lands II: From the Second Temple Period to the End of the Nineteenth Century. Jerusalem (Hebrew). Boneh Y. and Baida U. 1977. Water Sources and Their Exploitation in Judea and Samaria. In S. Shmueli, D. Grossman and R. Zeevi (eds.). Judea and Samaria: Studies in Settlement Geography I. Jerusalem. Pp. 34–48 (Hebrew).

Ben-David H. 1998. Oil Presses and Olive Oil Production in the Golan in the Mishnaic and Talmudic Periods. ʿAtiqot 34: 1–62 (Hebrew; English summary, pp. 5*– 6*). Ben-Herut Z. 1977. The Advancement of Agriculture in Judea and Samaria. In S. Shmueli, D. Grossman and 225

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods Bonfil D.J. and Hadas A. 2011. Ancient Field-Crop Yields and Land-Carrying Capacity. In Y. Dagan. The Ramat Beit Shemesh Regional Project: Landscape of Settlement: From the Paleolithic to the Ottoman Periods (IAA Reports 47). Jerusalem. Pp. 199–210.

Clermont-Ganneau C. 1896. Archaeological Researches in Palestine during the Years 1873–1874 II. London. Conder C.R. and Kitchener H.H. 1882. The Survey of Western Palestine II: Samaria. London. Crawford O.G.S. 1953. Archaeology in the Field. New York.

Brandl B. and Itach G. 2019. A Locally-made Scaraboid from Khallat es-Sihrij Near Tel Aphek and Its NeoAssyrian Connection. Tel-Aviv 46: 211 – 226.

Dagan Y. 2004. Results of the Survey: Settlement Patterns in the Lachish Region. In D. Ussishkin (ed.). The Renewed Archaeological Excavations at Lachish (1973–1994) (Tel Aviv Monograph Series 22). Tel Aviv. Pp. 2672–2690.

Bradford J. 1957. Ancient Landscape: Studies in Field Archaeology. London. Brawer M. 1977. Creep and Dispersion of Villages in Judea and Samaria. In S. Shmueli, D. Grossman and R. Zeevi (eds.). Judea and Samaria: Studies in Settlement Geography II. Jerusalem. Pp. 367–382 (Hebrew).

Dagan Y. 2006. Archaeological Survey of Israel. Map of Amaẓya (109). Jerusalem (Hebrew). Dagan Y. 2010. The Ramat Bet Shemesh Regional Project: The Gazette (IAA Reports 46). Jerusalem.

Brock S.P. 1977. A Letter Attributed to Cyril of Jerusalem on the Rebuilding of the Temple. BSOAS 40: 267–286.

Dagan Y. 2011a. Archaeological Surveys and Excavations along the Cross-Israel Highway: A Case Study for Data Collection from the Field. ʿAtiqot 64: 3*–61* (Hebrew; English summary, pp. 153–155).

Broshi M. 1982. The Population of Eretz Israel in the Roman-Byzantine Period. In Z. Baras, S. Safrai, M. Stern and Y. Tsafrir (eds.). Eretz Israel from the Destruction of the Second Temple to the Muslim Conquest I. Jerusalem. Pp. 442–445 (Hebrew).

Dagan Y. 2011b. The Ramat Bet Shemesh Regional Project: Landscape of Settlement: From the Paleolithic to the Ottoman Periods (IAA Report 47). Jerusalem.

Broshi M. 1986. The Capacity of Eretz Israel in the Byzantine Period and Its Demographic Implication. In A. Kasher, A. Oppenheimer and U. Rappaport (eds.). Man and Land in Eretz Israel in Antiquity. Jerusalem. Pp. 49–56 (Hebrew).

Dahari U. 1996. Ḥorbat Tinshemet, Church of St. Bacchus. ESI 18: 67*–68*. Dahari U. 2003. Excavations at Ḥorbat Ḥani. Qadmoniot 126: 102–106 (Hebrew).

Broshi M. 2001. Methodology of Population Estimates: The Roman Byzantine Period as a Case Study. In M. Broshi. Bread, Wine, Walls and Scrolls (Journal for the Study of the pseudepigrapha, Supplement Series 36). London. Pp. 86–92.

Dahari U. 2012. The Church of St. Bacchus near Horvat Tinshemet. In: D. Chrupcala. (ed.). Christ Is Here!: Studies in Biblical and Christian Archaeology in Memory of Michele Piccirillo. ofm (SBF 52). Milan. Pp. 105–124.

Bull R.J. 1966. A Roman Veteran’s Epitaph from Azzun, Jordan. PEQ 98: 163–165.

Dahari U. and ʿAd U. 1999. Shoham Bypass Road. ESI 20: 56*–59*.

Butcher K. 2003. Roman Syria and the Near East. Los Angeles.

Dahari U. and ʿAd U. 2008. Shoham. In E. Stern (ed.). NEAEHL 5. Pp. 2037–2039.

Cameron, A. 1993. The Mediterranean World in Late Antiquity, AD 395–600. London and New York.

Dahari U. and Zelinger Y. 2014. The excavation at Ḥorvat Ḥani: final report and survey on nuns and nunneries in Israel. In G.C. Bottini, L.D. Chrupcala and J. Patrich (eds.). Knowledge and Wisdom: Archaeological and Historical Essays in Honour of Leah Di Segni (SBF 54). Milan. Pp. 179–203.

Chevallier R. 1976. Roman Roads. London. Chitty D.J. 1966. The Desert City: An Introduction to the Study of Egyptian and Palestinian Monasticism under the Christian Empire. Oxford.

Dan J. and Raz T. 1970. Soil Association Map of Israel. Tel Aviv (Hebrew).

Clark C. and Haswell M. 1967. The Economics of Subsistence Agriculture. New York.

Dan J., Fine P. and Lavee H. 2007. The Soils of the Land of the Israel. Jerusalem (Hebrew). 226

Amit Shadman: Bibliography Dan Y. 1976. Social Life in Eretz Israel in the Byzantine Period in the Sixth and Seventh Centuries. Ph.D. dissertation, The Hebrew University of Jerusalem (Hebrew).

Dar S. 1986c. Agriculture and Agricultural Produce in Eretz-Israel in the Roman-Byzantine Period. In A. Kasher, A. Oppenheimer and U. Rappaport (eds.). Man and Land in Eretz-Israel in Antiquity. Jerusalem. Pp. 142–169 (Hebrew).

Dan Y. 1977. Lands in Judea and Samaria. In S. Shmueli, D. Grossman and R. Zeevi (eds.). Judea and Samaria: Studies in Settlement Geography I. Jerusalem. Pp. 14–28 (Hebrew).

Dar S. 1996. The Connection between the Family and the Dwelling House in Ancient Israel. Eretz Israel 25: 151–157 (Hebrew).

Dan Y. 1984. Urban Life in Eretz Israel in Late Antiquity. Jerusalem (Hebrew).

Dar S. 1998. Sumaqa – A Jewish Village on the Carmel. Tel Aviv (Hebrew).

Dan Y. 1990a. Trade within and Trade without the Land of Israel in Second Temple Times. In B.Z. Kedar, T. Dothan and S. Safrai (eds.). Commerce in Palestine throughout the Ages. Jerusalem. Pp. 91–107 (Hebrew).

Dar S. 2000. Ancient Agricultural Fields in Samaria and Aerial Photography. In B.Z. Kedar and A. Danin (eds.). Remote Sensing: The Use of Aerial Photographs and Satellite Images in Israel Studies. Jerusalem. Pp. 155–165 (Hebrew).

Dan Y. 1990b. Economic Life in Eretz-Israel in the Byzantine Period. In B.Z. Kedar, T. Dothan and S. Safrai (eds.). Commerce in Palestine throughout the Ages. Jerusalem. Pp. 181–194 (Hebrew).

Dar S. 2002. Samaritan Rebellions in the Byzantine Period – The Archaeological Evidence. In E. Stern and H. Eshel (eds.). The Samaritans. Jerusalem. Pp. 444–453 (Hebrew).

Dan Y. 2006. Economic Life in Eretz-Israel during the Byzantine Period (the Fifth to Seventh Centuries). In Y. Dan (ed.). Chapters in the History of Trade in the Land of Israel during the Roman-Byzantine Period. Jerusalem. Pp. 58–73 (Hebrew).

Dar S. 2003. Raqit – Marinus Estate on the Carmel. Tel Aviv (Hebrew). Dar S. 2015. The Function of the Underground Complexes in the Bar Kokhba War. In A. Tavger, Z. Amar and M. Billig (eds.). In the Highland’s Depth 5. Ariel. Pp. 111–122 (Hebrew).

Dar S. 1978. The Historical Background of the Settlements of the Hermon. In S. Applebaum (ed.). The Hermon and Its Foothills. Tel Aviv. Pp. 52–120 (Hebrew).

Dar S. and Applebaum S. 1973. The Roman Road Antipatris to Caesarea. PEQ 105: 91–99.

Dar S. 1980. Khirbet Jemaʿin – A Village from the Period of the Monarchy. Qadmoniot 51–52: 97–100 (Hebrew).

Dar S. and Safrai Z. 1984. Kh. el-Bireh. ESI 3: 11*–13*.

Dar S. 1981. Roads and Fortresses in the Beit-Lid Region of Samaria. Eretz Israel 15: 376–382 (Hebrew).

Dar S. and Ziegelmann A. 1997. H. el-Kerak in the Carmel. In S. Dar and Z. Safrai (eds.). The Village in Ancient Israel. Tel Aviv. Pp. 161–191 (Hebrew).

Dar S. 1982a. The Settlement Pattern of Western Samaria in the Periods of the Second Temple, Mishna, Talmud and the Byzantine Period. Ph.D. dissertation, Tel Aviv University (Hebrew).

Dar S., Tepper Y. and Safrai Z. 1986. Um Rihan: A Village of the Mishna. Tel Aviv (Hebrew).

Dar S. 1982b. Ancient Farmsteads in the Nahal Beit ʿArif Region. Nofim 16: 47–60 (Hebrew).

Dauphin C.M. 2000. Bishop Polychronius of Antipatris. In M. Kochavi, P. Beck and E. Yadin (eds.). AphekAntipatris I: Excavations of Areas A and B: The 1972–1976 Seasons (Tel Aviv Monograph Series 19). Tel Aviv. Pp. 56–58.

Dar S. 1986a. The Rural Landscape of Samaria from the Time of the First Temple until the End of the Byzantine Period: Concluding Seven Years of Research and Excavation in Western Samaria. Nofim 18–19: 7–30 (Hebrew).

Dauphin C.M. and Gibson S. 1992-93. Ancient Settlements in Their Landscapes: The Results of Ten Years of Survey on the Golan Heights (1978–1988). BAIAS 12: 7–31.

Dar S. 1986b. Hirbet Jemaʿin: A First Temple Village in Western Samaria. In S. Dar and Z. Safrai (eds.). Shomron Studies. Tel Aviv. Pp. 13–73 (Hebrew).

227

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods Davidovich U., Porat N., Avni Y. and Lipschitz O. 2011. A New Method for Terrace Dating: A Case Study from Ramat Rahel. In E. Baruch, A. Levy-Reifer and A. Faust (eds.). New Studies in Jerusalem 17. Ramat-Gan. Pp. 473–492. (Hebrew).

Eshel H. and Zissu B. 2015. The Bar Kokhba Revolt – An Archaeological Perspective. Jerusalem (Hebrew).

Dayan A. 2015. Monasteries in the Northern Judean Shephelah and the Samaria Western Slopes during the Byzantine and Early Islamic Periods. Ph.D. dissertation, Bar-Ilan University. Ramat-Gan (Hebrew).

Evenari M., Shanan L. and Tadmor N. 1980. The Negev: The Challenge of a Desert. Jerusalem (Hebrew).

Evenari M., Shanan L. and Tadmor N. 1971. The Negev: The Challenge of a Desert. Harvard.

Fabian P. and Goldfus H. 2004. A Byzantine Farmhouse, Terraces and Agricultural Installations at the Goral Hills near Beʾer Sheva.ʿAtiqot 47: 1*–14* (Hebrew; English summary, pp. 209–210).

Dilke O.A.W. 1971. The Roman Land Surveyors: An Introduction to the Agrimensores. Newton Abbot. Dilke O.A.W. 1985. Greek and Roman Maps. London.

Faust A. 2003. Judea in the Sixth Century B.C.E.: A Rural Perspective. PEQ 135: 37–53.

Di Segni L. 2001. Monk and Society: The Case of Palestine. In J. Patrich (ed.). The Sabaite Heritage in the Orthodox Church from the Fifth Century to the Present (Orientalia Lovaniensia Analecta 98). Leuven. Pp. 31–36.

Faust A. 2005. The Israelite Society in the Period of the Monarchy: An Archaeological Perspective. Jerusalem (Hebrew).

Di Segni L. 2005. Cyril of Scythopolis: Lives of Monks of the Judean Desert. Jerusalem (Hebrew).

Faust A. 2006. Farmsteads in the Foothills of Western Samaria: A Reexamination. In A.M. Maeir and P.R. de Miroschedji (eds.). “I Will Speak the Riddles of Ancient Times”: Archaeological and Historical Studies in Honor of Amihai Mazar on the Occasion of His Sixtieth Birthday. Winona Lake, IN. Pp. 477–504.

Dorsey D.A. 1991. The Roads and Highways of Ancient Israel. Baltimore. Drezner L. 2014. Rosh Ha-ʿAyin, Qurnat Haramiya. ESI 126.

Faust A. 2011. How Were the Israelites Buried? The Lack of Iron Age I Burials in the Highlands in Context. In A. Tavgar, Z. Amar and M. Billig (eds.). In the Highland’s Depth: Ephraim Range and Binyamin Research Studies. Pp. 13–32 (Hebrew). Faust A. and Erlich A. 2008. The Hasmonean Policy toward the Gentile Population in Light of the Excavations at Kh. Er-Rasm and Additional Rural Sites. Jerusalem and Eretz Israel 6: 5–32 (Hebrew).

Edelstein G., Milevski I. and Aurant S. 1998. Villages, Terraces and Stone Mounds: Excavations at Manḥata, Jerusalem, 1987–1989 (IAA Reports 3). Jerusalem. Eidelman A. and Anmer L. 2014. Geological Sites in Israel. Jerusalem (Hebrew). Eitam D. 1979. Olive Presses of the Israelite Period. Tel Aviv 6: 146–155.

Faust A. and Katz H. 2012. Survey, Shovel Tests and Excavations at Tel ʿEton: On Methodology and Site History. Tel Aviv 39: 158–185.

Eitan A. 1967. A Sarcophagus and Decorated Arch from the Mausoleum at Rosh Ha-ʿAyin. Eretz Israel 8: 114– 118 (Hebrew).

Faust A. and Safrai Z. 2015. The Settlement History of Ancient Israel: A Quantitive Analysis. Ramat-Gan (Hebrew).

Eliaz C. 2002. Rosh Ha-ʿAyin. ESI 114: 112*. Elisha Y. 2010. Rosh Ha-Ayin. ESI 122.

Feliks Y. 1982. Jewish Agriculture in the Period of the Mishnah and Talmud. In Z. Baras, S. Safrai, M. Stern and Y. Tsafrir (eds.). Eretz Israel from the Destruction of the Second Temple to the Muslim Conquest I. Jerusalem. Pp. 419–441 (Hebrew).

Ellenblum R. 1992. Construction Methods in Frankish Rural Settlement. In B.Z. Kedar (ed.). The Horns of Hattin: Proceedings of the 2nd Conference of the Society for the Study of the Crusades and the Latin East, Jerusalem and Haifa, 2–6 July 1987. Jerusalem. Pp. 168–189.

Finkelstein I. 1977. The Roman Road from Gophna to Antipatris. In M. Broshi (ed.). Between Hermon and Sinai: Studies in the History, Archaeology and Geography of Eretz Israel. Tel Aviv. Pp. 171–180 (Hebrew).

Elitzur Y. and Ben-David C. 2007. Deir in Hebrew, Aramaic and Arabic and Toponyms of the “Deir-X” Type. Cathedra 123: 13–38 (Hebrew). 228

Amit Shadman: Bibliography Finkelstein I. 1978a. Rural Settlement in the Foothills and the Yarkon Basin. M.A. thesis, Tel Aviv University (Hebrew).

dissertation, (Hebrew).

Bar-Ilan

University.

Ramat-Gan

Fixler Y. and Silman Y. 2000. Computerized Processing of Aerial Photographs for Use in Archaeological Research. In B.Z. Kedar and A. Danin (eds.). Remote Sensing: The Use of Aerial Photographs and Satellite Images in Israel Studies. Jerusalem. Pp. 203–211 (Hebrew).

Finkelstein I. 1978b. Historical-Geographical Notes on the Description of Eretz-Israel in the Peutinger Map. Cathedra 8: 71–79 (Hebrew). Finkelstein I. 1981. Israelite and Hellenistic Farms in the Foothills and the Yarkon Basin. Eretz Israel 15: 331–348 (Hebrew).

Fowler P.J. 1975. Continuity in the Landscape? A Summary of Some Local Archaeology in Wiltshire, Somerset and Gloucestershire. In P.J. Fowler (ed.). Recent Work in Rural Archaeology. Bradford-on-Avon. Pp. 121–136.

Finkelstein I. 1988. The Archaeology of the Israelite Settlement. Jerusalem. Finkelstein I. 1990. A Few Notes on Demographic Data from Recent Generations and Ethnoarchaeology. PEQ 122: 47–52.

Frankel R. 1984. The History of the Processing of Wine and Oil in the Galilee in the Period of the Bible, the Mishna and the Talmud. Ph.D. dissertation, Tel Aviv University (Hebrew).

Finkelstein I. 1997. Method of Field Survey and Data Recording. In I. Finkelstein and Z. Lederman (eds.). Highland of Many Cultures: The Southern Samaria Survey. The Sites (Tel Aviv Monograph Series 14). Tel Aviv. Pp. 11–24.

Frankel R. 1986. Ancient Olive Pressing Installations (Eretz Israel Museum Publications, Man and His Labor Series 7). Tel Aviv (Hebrew).

Finkelstein I. and Lederman Z. 1997. Highland of Many Cultures: The Southern Samaria Survey. The Sites (Tel Aviv Monograph Series 14). Tel Aviv.

Frankel R. 1997. The Wine Press in Eretz Israel and Its Vicinity in the Byzantine Period. In S. Dar and Z. Safrai (eds.). The Village in Ancient Israel. Tel Aviv. Pp. 193–207 (Hebrew).

Fischer M. 1983. Khirbet Zikhrin – 1982. ESI 2: 114–115.

Frankel R. 2003. Mills and Querns in Talmudic Literature – A Reappraisal. Cathedra 110: 43–60 (Hebrew).

Fischer M. 1984. Khirbet Zikhrin – 1984. ESI 3: 113–115. Fischer M. 1985. Excavations at Horbat Zikhrin. Qadmoniot 71/72: 112–121 (Hebrew).

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 International Series 1972). Oxford.

Fischer M. 1989. An Early Byzantine Settlement at Kh. Zikrin (Israel). Actes Du XI Congrès International D’archèologie Chrètienne. Rome. Pp. 1787–1807. Fischer M. 1993. Horbat Zikhrin – 1987/1989. ESI 12: 40–44.

Frankel R. and Kochavi M. 2000. Identification of the Site. In: M. Kochavi, P. Beck and E. Yadin (eds.). Aphek-Antipatris I: Excavation of Areas A and B, The 1972–1976 Seasons. Tel Aviv. Pp. 9–15.

Fischer M. 2008. Zikhrin, Horvat. In E. Stern (ed.). NEAEHL 5. Pp. 2083–2085.

Frayn J.M. 1979. Subsistence Farming in Roman Italy. London.

Fischer M. and Tal O. 1999. The Hellenistic Period. In I. Roll and O. Tal (eds.). Apollonia-Arsuf I: The Persian and Hellenistic periods. Tel Aviv. Pp. 223–261.

Frumkin A. 1992. Karst Origin of the Upper Erosion Surface in the Northern Judean Mountains, Israel. IJES 41: 169–176.

Fixler Y. 1991. Computerized Processing of Panchromatic Aerial Photographs for the Purpose of Mapping Archaeological Sites from the Mishnaic and Talmudic Periods in the Lod Shephelah. M.A. thesis, Bar-Ilan University. Ramat-Gan (Hebrew).

Gadot Y., Davidovich U., Avni G., Avni Y., Avraham A., Kisilevitz S., Ein-Mor D. and Porat N. 2015. The Formation of a Mediterranean Terraced Landscape: The Case of Nahal Refaʾim. In G. Stiebel, O. PelegBarkat, D. Ben-Ami and Y. Gadot (eds.). New Studies in the Archaeology of Jerusalem and Its Region 9. Jerusalem. Pp. 118–142 (Hebrew).

Fixler Y. 2001. Mapping and Analysis of Roman Roads in Palestine Based on Aerial Photographs. Ph.D. 229

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods Gadot Y., Davidovich U., Avni G., Avni Y., Piasetzky M., Faershtein F., Golan D. and Porat N. 2016. The formation of a Mediterranean terraced landscape: Mount Eitan, Judean Highlands, Israel. Journal of Archaeological Science: Reports 6. Pp. 397–417.

Gibson S. 2003. From Wildscape to Landscape: Landscape Archaeology in the Southern Levant – Methods and Practice. In A.M. Maeir, D. Shimon and Z. Safrai (eds.). The Rural Landscape of Ancient Israel (BAR International Series 1121). Oxford. Pp. 1–25.

Gal A. and Ramon U. 2012. Survey of the Hills Axis. Open Landscape Institute. http://deshe.org.il/ wp-content/uploads/2019/01/tzir_hagvaot.pdf (Hebrew).

Gibson S. 2015. The Archaeology of Agricultural in the Mediterranean Zone of the Southern Levant and the Use of the Optically Stimulated Luminescence (OSL) Dating Method. Erlanger Geographische Arbeiten 42: 295–314.

Gardiner A.H. 1947. Ancient Egyptian Onomastica (I–II). Oxford.

Gibson S. and Edelstein G. 1985. Investigating Jerusalem’s Rural Landscape. Levant 17: 139–155.

Gardiner A.H. 1961. Egypt of the Pharaohs. Oxford.

Gibson S., Ibbs B. and Kloner A. 1991. The Sataf Project of the Landscape Archaeology in the Judean Hills: A Preliminary Report on the Four Seasons of Survey and Excavation (1987–89). Levant 23: 29–54.

Garnsey P. 1988. Famine and Food Supply in the GraecoRoman World: Responses to Risk and Crisis. Cambridge. Garnsey P. and Saller R. 1987. The Roman Empire: Economy, Society and Culture. Bloomsbury.

Gil M. 1981. Palestine under Muslim Rule (634–1099). In J. Prawer (ed.). The History of Eretz Israel VI: Under Moslem and Crusader Rule (634–1291). Jerusalem. Pp. 17–160 (Hebrew).

Gat Z. and Karni O. 2003. Climate and Agricultural Meteorology in Samaria and Judea. In Z. Gat and Y. Eshel (eds.). Climate and Agricultural Meteorology in Samaria and Judea. Ariel. Pp. 7–44 (Hebrew).

Gil M. 1983. Palestine during the First Muslim Period (634– 1099) I. Tel Aviv (Hebrew).

Getzov N. 2005. Shaft Planters for Fruit Trees in Quarries of the Late Roman Period in the Western Galilee. ʿAtiqot 48: 99–104 (Hebrew; English summary, pp. 158–159).

Gitin S. 1990. Gezer III: A Ceramic Typology of the Late Iron Age II, Persian and Hellenistic Periods at Tell Gezer (Annual of the Nelson Glueck School of Biblical Archaeology 3). Jerusalem. Glueck N. 1960. Explorations on Eastern Palestine. Tel Aviv (Hebrew).

Geus C.H.J. de. 1975. The Importance of Archaeological Research into the Palestinian Agriculture Terraces, with an Excursus on the Hebrew Word gbī. PEQ 107: 65–74.

Golani A. 2005. Salvage Excavations in the Modiʿin Landscape. ‘Atiqot 50: 73–97.

Geva H. 2007. Estimating Jerusalem’s Population in Antiquity: A Minimalist View. Eretz Israel 28: 60–65 (Hebrew).

Goldreich Y. 1998. The Climate of Israel: Observation, Research and Application. Jerusalem (Hebrew).

Geva H. 2014. Jerusalem’s Population in Antiquity: A Minimalist View. Tel Aviv 41: 131–160.

Gophna R. and Beit-Arieh Y. 1997. Archaeological Survey of Israel. Map of Lod (80). Jerusalem (Hebrew).

Gibson S. 1995. Landscape Archaeology and Ancient Agriculture Field System in Palestine. Ph.D. dissertation, University of London. London.

Gophna R. and Porath Y. 1972. The Land of Ephraim and Manasseh. In M. Kochavi (ed.). Judaea, Samaria and the Golan: Archaeological Survey 1967–1968. Jerusalem. Pp. 195–241 (Hebrew).

Gibson S. 2001a. Archaeology of the Landscape and Environment: Reconstructing the Ancient Landscape Based on Archaeology. Ariel 147–148: 32–44 (Hebrew).

Gorzalczany A. 2006. Petrographic Analysis of the Tel Mikhal (Tel Michal) Pottery. ‘Atiqot 52: 55–65. Gorzalczany A. 2007. Survey and Salvage Excavations on the Menashe Spur, near Ḥorbat Bareqet and Kibbutz Regavim. ʿAtiqot 55: 83–107 (Hebrew; English summary, pp. 59–61).

Gibson S. 2001b. Agriculture Terraces and Settlement Expansion in the Highland of Early Iron Age Palestine: Is There A Correlation Between the Two? In A. Mazar (ed.). Studies in the Archaeology of the Iron Age in Israel and Jordan. Sheffield. Pp. 113–146. 230

Amit Shadman: Bibliography Govrin Y. 1992. Archaeological Survey of Israel. Map of Nahal Yattir (139). Jerusalem (Hebrew).

Haiman M. 1999. Rosh Ha-‘Ayin Area, Survey. ESI 110: 93* (Hebrew).

Granowsky A. 1949. The Land Regime in Palestine. Tel Aviv (Hebrew).

Haiman M. 2007. The Agricultural Landscape of the Yatir Area in the 6th–8th Centuries CE: Aspects of Economy and Ethnicity. In Y. Eshel (ed.). Judea and Samaria Research Studies – Proceedings of the Sixteenth Annual Meeting. Ariel. Pp. 131–146 (Hebrew).

Greenhut Z. 1998. Ḥorbat Ḥermeshit: The 1988–1990 Seasons. ʿAtiqot 34: 121–172 (Hebrew). Grossman D. 1994. Expansion and Desertions: The Arab Village and Its Offshoots in Ottoman Palestine. Jerusalem (Hebrew).

Haiman M. 2012a. Churches and Agriculture in Palestine in the Late Byzantine Period – Results from a Remote Sensing Landscape Project. In E. Baruch, A. LevyReifer and A. Faust (eds.). New Studies on Jerusalem 17. Ramat-Gan. Pp. 403–425 (Hebrew).

Grossman D. 2004. Arab Demography and Early Jewish Settlement in Palestine: Distribution and Population Density during the Late Ottoman and Early Mandate Periods. Jerusalem (Hebrew).

Haiman M. 2012b. Dating the Agricultural Terraces in the South Levant Desert, The Spatial Context Argument. Journal of Arid Environments 30: 1–7.

Gudovitz S. 1996. Mazor – 1993. ESI 18: 57*–58*.

Haiman M. 2014. The Negev Emergency Survey. Devar Baʿavar – Newsletter of the Israel Antiquities Authority 21: 20–21 (Hebrew).

Guérin V. 1984 (transl. H. Ben-Amram). A Geographic, Historical and Archaeological Description of the Land of Israel 5: Samaria. Jerusalem (Hebrew).

Haiman M. and Barda L. 2005. Ḥorbat Anusha, Development Survey in the Area. ESI 117.

Guz-Zilberstein B. 1995. The Typology of the Hellenistic Coarse Ware and Selected Loci of the Hellenistic and Roman Periods. In E. Stern (ed.) Excavations in Dor, Final Report 1B: Areas A and C: The Finds (Qedem 2). Jerusalem. Pp. 289–433.

Hammer E. 2014. Local Landscape Organization of Mobile Pastoralists in Southeastern Turkey. Journal of Anthropological Archaeology 35: 269–288.

Hadas A. 2007. Vine and Wine in the Archaeology of Eretz Israel. Tel Aviv (Hebrew).

Har-El M. and Nir D. 1995. The Geography of Eretz Israel. Tel Aviv (Hebrew).

Haddad E. 2007. A Burial Cave from the First–Second Centuries CE and Double Arcosolia Tombs from the Fourth–Fifth Centuries CE on the Fringes of Ḥorbat Zikhrin. ʿAtiqot 56: 45–57 (Hebrew; English summary, pp. 74*–75*).

Harpaz-Segal A. 2012. The Eretz-Israel Farmhouse in the Hellenistic to the Early Muslim Period: Aspects of Daily Life. Ph.D. dissertation, Bar-Ilan University. RamatGan (Hebrew). Hartman M. 1987. Maavar Afek: A Background for the Teacher and Guide. Tel Aviv (Hebrew).

Haddad E. 2011. Rosh Ha-ʿAyin, Mizpe Afeq: Remains of an Agricultural Settlement from the Iron Age to the Hellenistic Period. ʿAtiqot 65: 53–63 (Hebrew; English summary, p. 67*).

Hayes J. 1972. Late Roman Pottery. London: The British School at Rome.

Haddad E., Tendler A., Shadman A., Torge H. and Itach G. 2015. An Administrative Building from The Persian Period East of Rosh Ha-ʿAyin. IEJ 65: 50–68.

Hestrin R. and Yeivin Z. 1977. Oil from the Presses of Tirat Yehuda. Biblical Archaeologist 40: 29–31. Hingley R. 1990. Rural Settlement in Roman Britain. London.

Haiman M. 1986. Archaeological Survey of Israel. Map of Har Hamran – Southwest (198). Jerusalem (Hebrew).

Hirschfeld Y. 1981. Ancient Winepresses in the Area of Ayalon Park. Eretz Israel 15: 383–390 (Hebrew).

Haiman M. 1995. Agriculture and Nomad-State Relations in the Negev Desert in the Byzantine and Early Islamic Period. BASOR 279: 5–29.

Hirschfeld Y. 1990. List of the Byzantine Monasteries in the Judean Desert. In: G.C. Bottini, L. Di Segni and E. Alliata (eds.). Christian Archaeology in the Holy Land: New Discoveries: Essays in Honour of Virgilio C. Corbo (SBF 36). Jerusalem. Pp. 1–90.

Haiman M. 1998. Rosh Ha-ʿAyin Area, Survey. ESI 20: 45*–46*. 231

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods Hirschfeld Y. 1997a. Farms and Villages in Byzantine Palestine. Dumbarton Oaks Papers 51:33–71.

Kaplan J. 1985. The Mausoleum at Mazor. Eretz Israel 18: 408–418 (Hebrew).

Hirschfeld Y. 1997b. Villages, Manor Houses and Farmsteads in Byzantine Palestine. In S. Dar and Z. Safrai (eds.). The Village in Ancient Israel. Tel Aviv. Pp. 75–129 (Hebrew).

Karmon Y. 1960. An Analysis of Jacotin’s Map of Palestine. IEJ 10: 155–173. Karmon Y. 1984. Geography and Archaeology in the Study of Eretz Israel’s Past. In R. Zeevi (ed.). Israel – People and Land. Tel Aviv. Pp. 147–156 (Hebrew).

Hirschfeld Y. 2000 (ed.). Ramat Hanadiv Excavations: Final Report of the 1984–1998 Seasons. Jerusalem.

Kasher A. 1988. Canaan, Philistia, Greece and Israel: Relations of the Jews in Eretz-Israel with the Hellenistic Cities (332 BCE–70 CE). Jerusalem (Hebrew).

Hirschfeld Y. 2002a. Deir Qalʿa and the Monasteries of Western Samaria. In J.H. Humphrey (ed.). The Roman and Byzantine Near East 3 (JRA Supplementary Series 49). Portsmouth, RI. Pp. 155–189.

Katsenelson J. 1966. The Climate of Eretz Israel and Its Surroundings. In H. Halperin (ed.). The Encyclopedia of Agriculture I. Tel Aviv. Pp. 27–62 (Hebrew).

Hirschfeld Y. 2002b. Desert of the Holy City: The Judean Desert Monasteries in the Byzantine Period. Jerusalem (Hebrew).

Kedar B.Z. 1985. The Arab Conquest and Agriculture: A Seventh-Century Apocalypse, Satellite Imagery, and Palynology. Asian and African Studies 19: 1–15.

Hoade E. 1978. Guide to the Holy Land. Jerusalem. Ilan S. 1984. The Traditional Arab Agriculture. Kardom 34: 22–29 (Hebrew).

Kennedy H. 1994. Crusader Castles. Cambridge. Kennedy H. 2006. The Byzantine and Early Islamic Near East. Aldershot.

Isaac B. and Roll I. 1982. Roman Roads in Judea I: The Legio–Scythopolis Road (BAR International Series 141). Oxford. Ish-Shalom M 1965. Christian Travels in the Holy Land. Descriptions and Sources on the History of the Jews in Palestine. Tel Aviv (Hebrew).

Kennedy H. and Andrews S.T. 1985. The Last Century of Byzantine Syria: A Reinterpretation (Byzantine Forschungen 10). Amsterdam. Pp. 141–183. Kenyon K.M. 1957a. Digging up Jericho. London.

Isser A. and Govrin Y. 1992. Climate Change and Desertification of the Negev at the End of the Byzantine Period. Cathedra 61: 67–83 (Hebrew).

Kenyon K.M. 1957b. Pottery: Hellenistic and Later. In J.W. Crowfoot, G.M. Crowfoot and K.M. Kenyon (eds.). The Objects from Samaria (Samaria-Sebaste 3). London. Pp. 217–235.

Jones A.H.M. 1971. The Greek City: From Alexander to Justinian. Oxford.

Khalidi W. (ed.) 1992. All that Remains: The Palestinian Villages Occupied and Depopulated by Israel in 1948. Washington, D.C..

Kallai Z. 1960. The Northern Boundaries of Judah: From the Settlement of the Tribes until the Beginning of the Hasmonean Period. Jerusalem (Hebrew).

Klein E. 2011. Aspects of the Material Culture of Rural Judea during the Late Roman Period (135–324 CE). Ph.D. dissertation, Bar-Ilan University. Ramat-Gan (Hebrew).

Kalbanner L. 1999. The Climate of Israel in Light of its Causes. Ramat Gan. Kapitaikin L.A. 2006. The Pottery from the IAA Excavations at Tel Mikhal (Tel Michal). ‘Atiqot 52: 21–56.

Kloner A. and Zissu B. 2003. The Necropolis of Jerusalem in the Second Temple Period. Jerusalem (Hebrew).

Kaplan J. 1967. The Yannai Line. In S. Applebaum (ed.). Roman Frontier Studies, 1967: The Proceedings of the Seventh International Congress Held at Tel Aviv. Tel Aviv. Pp. 201–205.

Kochavi M. 1972. Judea. In M. Kochavi (ed.). Judaea, Samaria and the Golan: Archaeological Survey 1967–1968. Jerusalem. Pp. 19–89 (Hebrew). Kochavi M. 1977. An Ostracon of the Period of the Judges from ʿIzbat Sartah. Tel Aviv 4: 1–13.

Kaplan J. 1983. Mazor, the Mausoleum. ESI 2: 66–67.

232

Amit Shadman: Bibliography Kochavi M. 1989. Aphek-Antipatris: Five Thousand Years of History. Tel Aviv (Hebrew).

Lev-Yadun S. 1997. Flora and Climate in Southern Samaria: Past and Present. In I. Finkelstein and Z. Lederman (eds.). Highland of Many Cultures: The Southern Samaria Survey. The Sites (Tel Aviv Monograph Series 14). Tel Aviv. Pp. 85–102.

Kochavi M. and Beit-Arieh Y. 1994. Archaeological Survey of Israel. Map of Rosh Ha-ʿAyin (78). Jerusalem (Hebrew).

Levine L.I. 1984. Roman Rule in Palestine at the End of the Second Temple Period (63 B.C.E.–74 C.E.). In M. Stern (ed.). The History of the Land of Israel IV: The Roman-Byzantine Period. Jerusalem. Pp. 9–247 (Hebrew).

Kogan-Zehavi E. 2014. The Rural Settlement in the Judean Foothills in the Persian and Early Hellenistic Periods, in Light of the Excavations in Ramat Bet Shemesh. In G. Stiebel, O. Peleg-Barkat, D. Ben-Ami and Y. Gadot (eds.). New Studies in the Archaeology of Jerusalem and Its Region 8. Jerusalem. Pp. 120–133 (Hebrew).

Levy-Rubin M. 2002. The Samaritans during the Early Muslim Period According to the Continuatio to the Chronicle of Abu ʿL-Fath. In E. Stern and H. Eshel (eds.). The Samaritans. Jerusalem. Pp. 562–586 (Hebrew).

Kolska-Horwitz L. 2011. The Natural Landscape of Ramat Bet Shemesh. In Y. Dagan (ed.) The Ramat Bet Shemesh Regional Project: Landscape of Settlement from the Paleolithic to the Ottoman Periods (IAA Reports 47). Jerusalem. Pp. 180–198.

Limor O. 1998. Holy Land Travels: Christian Pilgrims in Late Antiquity. Jerusalem (Hebrew).

Korenfeld I. 2014. Development Surveys in the Backdrop of Fifty Years of the Archaeological Survey of Israel. Devar Baʿavar – Newsletter of the Israel Antiquities Authority 21: 24–26 (Hebrew).

Liphschitz N. 1984. Dendroarchaeological Studies – Horbat Bureq (report no. 18). Tel Aviv (Hebrew). Liphschitz N. 1986–7. Paleobotanical Remains from Mount Ebal. Tel Aviv 13–14: 190–191.

Kraemer C.J. 1958. Excavations at Nessana (Auja Hafir, Palestine). Vol. 3: Non-Literary Papyri. Princeton.

Liphschitz N. 1989. Dendroarchaeological Studies – Nahal Qanah Cave (report no. 184). Tel Aviv (Hebrew).

Krauss S. 1924. The Antiquity of the Talmud I. Berlin and Vienna (Hebrew).

Liphschitz N. 1991. The History of Vegetation in Samaria. In Z.H. Erlich (ed.). Shomron and Benjamin II. Ariel. Pp. 27–33 (Hebrew).

Landau Y.H. 1966. A Greek Inscription Found Near Hefzibah. IEJ 16: 54–70.

Liphschitz N. and Waisel Y. 1980. Dendroarchaeological Studies – Karnei Shomron (report no. 85). Tel Aviv (Hebrew).

Lange S. 1965. Architettura della cruciate in Palestina. Como. Langgut D. Finkelstein I. and Litt T. 2013. Climate and the Late Bronze Collapse: New Evidence from the Southern Levant. Tel Aviv 40: 149–175.

Lipschits O. 2004. The Fall and Rise of Jerusalem. Jerusalem (Hebrew). Loewenstamm S.E. 1950. Aphek. In E.Z. Sukenik (ed.). Encyclopedia Biblical I. Jerusalem. Pp. 501–503 (Hebrew).

Lapp N.L. (ed.) 1981. The Third campaign at Tel el-Ful: The Excavations of 1964 (AASOR 45). Boston. Lavi S. 1976. The Olive. In S. Homsky, R. Karschon, Schiller G. and Swirski E. (eds.). The Encyclopedia of Agriculture III. Tel Aviv. Pp. 201–208 (Hebrew).

London A. 2009. The Cultivation of Fruit Trees during the Roman and Byzantine Periods Based on Rabbinical Sources and Classical Literature. Ph.D. dissertation, Bar-Ilan University. Ramat-Gan (Hebrew).

Leibner U. 2004. History of Settlement in the Eastern Galilee during the Hellenistic, Roman and Byzantine Periods in Light of an Archaeological Survey. Ph.D. dissertation, Bar-Ilan University. Ramat-Gan (Hebrew).

MacDonald B. 1988. The Wadi el-Hasa Archaeological Survey 1979–1983, West-Central Jordan. Waterloo. Magen Y. 2002a. The Samaritans in the RomanByzantine Period. In E. Stern and H. Eshel (eds.). The Samaritans. Jerusalem. Pp. 213–244 (Hebrew).

Leibner U. 2009. Settlement and History in Hellenistic, Roman and Byzantine Galilee: An Archaeological Survey of the Eastern Galilee (Texts and Studies in Ancient Judaism 127). Tübingen. 233

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods Magen Y. 2002b. The Areas of Samaritan Settlement in the Roman-Byzantine Period. In E. Stern and H. Eshel (eds.). The Samaritans. Jerusalem. Pp. 245–271 (Hebrew).

and the Galilee Text and Context in the Greco- Roman and Byzantine Periods (South Florida Studies in the History of Judaism 143). Atlanta. Melamed E.Z. 1950. The Onomasticon of Eusebius. Jerusalem (Hebrew).

Magen Y. 2002c. The Stone Vessel Industry in the Second Temple Period. Jerusalem.

Mills P. J. and Reynolds P. 2014 Amphorae and Specialized Coarsewares of Ras al Bassit, Syria: Local Products and Exports. In N. Poulou-Papadimitriou, E. Nodarou and V. Kilikoglou (Eds.) LRCW4 Late Roman Coarse Wares, Cooking Wares and Amphorae in the Mediterranean: Archaeology and Archaeometry: The Mediterranean: A Market without Frontiers. BAR International Series 2616 (1).

Magen Y. 2008a. Late Roman Fortresses and Towers in Southern Samaria and Northern Judea. In Y. Magen (ed.). Judea and Samaria Researches and Discoveries (JSP 6). Jerusalem. Pp. 177–246. Magen Y. 2008b. Oil Production in the Land of Israel in the Early Islamic Period. In Y. Magen (ed.). Judea and Samaria Researches and Discoveries (JSP 6). Jerusalem. Pp. 275–278.

Naʾaman N. 1990. Population Changes in Palestine Following Assyrian Deportations. Cathedra 54: 43–62 (Hebrew).

Magen Y. 2012. A Roman Fortress and Byzantine Monastery at Khirbet Deir Samʿan. In Y. Magen (ed.). Churches and Monasteries in Samaria and Northern Judea (JSP 15). Jerusalem. Pp. 9–106.

Naʾaman N. 2009. Did the City of Dor Serve as the Capital of an Assyrian Province? Eretz Israel 29: 310– 317 (Hebrew).

Magen Y. and Aizik N. 2012. A Late Roman Fortress and Byzantine Monastery at Deir Qalʿa. In Y. Magen (ed.). Churches and Monasteries in Samaria and Northern Judea (JSP 15). Jerusalem. Pp. 107–156.

Naʾaman N. and Zadok R. 2000. Assyrian Deportations to the Province of Samaria in the Light of Two Cuneiform Tablets from Tel Hadid. Tel Aviv 27: 159– 188.

Magen Y. and Finkelstein I. 1993. Archaeological Survey of the Hill Country of Benjamin. Jerusalem (Hebrew).

Neeman Y. 1995. Or ʿAqiva – Byzantine Road. ESI 103: 47–48 (Hebrew).

Magen Y. and Kagan E. 2012. Corpus of Christian Sites in Samaria and Northern Judea (Christians and Christianity 1 and JSP 13). Jerusalem.

Negev A. 1983. Masters of the Desert: The Story of the Nabateans. Jerusalem (Hebrew).

Magness J. 1993. Jerusalem Ceramic Chronology. Sheffield.

Nevo Y.D. 1991. Pagans and Herders: A Re-Examination of the Negev Runoff Cultivation System in the Byzantine and Early Arab Periods. Midreshet Ben-Gurion.

Majcherek G. 1995. Gazan Amphorae: Typology Reconsidered. In H. Meyza and J. Mlynarczy eds. Hellenistic and Roman Pottery in the Eastern Mediterranean: Advances in Scientific Studies: Acts of the II Nieborow Pottery Workshop, Nieborow 20 December 1993. Warsaw. Pp. 163-178.

Nir D. 1989. The Geomorphology of Eretz Israel. Jerusalem (Hebrew). Parpola S. 1997. State Archives of Assyria Bulletin 1. Helsinki.

Mattingly D.J. 1988. Megalithic Madness and Measurement. Or How Many Olives Could an Olive Press Press? Oxford Journal of Archaeology 7: 177–195.

Pastor J. 1997. Land and Economy in Ancient Palestine. London.

Mazar A. and Wachtel I. 2015. Hurvat ʿEres: A FourthCentury BCE Fortress West of Jerusalem. IEJ 65: 214–244.

Patrich J. 1995. The Judean Desert Monasticism in the Byzantine Period: The Institutions of Sabas and His Disciples. Jerusalem (Hebrew).

Mazar B., Dothan T. and Dunayevsky I. 1966. En-Gedi Excavations in 1961–1962 ‘Atiqot 5. Jerusalem.

Paz Y., Mizrahi S. and Grossman L. 2015. Ancient Fields: An EBIII Case Study from Tel Yarmuth. In G. Stiebel, O. Peleg-Barkat, D. Ben-Ami and Y. Gadot (eds.). New Studies in the Archaeology of Jerusalem and Its Region 9. Jerusalem. Pp. 91–98 (Hebrew).

McCollough C.T. and Edwards D.R. 1997. Transportation of Space: The Roman Road at Sepphoris. In D.R. Edwards and C.T. McMcollough (eds.). Archaeology 234

Amit Shadman: Bibliography Peleg Y. and Feller Y. 2004a. Betar ʿIllit (West). ESI 116.

Rappaport U. 2013. The House of the Hasmoneans: The People of Israel in the Land of Israel in the Hasmonean Period. Jerusalem (Hebrew).

Peleg Y. and Feller Y. 2004b. Shimʿa, Industrial Zone. ESI 116.

Ravikovitch S. 1981. The Soils of Israel: Formation, Nature and Properties. Tel Aviv (Hebrew).

Percival J. 1940. The Villa Economy: Problems and Perspectives. In K. Branigan and D. Miles (eds.). The Economies of Romano-British Villas. Sheffield.

Raviv D. 2013. Magnificent Tombs from the Second Temple Period in Western Samaria – New Insights. In A. Tavger, Z. Amar and M. Billig (eds.). In the Highland’s Depth 3. Ariel. Pp. 109–142 (Hebrew).

Pinthus M. and Amir Y. 1994. Wheat. In Y. Arnon (ed.). The Encyclopedia of Agriculture IV. Tel Aviv. Pp. 305– 338 (Hebrew).

Raviv D. and Elinson N. 2014. The Samaria Travel Guide II. Shiloh (Hebrew).

Porath Y. 1989. Hydraulic Plaster in Aqueducts as a Chronological Indicator. In D. Amit, Y. Hirschfeld and J. Patrich (eds.). The Aqueducts of Israel. Jerusalem. Pp. 69–76 (Hebrew).

Reich R. 1992. The Beth-Zur Citadel II: A Persian Residency? Tel Aviv 19: 113–123.

Porath Y. 2002. Hydraulic Plaster in Aqueducts as a Chronological Indicator. In: D. Amit, J. Patrich and Y. Hirschfeld (eds.). The Aqueducts of Israel (JRA 46). Portsmouth.

Reich R. and Shukron E. 2003. The Urban Development of Jerusalem in the Late Eight Century B.C.E. In A.G. Vaughn and A.E. Killebrew (eds.). Jerusalem in Bible and Archaeology: The First Temple Period (Society of Biblical Literature Symposium Series 18). Atlanta. Pp. 209–218.

Portugali Y. 1982. A Field Methodology for Regional Archaeology (The Jezreel Valley Survey 1981). Tel Aviv 9: 170–188.

Reifenberg A. 1950. Milhemet haMizra vehaYeshimon. Jerusalem (Hebrew).

Prawer J. 1963. A History of the Latin Kingdom of Jerusalem (2 vols.). Jerusalem (Hebrew).

Rey E.G. 1883. Les colonies Franques de Syria aux XII et XIIIème siècles. Paris.

Prawer J. 1981. Preface. In J. Prawer and H. Ben-Shammai (eds.). The History of Jerusalem: Crusaders and Ayyubids (1099–1250). Jerusalem. Pp. 11–14 (Hebrew).

Riley J.A. 1975. The pottery from the first session of excavations in the Caesarea hippodrome. BASOR 218.

Press Y. 1951–1955. A Topographical-Historical Encylopaedia of Palestine (4 vols.). Jerusalem (Hebrew).

Robinson E. and Smith E. 1856. Biblical Researches in Palestine and Adjacent Regions: A Journal of Travels in the Years 1838 & 1852 (3 vols.). London.

Pringle D. 1994. Towers in Crusader Palestine. Château Gaillard 16: 337–350.

Richardson J. 1976. Roman Provincial Administration: 227 BC to AD 117. London.

Pringle D. 1997. Secular Building in the Crusader Kingdom of Jerusalem: An Archaeological Gazetteer. Cambridge.

Roll I. 1976. The Roman Road Network in Eretz-Israel. Qadmoniot 9: 38–50 (Israel).

Rahmani L.Y. 1970. The Coins from Tirat Yehuda. ʿAtiqot 6: 68–69 (Hebrew).

Roll I. 1996. The Types of Roads in Eretz Israel during the Roman-Byzantine Period. In G. Barkai and E. Schiller (eds.). 120th Volume of Ariel: A Selection of Essays on the Land of Israel. Jerusalem. Pp. 151–156 (Hebrew).

Rappaport U. 1981. The Hasmonean State (160–37 B.C.E.). In M. Stern (ed.). The History of Eretz Israel: The Hellenistic Period and the Hasmonean State (332–37 B.C.E.). Jerusalem. Pp. 191–273 (Hebrew). Rappaport U. 1995. The Hasmonean State from Jonathan Apphus to Antigones Mattathias. In D. Amit and H. Eshel (eds.). The Hasmonean Era. Jerusalem. Pp. 61–76 (Hebrew).

Roll I. 2000. Ancient Roads in Eretz Israel and Their Expression in Aerial Photographs. In B.Z. Kedar and A. Danin (eds.). Remote Sensing: The Use of Aerial Photographs and Satellite Images in Israel Studies. Jerusalem. Pp. 131–141 (Hebrew).

235

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods Roll I. and Ayalon E. 1983. Roman Roads in Western Samaria and Northern Judea. Israel – People and Land 1: 131–146 (Hebrew).

Safra D. 1983. Water Pollution and Purification in the Lower Yarkon Basin. In D. Grossman (ed.). Between the Yarkon and Ayalon. Ramat-Gan. Pp. 35–44 (Hebrew).

Roll I. and Ayalon E. 1986a. Paving Methods and Routes of the Roman Roads in Western Samaria. Nofim 18– 19: 31–41 (Hebrew).

Safrai S. 1965. Pilgrimage to Jerusalem in the Second Temple Period. Tel Aviv (Hebrew).

Roll I. and Ayalon E. 1986b. Roman Roads in Western Samaria. PEQ 118: 113–134.

Safrai Z. 1980a. A Roman Road from Lydda to Jerusalem. Nofim 55–64 (Hebrew).

Roll I. and Ayalon E. 1987. Highways and Roads in the Sharon Plain during the Roman and Byzantine Periods. Israel – People and Land 4: 147–162 (Hebrew).

Safrai Z. 1980b. Borders and Government in Eretz Israel during the Mishnaic and Talmudic Periods. Tel Aviv (Hebrew).

Roll I. and Ayalon E. 1989. Apollonia and Southern Sharon: Model of a Coastal City and Its Hinterland. Tel Aviv (Hebrew).

Safrai Z. 1981. The Conquest of Jerusalem in 68–69 C.E. In A. Oppenheimer, U. Rappaport and M. Stern (eds.). Jerusalem in the Second Temple Period: Abraham Schalit Memorial Volume. Jerusalem. Pp. 320–339 (Hebrew).

Ron Z. 1977a. Stone Huts as an Expression of Terraced Agriculture in the Judean and Samarian Hills. Ph.D. dissertation, Tel Aviv University (Hebrew).

Safrai Z. 1983a. Family Structure during the Period of the Mishnah and the Talmud. Milet 1: 129–156 (Hebrew).

Ron Z. 1977b. The Distribution of Terrace Agriculture in the Jerusalem Hills. In S. Shmueli, D. Grossman and R. Zeevi (eds.). Judea and Samaria: Studies in Settlement Geography I. Jerusalem. Pp. 210–229 (Hebrew).

Safrai Z. 1983b. The Question of the Spatial Structure of the Settlement in the Galilee in the Times of the Mishna and the Talmud. In A. Shmueli, A. Sofer and N. Kliot (eds.). The Land of Galilee I. Haifa. Pp. 269–288 (Hebrew).

Rosenthal-Heginbottom R. and Sivan R. 1978. Ancient Lamps in the Schloessinger Collection (Qedem 8). Jerusalem.

Safrai Z. 1986a. The Influence of Demographic Stratification on the Agricultural and Economic Structure during the Mishnaic and Talmudic Period. In A. Kasher, A. Oppenheimer and U. Rappaport (eds.). Man and Land in Eretz Israel in Antiquity. Jerusalem. Pp. 20–48 (Hebrew).

Roth I. and Flexer A. 1977. Minerals of Judea and Samaria and Their Exploitation. In S. Shmueli, D. Grossman and R. Zeevi (eds.). Judea and Samaria: Studies in Settlement Geography I. Jerusalem. Pp. 3–13 (Hebrew).

Safrai Z. 1986b. Shechem in the Days of the Mishnah and the Talmud, 63 BCE–637 CE. In S. Dar and Z. Safrai (eds.). Shomron Studies. Tel Aviv. Pp. 83–126 (Hebrew).

Rubin R. 1990. The Negev as a Settled Land: Urbanization and Settlement in the Desert in the Byzantine Period. Jerusalem (Hebrew). Rubin R. 1998. The Roman-Byzantine Empire and Its Desert Frontiers: The Negev vs. Tripolitania – A Comparative Study. Cathedra 89: 63–82 (Hebrew).

Safrai Z. 1987. The Economic Implication of Olive Production in the Mishnah and Talmud Periods. In M. Heltzer and D. Eitam (eds.). Olive Oil in Antiquity: Israel and Neighbouring Countries from Neolith to Early Arab Period, Conference 1987. Haifa. 176–182.

Rupp D.W. 1983. Archaeology Survey: Definitions. In D.R. Keller and D.W. Rupp (eds). Archaeological Survey in the Mediterranean Area (BAR International Series 155). Oxford. Pp. 17–30.

Safrai Z. 1990a. The Administration of the Rural Road System in the Mishnaic and Talmud Period. In B.Z. Kedar, T. Dothan and S. Safrai (eds.). Commerce in Palestine throughout the Ages. Jerusalem. Pp. 159–180 (Hebrew).

Russell J. 2001. The Persian Invasions of Syria/ Palestine and Asia Minor in the Reign of Heraclius: Archaeological, Numismatic and Epigraphic Evidence. In E. Kountoura-Galake (ed.). The Dark Centuries of Byzantium (7th–9th C.) (National Hellenic Research Foundation, Institute for Byzantine Research, International Symposium 9). Athens. Pp. 43–71.

Safrai Z. 1990b. Commerce in Eretz-Israel in the Roman Period. In B.Z. Kedar, T. Dothan and S. Safrai (eds.). Commerce in Palestine throughout the Ages. Jerusalem. Pp. 108–139 (Hebrew). 236

Amit Shadman: Bibliography Safrai Z. 1994a. The Onomasticon of Eusebius and the Jewish Population of the Southern Hebron Hills. Cathedra 72: 23–33 (Hebrew).

Safrai Z. and Sion O. 2007. Nomad Settlement in Palestine During the Late Byzantine–Early Moslem Period. In D.R. Edwards (ed.). The Archaeology of Difference: Gender, Ethnicity, Class and the “Other” in Antiquity. Studies in Honor of Eric M. Meyers (AASOR 60/61). Boston. Pp. 397–411.

Safrai Z. 1994b. The Economy of Roman Palestine. London. Safrai Z. 1995a. The Jewish Community in Eretz Israel in the Time of the Mishnah and Talmud. Jerusalem (Hebrew).

Sasson A. 2014. The Hebrew Stone Industry at the Foothills in Modern Times – Migdal Tzedek as a Center of the Stone and Quicklime Industry. In A. Tavger, Z. Amar and M. Billig (eds.). In the Highland’s Depth 4. Ariel. Pp. 159–173 (Hebrew).

Safrai Z. 1995b. The Shephela of Samaria in the Byzantine Period. In Z.H. Erlich and Y. Eshel (eds.). Judea and Samaria Research Studies – Proceedings of the Fourth Annual Meeting. Ariel. Pp. 189–208 (Hebrew).

Segal A., Kletter R. and Ziffer I. 2006. A Persian-Period Building from Tel Yaʿoz (Tell Ghaza). ʿAtiqot 52: 1*– 24* (Hebrew; English summary, p. 203).

Safrai Z. 1995c. The Lod Shephelah in the Byzantine Period. In R. Bar-Ner and A. Buch (eds.). The Yarkon Basin: Monuments and Routes. Tel Aviv. (Hebrew).

Shadman A. 2008. Naḥal Narbeta. ESI 120.

Safrai Z. 1997a. Settlements in the Shephela of Samaria during the Byzantine Period. In Y. Eshel (ed.). Judea and Samaria Research Studies – Proceedings of the Sixth Annual Meeting. Ariel. Pp. 217–234 (Hebrew).

Shadman A. 2014. Rosh Ha-ʿAyin (South). ESI 126. Shadman A., Tendler A. and Marcus J. 2015. A Farmhouse of the Iron Age II and the Persian and Hellenistic Periods at Rosh Ha-ʿAyin. Qadmoniot 150: 84–88 (Hebrew).

Safrai Z. 1997b. The Village in Judea. In S. Dar and Z. Safrai (eds.). The Village in Ancient Israel. Tel Aviv. Pp. 11–73 (Hebrew).

Shai I. and Uziel J. 2014. Addressing Survey Methodology in the Southern Levant: Applying Different Methods for the Survey of Tel Burna, Israel. IEJ 64: 172–190.

Safrai Z. 1997c. The Size of the Population in the Land of Israel in the Roman-Byzantine Period. In Y. Friedman, Z. Safrai and J. Schwartz (eds.). Hikrei Eretz: Studies in the History of the Land of Israel Dedicated to Prof. Yehuda Feliks. Ramat-Gan. Pp. 277– 306 (Hebrew).

Shiloh Y. 1980. The Population of Iron Age Palestine in the Light of Sample Analysis of Urban Plans, Areas and Population Density. BASOR 239: 25–35.

Safrai Z. 1998. Ancient Field Structures – The Villages in Eretz-Israel during the Roman Period. Cathedra 89: 7–40 (Hebrew).

Simons J.J. 1937. Handbook for the Study of Egyptian Topographical Lists Relating to Western Asia. Leiden. Singer-Avitz L. 1989. Local Pottery of the Persian Period (Tel Michal): Strata XI–VI. In Z. Herzog, G. Rapp and O. Negbi (eds.). Excavations at Tel Michal, Israel (Tel Aviv Monograph Series 8). Minneapolis and Tel Aviv. Pp. 115–144.

Safrai Z. and Dar S. 1997a. Introduction. In S. Dar and Z. Safrai (eds.). The Village in Ancient Israel. Tel Aviv. Pp. 7–10 (Hebrew). Safrai Z. and Dar S. 1997b. Horbat Bira – An Estate in the Lod Shephelah. In Y. Friedman, Z. Safrai and J. Schwartz (eds.). Hikrei Eretz: Studies in the History of the Land of Israel Dedicated to Prof. Yehuda Feliks. Ramat-Gan. Pp. 57–107 (Hebrew).

Sion O. 2001. Settlement History in the Central Samarian Region in the Byzantine Period. Ph.D. dissertation, The Hebrew University of Jerusalem (Hebrew). Sion O. 2004. The Agrarian Situation in Central Samaria and Its Effect on the Samaritan Revolts during the Byzantine Period. In Y. Eshel (ed.). Judea and Samaria Research Studies – Proceedings of the Thirteenth Annual Meeting. Ariel. Pp. 189–196 (Hebrew).

Safrai Z. and Fixler Y. 1993. Horbat Kesfa – A Village and Its Land. In Z.H. Erlich (ed.). Shomron and Benjamin III. Ariel. Pp. 204–223 (Hebrew). Safrai Z., Grossman D., Frenkel J. and Rainey A.F. 1990. The Roads of the Sharon. In A. Degani, D. Grossman and S. Shmueli (eds.). Hasharon Between Yarkon and Carmel. Tel Aviv. Pp. 249–261 (Hebrew).

Sion O., ʿAd U., Haiman M. and Parnos G. 2007. Excavations and Surveys at Ḥorbat Anush and Ḥorbat Leved in the Samarian Shephelah. ʿAtiqot 55: 109–159 (Hebrew; English summary, pp. 62–64). 237

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods Sion O., Haiman M. and Artzi B. 2008. Ẓur Natan. ESI 120.

Tal O. 2006. The Archaeology of Hellenistic Palestine: Between Tradition and Renewal. Jerusalem (Hebrew).

Sion D. and Hartal M. 2003. A New Tetrarchic BoundaryStone from the Nothern Hula Valley. Scripta Classica Israelica 22. Pp. 233–239.

Tal O. and Taxel I. 2015. Samaritan Cemeteries and Tombs in the Central Coastal Plain. Archaeology and History of the Samaritan Settlement outside Samaria (ca. 300–700 CE) (Aegypten und Altes Testament 82). Munster.

Smithline H. 2009. Pottery and a Small Find from the Persian and Hellenistic Periods. In N. Getzov, R. Lieberman-Wander, H. Smithline and D. Syon (eds.). Horvat ʿUẓa: The 1991 Excavation I: The Early Periods (IAA Reports 41). Jerusalem. Pp. 136–149.

Taxel I. 2005. The Transition between the Byzantine and the Early Islamic periods (the 7th century CE) as seen through rural settlement: Horbat Zikrin as a case study. M.A. thesis, Tel Aviv University (Hebrew).

Smithline H. 2013. A Unique Hellenistic Pottery Assemblage from ʿAkko. ‘Atiqot 76: 71–103.

Taxel I. 2006. Mazor (Elʿad). ESI 118.

Sneh B. and Hochberg N. 1976. Grapevine. In S. Homsky, R. Karschon, Schiller G. and Swirski E. (eds.). The Encyclopedia of Agriculture III. Tel Aviv. Pp. 172–195 (Hebrew).

Taxel I. 2010. Aspects of Material Culture of the Rural Settlement in the Province of Palaestina Prima in the Fifth–Seventh Centuries CE. Ph.D. dissertation, Tel Aviv University (Hebrew).

Spanier Y. and Sasson A. 2001. Lime Kilns in the Land of Israel: Seminar in Memory of Shmuel Avitzur. Tel Aviv (Hebrew).

Taxel I. 2013. Identifying Social Hierarchy through House Planning in the Villages of the Late Antique Palestine: The Case of Horvat Zikhrin. Antiquité Tardive 21: 149–166.

Stern E. 1978. Excavations at Tel Mevorakh 1973–1976: From the Iron Age to the Roman Period (Qedem 9). Jerusalem.

Tcherikover A. 1961. The Land of Israel in Light of the Zenon Papyri. In The Jews in the Greco-Roman World. Tel Aviv. Pp. 33–82 (Hebrew).

Stern M. 1981. The Land of Israel in the Hellenistic Period (332–160 B.C.E.) In M. Stern (ed.). The History of Eretz Israel: The Hellenistic Period and the Hasmonean State (332–37 B.C.E.). Jerusalem. Pp. 191–273 (Hebrew).

Tendler A. 2013a. Rosh Ha-ʿAyin, Qurnat Haramiya. ESI 125.

Stern E. 1982. The Material Culture of the Land of the Bible in the Persian Period 538–332 B.C.E. Warminster.

Tendler A. 2013b. Rosh Ha-ʿAyin (Qurnat Haramiya East). ESI 125.

Stern E. 1995. Local Pottery of the Persian Period. In E. Stern (ed.). Excavations at Dor: Final Report 1B: Areas A and C: The Finds (Qedem 2). Jerusalem. Pp. 51–92.

Tendler A. 2014. Ḥorbat Migdal Afeq. ESI 126. Tendler A. 2015. Rosh Ha-ʿAyin, Nahal Susi. ESI 127.

Stern E. 2001. Archaeology of the Land of the Bible II: The Assyrian, Babylonian and Persian period (732–332 B.C.E.). New York.

Tendler A. and Shadman A. 2015. An Administrative Building from the Late Persian Period in the Foothills of Western Samaria. In A. Tavger, Z. Amar and M. Billig (eds.). In the Highland’s Depth 5. Ariel. Pp. 193–207 (Hebrew).

Stroumsa R. 2008. People and Identities in Nessana. Durham. Sussman V. 2017. Late Roman to Late Byzantine / Early Islamic Period Lamps in the Holy Land: The Collection of the Israel Antiquities Authority. Oxford: Archaeopress.

Tepper Y. 1986. The Rise and Fall of Dove-Raising. In A. Kasher, A. Oppenheimer and U. Rappaport (eds.). Man and Land in Eretz Israel in Antiquity. Jerusalem. Pp. 170–196 (Hebrew).

Taaffe E.J. and Gauthier H.L. 1973. Geography of Transportation. New York.

Tepper Y. 1997. Stables in the Land of Israel in the Roman–Byzantine Periods. In S. Dar and Z. Safrai (eds.). The Village in Ancient Israel. Tel Aviv. Pp. 229– 274 (Hebrew).

Tal O. 1999. The Persian Period. In I. Roll and O. Tal (eds.). Apollonia-Arsuf I: The Persian and Hellenistic periods. Tel Aviv. Pp. 83–222, 291–296.

Tepper Y. 2009. Midrakh ʿOz, Khirbat el-Khishash, Survey. ESI 121. 238

Amit Shadman: Bibliography Tepper Y. and Tepper Y. 2013. The Roads that Bear the People: Pilgrimage Roads to Jerusalem in the Second Temple Period. Tel Aviv (Hebrew).

Tsuk T. 2003. Water Supply in Byzantine Shivta. Qadmoniot 125: 18–24 (Hebrew). Tsuk T. 2011. Water at the End of the Tunnel. Jerusalem (Hebrew).

Thomsen P. 1917. Die römischen Meilenstein der Provinzen Syria, Arabia und Palastina. ZDPV 40: 1–86.

Tsuk T., Bordowicz I and Taxel I. 2016. Majdal Yābā: The History and Material Culture of a Fortified Village in Late Ottoman- and British Mandate-Palestine. Journal of Islamic Archaeology 3.1: 37–88.

Thompson D.J. 2005. The Ptolemies and Egypt. In A. Erskine (ed.). A Companion to the Hellenistic World. Oxford. Pp. 105–120.

Turtledove H. 1982. The Chronicle of Theophares University of Pensylvenia Press Pensylvenia.

Torge H. 2007. The Foothills of the Yarkon Basin during the Iron Age. M.A. thesis, Tel Aviv University (Hebrew).

Van Beek G.W. 1958. Frankincense and Myrrh in Ancient South Africa. Journal of the American Oriental Society 78: 141–152.

Torge H. 2012a. Pottery from the Persian and Hellenistic Periods at Ḥorbat Burin (East). ʿAtiqot 70: 33–38 (Hebrew; English summary, p. *82).

Vilnay Z. 1977a. Migdal Afeq. In Z. Vilnay (ed.). Ariel Encyclopedia. Tel Aviv. Pp. 4164 (Hebrew).

Torge H. 2012b. Settlement Remains from the Late Hellenistic and Early Roman Periods at Khirbat Burnat (Southwest). ʿAtiqot 69: 1*–68* (Hebrew; English summary, pp. 157–159).

Vilnay Z. 1977b. Migdal Tzedeq. In Z. Vilnay (ed.). Ariel Encyclopedia. Tel Aviv. Pp. 4187–4188 (Hebrew).

Torge H. 2018. Rosh Ha-‘Ayin Qurnat Hasamiya. ESI 130.

De Vincenz A. 2005. The Pottery from Khirbet el-Jiljil (First Season) BAIAS 23: 111-138.

Treacy J.M. and Denevan W.M. 1994. The Creation of Cultivable Land through Terracing. In N.F. Miller and K.K. Gleason (eds.). The Archeology of Garden and Field. Philadelphia. Pp. 91–110.

Waisel Y., Pollak G. and Y. Cohen. 1978. The Ecology of the Vegetation of Israel. Petah Tiqvah (Hebrew). Weiss D., Zissu B. and Solimany G. 2004. Archaeological Survey of Israel. Map of Nes Harim (104). Jerusalem (Hebrew).

Tsafrir Y. 1984. Eretz Israel from the Destruction of the Second Temple to the Muslim Conquest II. Jerusalem (Hebrew).

Wensinck A.J. 1987. Khamr. First Encyclopedia of Islam IV. Leiden. Pp. 894–897.

Tsafrir Y. 1996. Some Notes on the Settlement and Demography of Palestine in the Byzantine Period: The Archaeological Evidence. In J.D. Seger (ed.). Retrieving the Past: Essays on Archaeological Research and Methodology in Honor of Gus W. Van Beek. Winona Lake, IN. Pp. 269–283.

White K.D. 1970. Roman Farming. London. Wilkinson J. 1977. Jerusalem Pilgrims before the Crusades. Jerusalem.

Tsafrir, Y., Di Segni, L. and Green, J. (eds.) 1994. Tabula Imperii Romani: Iudaea Palaestina. Jerusalem.

Yaalon D.H. 1995. Soils. In T. Sofer (ed.). The New Israel Atlas – The National Atlas. Tel Aviv. Pp. 23 (Hebrew).

Tsuk T. 1982. Survey of Western Samaria. Jerusalem (Hebrew).

Yeivin Z. 1987. On the “Medium-Sized City”. Eretz Israel 19: 59–71 (Hebrew).

Tsuk T. 1993. The Jewish Community of Western Samaria in Light of Information from the Burial Caves at Migdal Tzedek. Niqrot Zurim 19: 67–76 (Hebrew; English summary, p. 141).

Yeivin Z. and Edelstein G. 1970. Excavations at Tirat Yehuda. ʿAtiqot 6: 56–67 (Hebrew). Yitach M. and Baruch Y. 2001. Surveyed Byzantine Churches in the Hill Country of Benjamin. In Y. Eshel (ed.). Judea and Samaria Research Studies – Proceedings of the Tenth Annual Meeting. Ariel. Pp. 159–170 (Hebrew).

Tsuk T. 1994. The Water Supply of Horvat Zikhrin, A Byzantine-Period Village in the Samaria Foothills. Israel – People and Land 7–8: 133–148 (Hebrew).

239

The Rural Hinterland of Antipatris from the Hellenistic to the Byzantine Periods Yron-Lubin M. 1997. Ḥorbat Qedumim). ESI 19: 46*–49*.

Ḥermeshit

(Neot

Zertal A. 2012. The Manasseh Hill Country Survey V: The Middle Jordan Valley, from Wadi Fasael to Wadi ʿAujah. Tel Aviv (Hebrew).

Zelinger Y. 2009. The Rural Settlements in the Shephelah of Lod (Lydda) during the Second Temple Period. Ph.D. dissertation, Bar-Ilan University. Ramat-Gan (Hebrew).

Zertal A. and Mirkam N. 2000. The Manasseh Hill Country Survey III: From Nahal ʿIron to Nahal Shechem. Tel Aviv (Hebrew).

Zelinger Y. and Amit D. 2001. Mazor (Elʿad), Sites 17a–c, 34a, 35a, 95a–e, K4, K5. ESI 113: 56*–58*.

Ziegelmann A. 1982. The Roman Road from Caesarea to Haifa (Akko). Nofim 16: 35–46 (Hebrew).

Zertal A. 1992. The Manasseh Hill Country Survey I: The Shechem Syncline. Tel Aviv (Hebrew).

Zilberbod I. and Amit D. 2001. Mazor (Elʿad). ESI 113: 45*–46*.

Zertal A. 1996. The Manasseh Hill Country Survey II: The Eastern Valleys and the Fringes of the Desert. Tel Aviv (Hebrew).

Zissu B. 1996. Two Herodian Columbarium Towers at Khirbet ʿAleq and Khirbet Abu Hof. Qadmoniot 112: 100–106 (Hebrew).

Zertal A. 1999. The Province of Samaria during the Persian and Hellenistic Periods. In Y. Avishur and R. Deutsch (eds.). Michael: Historical, Epigraphical and Biblical Studies in Honor of Prof. Michael Heltzer. Tel Aviv. Pp. 75–98 (Hebrew).

Zissu B. 2001. Rural Settlement in the Judaean Hills and Foothills from the Late Second Temple Period to the Bar Kokhba Revolt. Ph.D. dissertation, Bar-Ilan University. Ramat-Gan (Hebrew). Zohar G. 1991. Settlement Distribution in the Southern Sharon during the Hellenistic, Roman and Byzantine Periods. M.A. thesis, Bar-Ilan University. Ramat-Gan (Hebrew).

Zertal A. 2005. The Manasseh Hill Country Survey IV: From Nahal Bezeq to the Sartaba. Tel Aviv (Hebrew).

240

LIST OF ABBREVIATIONS

AASOR BAIAS BAR BASOR BSOAS ESI IAA IEJ IJES JPOS JRA JSP NEAEHL PEQ Tel Aviv Monograph Series SBF ZDPV

Annual of the American Schools of Oriental Research Bulletin of the Anglo-Israel Archaeological Society Biblical Archaeology Review Bulletin of the American Schools of Oriental Research Bulletin of the School of Oriental and African Studies Excavations and Surveys in Israel Israel Antiquities Authority Israel Exploration Journal Israel Journal of Earth Sciences Journal of the Palestine Oriental Society Journal of Roman Archaeology Judea and Samaria Publications The New Encyclopedia of Archaeological Excavations in the Holy Land Palestine Exploration Quarterly Monograph Series of the Institute of Archaeology of Tel Aviv University Studium Biblicum Franciscanum, Collectio minor Zeitschrift des Deutschen Palästina-Vereins

243