Archaeological Perspectives on Houses and Households in Third Millennium Mesopotamian Society [1 ed.] 9781527516809

Citation preview

Archaeological Perspectives on Houses and Households in Third Millennium Mesopotamian Society

Archaeological Perspectives on Houses and Households in Third Millennium Mesopotamian Society By

Alessandra Salvin

Archaeological Perspectives on Houses and Households in Third Millennium Mesopotamian Society By Alessandra Salvin This book first published 2017 Cambridge Scholars Publishing Lady Stephenson Library, Newcastle upon Tyne, NE6 2PA, UK British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library Copyright © 2017 by Alessandra Salvin All rights for this book reserved. No part of this book may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without the prior permission of the copyright owner. ISBN (10): 1-5275-0338-0 ISBN (13): 978-1-5275-0338-0

To Maria Cristina Fumagalli and Geremia Fumagalli, who loved history

TABLE OF CONTENTS

List of Illustrations ..................................................................................... xi Acknowledgements ................................................................................. xvii Introduction ................................................................................................. 1 Chapter One ................................................................................................. 3 Theoretical Background 1.1 The Archaeological Significance of Houses .................................... 3 1.2 Research Context ............................................................................. 5 1.3 Research Issues and Questions ........................................................ 9 1.3.1 How large were Mesopotamian residential groups? .............. 10 1.3.2 How was space used in Mesopotamian houses?.................... 11 1.3.3 What evidence is there of variation in the wealth of Mesopotamian houses?......................................................... 12 1.3.4 What was private and public for the occupants of Mesopotamian houses?......................................................... 13 1.4 Spatial Studies................................................................................ 14 1.5 House, Household and Society in Upper and Lower Mesopotamia .................................................................................. 17 1.5.1 Lower Mesopotamia .............................................................. 19 1.5.2 Upper Mesopotamia .............................................................. 22 Chapter Two .............................................................................................. 25 Methodology 2.1 A Survey of Some Methodologies in Houses and Household Analysis .......................................................................................... 26 2.1.1 Typology ............................................................................... 26 2.1.2 Features ................................................................................. 27 2.1.3 Architectural Reconstruction ................................................. 29 2.1.4 Texts ...................................................................................... 29 2.2 Ground Plan Analysis .................................................................... 30 2.3 Space Syntax .................................................................................. 31 2.4 Ethnography and Ethnoarchaeology .............................................. 34 2.4.1 Kramer’s Village Ethnoarchaeology ..................................... 36 2.4.2 Archaeological Ethnography in Western Iran ....................... 40

viii

Table of Contents

2.4.3 Horne’s Study of Village Spaces ........................................... 51 2.4.4 Seeden’s Analysis of Syrian Villages.................................... 56 2.4.5 Villages in Jordan .................................................................. 56 2.4.6 Cafer Höyük .......................................................................... 57 2.4.7 The Kurdish Woman’s Life ................................................... 57 Chapter Three ............................................................................................ 69 North Region Case Study: Tell Melebiya, Titris Höyük and Tell Taya 3.1 Tell Melebiya ................................................................................. 70 3.1.1 House B1 ............................................................................... 75 3.1.2 House B2 ............................................................................... 78 3.1.3 House B4 ............................................................................... 80 3.1.4 House B7 ............................................................................... 83 3.1.5 House G1 ............................................................................... 86 3.1.6 House B6 ............................................................................... 89 3.1.7 House C2 ............................................................................... 90 3.1.8 House G2 ............................................................................... 92 3.2 Titris Höyük ................................................................................... 96 3.2.1 Building Unit I ...................................................................... 99 3.2.2 Building Unit II ................................................................... 101 3.2.3 Building Unit IV.................................................................. 103 3.2.4 Building Unit 1 .................................................................... 110 3.3 Tell Taya ...................................................................................... 114 3.3.1 Area AAr ............................................................................. 120 3.3.2 Area Py ................................................................................ 121 3.3.3 Area Qaa.............................................................................. 122 3.3.4 Area Qbb ............................................................................. 123 3.3.5 Area Tcc .............................................................................. 125 3.3.6 Area Ucc.............................................................................. 126 3.3.7 Area Vx ............................................................................... 127 3.3.8 Area Yr/s ............................................................................. 129 3.4 Summary ...................................................................................... 131 Chapter Four ............................................................................................ 133 Central Region Case Study: Khafajah and Tell Asmar 4.1 Khafajah ....................................................................................... 133 4.1.1 Khafajah Mound A .............................................................. 137 4.1.2 Khafajah Mound B and C .................................................... 174 4.1.3 Khafajah Mound D .............................................................. 174 4.2 Tell Asmar ................................................................................... 175 4.2.1 Stratum Vc........................................................................... 178

Archaeological Perspectives on Houses and Households in Third Millennium Mesopotamian Society

ix

4.2.2 Stratum Vb .......................................................................... 191 4.2.3 Stratum Va........................................................................... 197 4.3 Summary ...................................................................................... 222 Chapter Five ............................................................................................ 227 South Region Case Study: Tell Abu Salabikh and Shuruppak 5.1 Tell Abu Salabikh ........................................................................ 227 5.1.1 Building 5G65 ..................................................................... 231 5.1.2 Building 6H82 ..................................................................... 234 5.1.3 Southern Unit ...................................................................... 237 5.1.4 Burned Building .................................................................. 242 5.2 Shuruppak .................................................................................... 248 5.2.1 House Vlak-al...................................................................... 252 5.2.2 House Vlls-u ........................................................................ 254 5.2.3 House XVa-d ....................................................................... 259 5.2.4 House XVad ........................................................................ 262 Chapter Six .............................................................................................. 273 Discussion 6.1 How Large were Mesopotamian Residential Groups? ................. 273 6.2 How was Space used in Mesopotamian Houses?......................... 277 6.2.1 Open Courtyard ................................................................... 277 6.2.2 Features in Rooms ............................................................... 281 6.3 What Evidence is there of Variation in the Wealth of Mesopotamian Houses? ........................................................... 293 6.4 What was Private and Public for the Occupants of Mesopotamian Houses? ........................................................... 296 Chapter Seven.......................................................................................... 307 Conclusions Bibliography ............................................................................................ 311

LIST OF ILLUSTRATIONS

Fig. 2.1. (a) Ground plan of house (Kramer 1982b, Fig. 4.6); (b) Flowchart ..........37 Fig. 2.2. (a) Ground plan of house (Kramer 1982b, Fig. 4.7); (b) Flowchart ..........38 Fig. 2.3. (a) Ground plan of house (Kramer 1982b, Fig. 4.18); (b) Flowchart ........39 Fig. 2.4. Walls of piled-up straw-tempered mud (Watson 1979, 37).......................41 Fig. 2.5. Hasanabad hearth plan and section (Watson 1979, 124) ...........................42 Fig. 2.6. Hasanabad storage pit (Watson 1979, 126) ...............................................43 Fig. 2.7. Hasanabad metal plate (saj) for cooking bread (Watson 1979, 162) .........44 Fig. 2.8. Hasanabad flour storage chest (kenu) with opening stuffed with a rag (Watson 1979, 67) .............................................................................................45 Fig. 2.9. (a) Ground plan of house (Watson 1979, 129); (b) Flowchart ..................46 Fig. 2.10. (a) Ground plan of house (Watson 1979, 133); (b) Flowchart ................47 Fig. 2.11. (a) Ground plan of house (Watson 1979, 139); (b) Flowchart ................48 Fig. 2.12: (a) Ground plan of house (Watson 1979, 142); (b) Flowchart ................49 Fig. 2.13. (a) Ground plan of house (Watson 1979, 145); (b) Flowchart ................50 Fig. 2.14. (a) (Horne 1994, 93); (b) (Horne 1994, 205) ...........................................54 Fig. 2.15. (a) Ground plan (Horne 1994, 204); (b) Flowchart .................................55 Fig. 2.19. Wooden ladder leading to roof and inside of a house in Topzawa (Hansen 1961, 35, Figs. 25–26) ........................................................................58 Fig. 2.20. Sheikh Taifur’s house (Hansen 1961, 29, Fig. 21) ..................................59 Fig. 2.21. (a) Ground plan of Sheikh Taifur’s house (Hansen 1961, 33, Fig. 24); (b) Flowchart ..........................................................................................................60 Figure 2.22. Bird’s-eye view of Sheikh Taifur’s house (Hansen 1961, 30, Fig. 22)..............................................................................................................61 Fig. 2.23. Elevation of Sheikh Taifur’s house (Hansen 1961, 31, Fig. 23)..............62 Fig. 2.24. (a) Ground plan of a family house in Sulaimani (Hansen 1961, 40, Fig. 30); (b) Flowchart ......................................................................................63 Fig. 2.25. (a) Ground plan of a house for a newly-married couple in Sulaimani (Hansen 1961, 42, Fig. 31); (b) Flowchart ........................................................64 Fig. 3.1. Third-millennium Upper Mesopotamia (Lebeau 1993).............................71 Fig. 3.2. Tell Melebiya (Lebeau 1996) ....................................................................72 Fig. 3.3. Tell Melebiya: excavated area Chantier B (Lebeau 1996) ........................74 Fig. 3.4. (a) Plan of House B1 (Lebeau 1993); (b) Flowchart of House B1 ............75 Fig. 3.5. Internal space sizes and features of House B1 ..........................................76 Fig. 3.6. (a) Plan of House B2 (Lebeau 1993); (b) Flowchart; (c) Internal space sizes and features of House B2a........................................................................78 Fig. 3.7. (a) Plan of House B4 (Lebeau 1993); (b) Flowchart; (c) Internal space sizes and features of House B4 .........................................................................81 Fig. 3.8. (a) Plan of House B7 (Lebeau 1993); (b) Flowchart .................................84 Fig. 3.9. Internal space sizes and features of House B7 ..........................................84

xii

List of Illustrations

Fig. 3.10. (a) Plan of House G1 (Lebeau 1993); (b) Flowchart; (c) Internal space sizes and features of House G1 .........................................................................97 Fig. 3.11. (a) Plan of House B6 (Lebeau 1993); (b) Flowchart; (c) Internal space sizes and features of House B6 .........................................................................89 Fig. 3.12. (a) Plan of House C2 (Lebeau 1993); (b) Flowchart; (c) Internal space sizes and features of House C2 .........................................................................91 Fig. 3.13. (a) Plan of House G2 (Lebeau 1993); (b) Flowchart; (c) Internal space sizes and features of House G2 .........................................................................92 Fig. 3.14. Third millennium northern Mesopotamia (Nishimura 2012) ..................97 Fig. 3.15. Plan of Titris (Algaze and Misir 1994, 154)............................................98 Fig. 3.16. Titris Höyük Outer Town Architecture, with permission from the Titris Höyük Archaeological Expedition, Timothy Matney and Guillermo Algaze ...99 Fig. 3.17. Flowchart of Building Unit I .................................................................100 Fig. 3.18. Internal space sizes and features of Building Unit I ..............................100 Fig. 3.19. Building Unit 2, flowcharts ...................................................................102 Fig. 3.20. Internal space sizes and features of Building Unit II .............................103 Fig. 3.21. Building Unit 4VI (Algaze et al. 2001) .................................................104 Fig. 3.22. Flowchart of Building Unit IV c ...........................................................105 Fig. 3.23. Flowchart of Building Unit IV b ...........................................................105 Fig. 3.24. Internal space sizes and features of Building Unit IV c ........................106 Fig. 3.25. Internal space sizes and features of Building Unit IV b ........................106 Fig. 3.26. Outer Town (Matney et al. 1999) ..........................................................109 Fig. 3.27. Building Unit 1 Lower Town (Matney et al. 1997) ...............................111 Fig. 3.28. Internal space sizes and features of Building Unit 1, Lower Town .......111 Fig. 3.29. (a) Tell Taya (Reade 1973; Dibo 2016) ................................................115 Fig. 3.29. (b) Tell Taya, areas Tcc and Ucc Reade (1973) ....................................116 Fig. 3.29. (c) Tell Taya, areas Py, Qaa and Qbb (Reade 1973) .............................117 Fig. 3.29. (d) Tell Taya, area Vx (Reade 1973) .....................................................118 Fig. 3.29. (e) Tell Taya, areas AAr and Yr (Reade 1973) .....................................119 Fig. 3.30. (a) Plan of house in AAr (Reade 1973); (b) Flowchart .........................120 Fig. 3.31: (a) Plan of house in Py (Reade 1973); (b) Flowchart ............................122 Fig. 3.32. (a) Plan of house in Qaa (Reade 1973); (b) Flowchart ..........................122 Fig. 3.33. (a) Plan of houses in Qbb (Reade 1973); (b) Flowchart House I; (c) Flowchart House II ....................................................................................123 Fig. 3.34. (a) Plan of house in Tcc (Reade 1973); (b) Flowchart ..........................125 Fig. 3.35: (a) Plan of house in Ucc (Reade 1973); (b) Flowchart ..........................126 Fig. 3.36. (a) Plan of house in Vx (Reade 1973); (b) Flowchart ...........................128 Fig. 3.37. (a) Plan of house in Yr (Reade 1973); (b) Flowchart ............................130 Fig. 4.1. Diyala Region (Frankfort, Jacobsen, and Preusser 1932, Fig. 1) ............134 Fig. 4.2. Plan of Khafajah (Frankfort, Jacobsen, and Preusser 1932, Fig. 19) .......135 Fig. 4.3. Aerial view of Khafajah (Frankfort 1933, 154) .......................................136 Fig. 4.4. Aerial view of Khafajah (Frankfort 1934, Fig. 54) .................................136 Fig. 4.5. Khafajah contour map of Mound A showing excavated areas (Delougaz, Hill, and Lloyd 1967, plate 1) .......................................................138 Fig. 4.6. (a) Plan of House XXXII in grey and the adjacent temple in white (Delougaz, Hill, and Lloyd 1967); (b) Flowchart............................................139

Archaeological Perspectives on Houses and Households in Third Millennium Mesopotamian Society

xiii

Fig. 4.7. Internal space sizes and features of House XXXII and the adjacent temple .............................................................................................................140 Fig. 4.8. (a) Plan of House XXXV (Delougaz, Hill, and Lloyd 1967); (b) Flowchart...................................................................................................142 Fig. 4.9. Internal space sizes and features of House XXXV ..................................143 Fig. 4.10: (a) Plan of House XXXVIII (Delougaz, Hill, and Lloyd 1967); (b) Flowchart ............................................................................................................ 144 Fig. 4.11. Internal space sizes and features of House XXXVIII ............................145 Fig. 4.12. (a) Plan of House XLI (Delougaz, Hill, and Lloyd 1967); (b) Flowchart...................................................................................................147 Fig. 4.13. (a) Plan of House XXXIII and XXXIV (Delougaz, Hill, and Lloyd 1967); (b) Flowchart of House XXXIII; (c) Flowchart of House XXXIV ......148 Fig. 4.14. (a) Internal space sizes and features of House XXXIII; (b) Internal space sizes and features of House XXXIV ...................................149 Fig. 4.15. (a) Plan of House XXXVI and XXXVII (Delougaz, Hill, and Lloyd 1967); (b) Flowchart of House XXXVI; (c) Flowchart of House XXXVII ....150 Fig. 4.16. Internal space sizes and features of House XXXVI ..............................151 Fig. 4.17. (a) Plan of House XXXIX and XL (Delougaz, Hill, and Lloyd 1967); (b) Flowchart of House XXXIX; (c) Flowchart of House XL ........................153 Fig. 4.18. (a) Internal space sizes and features of House XXXIX; (b) Internal space sizes and features of House XL .............................................................154 Fig. 4.19. (a) Plan of House XLII and XLIII (Delougaz, Hill, and Lloyd 1967); (b) Flow chart of House XLII; (c) Flow chart of House XLIII .......................156 Fig. 4.20. (a) Internal space sizes and features of House XLII; (b) Internal space sizes and features of House XLIII ...................................................................157 Fig. 4.21. (a) Plan of House XLV (Delougaz, Hill, and Lloyd 1967); (b) Flowchart; (c) Internal space sizes and features of House XLV ................160 Fig. 4.22. (a) Plan of House XLVI (Delougaz, Hill, and Lloyd 1967); (b) Flowchart; (c) Internal space sizes and features of House XLVI...............162 Fig. 4.23. (a) Plan of House XLVII (Delougaz, Hill, and Lloyd 1967); (b) Flowchart; (c) Internal space sizes and features of House XLVII .............164 Fig. 4.24. (a) Plan of House XLVIII (Delougaz, Hill, and Lloyd 1967; (b) Flowchart; (c) Internal space sizes and features of House XLVIII ............167 Fig. 4.25. (a) Plan of House XLIX (Delougaz, Hill, and Lloyd 1967); (b) Flowchart; (c) Internal space sizes and features of House XLIX...............168 Fig. 4.26. (a) Plan of House L (Delougaz, Hill, and Lloyd 1967); (b) Flowchart; (c) Internal space sizes and features of House L......................170 Fig. 4.27. (a) Plan of House LI (Delougaz, Hill, and Lloyd 1967); (b) Flowchart; (c) Internal space sizes and features of House LI ....................172 Fig. 4.28. (a) Plan of House LIII (Delougaz, Hill, and Lloyd 1967); (b) Flowchart; (c) Internal space sizes and features of House LIII..................173 Fig. 4.29. Diyala Region (Frankfort, Jacobsen, and Preusser 1932, Fig. 1)……176 Fig. 4.30. Contour map of Tell Asmar (Delougaz, Hill, and Lloyd 1967, plate 23) ..........................................................................................................177 Fig. 4.31. Stratum Vc of Tell Asmar (Delougaz, Hill, and Lloyd 1967, plate 24) ..........................................................................................................179

xiv

List of Illustrations

Fig. 4.32. Stratum Vb of Tell Asmar (Delougaz, Hill, and Lloyd 1967, plate 25) ..........................................................................................................180 Fig. 4.33. (a) Flowchart; (b) Internal space sizes and features of House I Stratum Vc.......................................................................................................................... 181 Fig. 4.34. (a) Flowchart; (b) Internal space sizes and features of House III Stratum Vc.......................................................................................................................... 183 Fig. 4.35. (a) Flowchart; (b) Internal space sizes and features of House II Stratum Vc.......................................................................................................................... 184 Fig. 4.36. Reconstruction of House II Stratum Vb of Tell Asmar (Delougaz, Hill, and Lloyd 1967, plate 35) ...............................................................................186 Fig. 4.37. (a) Flowchart; (b) Internal space sizes and features of House II Stratum Vb ......................................................................................................................... 187 Fig. 4.38. (a) Flowchart of House II Stratum Va ...................................................189 Fig. 4.38. (b) Internal space sizes and features of House II Stratum Va ................190 Fig. 4.39(a). Flowchart of House I Stratum Vb .....................................................191 Fig. 4.39(b). Internal space sizes and features of House I Stratum Vb ..................192 Fig. 4.40. Reconstruction by Viviana Russo: Stratum Vb, remains of House I, reconstruction of elevation and features of House II and reconstruction of House IV completed with roof ........................................................................193 Fig. 4.41(a). Flowchart of House V Stratum Vb ...................................................194 Fig. 4.41(b). Internal space sizes and features of House V Stratum Vb ................194 Fig. 4.42(a). Flowchart of House VI Stratum Vb ..................................................196 Fig. 4.42(b). Internal space sizes and features of House VI Stratum Vb ...............196 Fig. 4.43. Stratum Va of Tell Asmar (Delougaz, Hill, and Lloyd 1967, plate 26) ..........................................................................................................197 Fig. 4.44(a). Flowchart of House I Stratum Va .....................................................199 Fig. 4.44. (b) Internal space sizes and features of House I Stratum Va ..................... 199 Fig. 4.45. (a) Flowchart; (b) Internal space sizes and features of House V Stratum Va.......................................................................................................................... 201 Fig. 4.46. (a) Flowchart; (b) Internal space sizes and features of House VI Stratum Va................................................................................................................ 202 Fig. 4.47(a). Flowchart of House VIII Stratum Va ................................................204 Fig. 4.48. (a) Flowchart of House X; (b) Flowchart of House XI; (c) Flowchart of House XII; (d) Flowchart of House XIII; (e) Flowchart of House XIV; (f) Flowchart of House XV; (g) Flowchart of House XVI; (h) Flowchart of House XVII ................................................................................................206 Fig. 4.49. (a) Internal space sizes and features of House X; (b) House XI; (c) House XII; (d) House XIII .........................................................................208 Fig. 4.50. (a) Internal space sizes and features of House XIV; (b) House XV; (c) House XVI .................................................................................................211 Fig. 4.51. (a) Internal space sizes and features of House XVII; (b) House XVIII; (c) House XIX .................................................................................................214 Fig. 4.52. (a) Flowchart of House XVIII; (b) Flowchart of House XIX; (c) Flowchart of House XX; (d) Flowchart of House XXI ...................................216 Fig. 4.53. (a) Internal space sizes and features of House XX; (b) House XXI ......217 Fig. 4.54. (a) Flowchart; (b) Internal space sizes and features of House XXII ......219

Archaeological Perspectives on Houses and Households in Third Millennium Mesopotamian Society

xv

Fig. 4.55. (a) Flowchart; (b) Internal space sizes and features of House XXV .....221 Fig. 4.56. Surface area distribution of houses in the central region .......................225 Fig. 5.1. Third-millennium southern Mesopotamia (Postgate 1994) .....................228 Fig. 5.2. Site plan of Tell Abu Salabikh (scale 1:10,000) (Postgate 1983) ............229 Fig. 5.3. Southeastern Quarter of Abu Salabikh Main Mound (Postgate 1990) ....230 Fig. 5.4. Area A of Abu Salabikh Main Mound (Matthews and Postgate 1987) ...231 Fig. 5.5. (a) Plan of House 5G65 (Matthews et al. 1994); (b) Flowchart; (c) Internal space sizes and features of House 5G65 .......................................233 Fig. 5.6. (a) Plan of House 6H82 (Matthews et al. 1994) ......................................235 Fig. 5.7. (b) Flowchart; (c) Internal space sizes and features of House 6H82 .......236 Fig. 5.8. Plan of Area E (Hansen 1974, Fig. 1) .....................................................238 Fig. 5.9. (a) Plan of Area E Southern Unit (Matthews et al. 1994); (b) Flowchart; (c) Internal space sizes and features of Area E Southern Unit ........................239 Fig. 5.10. Entrance to the Southern Unit (Hansen 1974, Fig. 2)............................240 Fig. 5.11. Partially excavated bowl set into the floor of room 39 in the Southern Unit (Hansen 1974, Fig. 7)..............................................................................240 Fig. 5.12. Room 29 in the Southern Unit (Hansen 1974, Fig. 3) ...........................241 Fig. 5.13. (a) Plan of the Burned Building (Matthews et al. 1994); (b) Flowchart; (c) Internal space sizes and features of the Burned Building ...243 Fig. 5.14. Courtyard (Room 4) with bitumen runners in the Burned Building (Hansen 1974, Fig. 13)....................................................................................244 Fig. 5.15. Fireplaces in room 7 in the Burned Building (Hansen 1974, Fig. 11) ...245 Fig. 5.16. Drain in room 6 in the Burned Building (Hansen 1974, Fig. 12) ..........246 Fig. 5.17. Third-millennium southern Mesopotamia (Postgate 1994) ...................249 Fig. 5.18. Plan of Fara (Starzmann 2007, Karte 1) ................................................250 Fig. 5.19. (a) Plan of House VIak-al (Heinrich 1931); (b) Flowchart; (c) Internal space sizes and features of House VIak-al....................................253 Fig. 5.20. House VIIs-u (Heinrich 1931, Fig. 7a)..................................................255 Fig. 5.21. (a) Plan of House VIIs-ur (Starzmann 2007); (b) Flowchart; (c) Internal space sizes and features of House VIIs-ur ....................................256 Fig. 5.22. Fireplace along the eastern wall of the court in House VIIs-u (Starzmann 2007, Fig. 41)...............................................................................258 Fig. 5.23. Excavation of area XVa-d (Heinrich 1931, 16) .....................................260 Fig. 5.24. (a) Plan of House XVa-d (Starzmann 2007); (b) Flowchart; (c) Internal space sizes and features of House XVa-d .....................................261 Fig. 5.25. Excavation of area XVad (Heinrich 1931, 16) ......................................263 Fig. 5.26. (a) Plan of House XVad west (Starzmann 2007); (b) Flowchart; (c) Internal space sizes and features of House XVad west ..............................265 Fig. 5.27. (a) Plan of House XVad east (Starzmann 2007); (b) Flowchart; (c) Internal space sizes and features of House XVad east ...............................268 Fig. 5.28. Surface area of houses in Fara...............................................................270 Fig. 6.1. Surface area of the sixty-eight buildings analysed ..................................274 Fig. 6.2. Estimation of the number of people per house ........................................276 Fig. 6.3. The presence and absence of open courtyards.........................................279 Fig. 6.4. Percentage of features found in 467 analysed spaces ..............................281

xvi

List of Illustrations

Fig. 6.5. Number of objects for a given space on the abscissa and the percentage of the space relative to the total house area on the ordinate ............................282 Fig. 6.6a. Locations of hearths and domestic ovens from Matthews and Postgate (1987), Algaze et al. (2001), Delougaz, Hill, and Lloyd (1967), Lebeau (1993), and Starzmann (2007).........................................................................283 Fig. 6.6b. Locations of hearths and domestic ovens, from Algaze et al. (2001), Delougaz, Hill, and Lloyd (1967), and Lebeau (1993)....................................284 Fig. 6.7a. Locations of hearths and domestic ovens, from Delougaz, Hill, and Lloyd (1967), and Lebeau (1993) .............................................................285 Fig. 6.7b. Locations of hearths and domestic ovens, from Delougaz, Hill, and Lloyd (1967) and Lebeau (1993) ..............................................................286 Fig. 6.8. Locations of hearths and domestic ovens, from Lebeau (1993) ..............287 Fig. 6.9a. Locations of benches and basins, from Matthews and Postgate (1987), Algaze et al. (2001), Delougaz, Hill, and Lloyd (1967), and Lebeau (1993) ..289 Fig. 6.9b. Location of benches and basins, from (Lebeau 1993) ...........................290 Fig. 6.10. Locations of cooking facilities, internal courtyards, and strode rooms (part 1) ............................................................................................................297 Fig. 6.11. Locations of cooking facilities, internal courtyards, and strode rooms (part 2) ............................................................................................................298 Fig. 6.12. Location of cooking facilities, internal courtyards, and strode rooms (part 3) ............................................................................................................299 Fig. 6.13. Locations of cooking facilities, internal courtyards, and strode rooms (part 4) ............................................................................................................300 Fig. 6.14. Locations of cooking facilities, internal courtyards, and strode rooms (part 5) ............................................................................................................301 Fig. 6.15. Locations of cooking facilities, internal courtyards, and strode rooms (part 6) ............................................................................................................302 Fig. 6.16. Locations of cooking facilities, internal courtyards, and strode rooms (part 7) ............................................................................................................303 Fig. 6.17. Locations of cooking facilities, internal courtyards, and strode rooms (part 8) ............................................................................................................304

ACKNOWLEDGMENTS

The work in this book is the result of my PhD thesis at University College London. A PhD thesis is never the accomplishment of only one person. Many people helped me from the beginning to the end of this project. First of all, I would like to thank Prof. Roger Matthews who accepted me as a student at UCL and suggested this subject, giving me the opportunity to work on Mesopotamian houses. The topic matched my interest in and passion for Ancient Near Eastern Architecture and social aspects, and developed other interests and passions such as ethnoarchaeology and theoretical archaeology. Many thanks to Dr. Karen Wright who shared the supervision with Roger at the beginning of my work and then became my first supervisor when Roger left UCL. Karen’s supervision obliged me to be rigorous, thorough, and consistent. Without her help and support, my work would certainly be of a lesser quality. She always gave me concrete and helpful suggestions, and encouraged me when I was feeling down. I thank Dr. Rachael Sparks, who substituted for Karen during her sabbatical, and Dr. Mark Altaweel, who became my second supervisor during the last years of the PhD. Mark has been always very fast and punctual in returning my chapters with corrections and useful comments. He has also been very supportive and helpful towards the examination process. For this dissertation I would also like to thank my upgrade panel members for their time, interest, and helpful comments, especially Prof. Cyprian Broodbank who took interest in my case in his role of graduate tutor at UCL and Prof. Todd Whitelaw for his insightful questions, very useful suggestions for my research, and for kindness. I would like to thank the examiners of my viva Dr. Marcus Widell and Dr. David Jeffreys for the very enjoyable experience of my examination and their useful constructive suggestions. Because of them, the viva examination became one of the highlights of my PhD journey.

xviii

Acknowledgments

The biggest thanks go to my husband Taro who made my PhD possible as a funding source and gave technical support, moral support, a model of inspiration, and much more. Thanks also to my son Akira who delayed my work but also supported me and actually helped me in his special way. This PhD took quite some time, and during this period I lived in five homes, four cities, and two countries. We made many friends in the different places and they are too many to mention here, but their presence has been precious. They are the family away from the family. Lastly, I would like to thank my family for all their love, help, and unconditional support, and also my parents-in-law for their love, encouragement, and support. Thank you.

INTRODUCTION

This work was proposed by Prof. Roger Matthews. The development of the book followed the research done for the PhD thesis. Chapter one reviews all the available publications about household archaeology. When the research questions were chosen all the possible methodologies of analysis were explored. One of the aims was to utilise evidence from different sites in order to cover as many variations as possible. One of the biggest challenges of this work has been to find a sufficiently versatile methodology applicable to the nature of the evidence, which is very heterogeneous and often incomplete. Chapter two shows that the analysis investigated various methodologies, some of which were partially discarded when they proved to be insufficiently useful to answer the research question or insufficiently versatile to apply to different types of evidence. This happened with, for example, the spatial analysis method: the calculation of the relative asymmetry was found to not be useful (see the formula in 2.3) and it has been not utilised, while the flow charts which are part of the same theory were found to be useful in analysing privacy. To examine variations and use of space, seven sites were analysed from north (Tell Melebiya, Titris Höyük, Tell Taya), central (Khafajah, Tell Asmar), and southern Mesopotamia (Tell Abu Salabikh and Shuruppak). Chapters three, four, and five feature detailed analyses of all the houses with an almost complete layout of the ground floor among the seven sites chosen for a total of sixty-eight house plans. In chapter six, the information gathered from the detailed analysis of the previous chapters was discussed in order to answer the four research questions. In all the analysis chapters, quantitative and qualitative methods were combined.

CHAPTER ONE THEORETICAL BACKGROUND

1.1 The Archaeological Significance of Houses “Il ne devrait pas être nécessaire d’insister longuement sur l’importance que rêvet la maison dans l’expression des traits culturels d’une civilisation....[les] archéologues...dans le monde mort où ils évouluent, “oublient” parfois cette humble manifestation de la créativité humaine. Parce qu’elle n’est pas porteuse, pensent-ils d’une signification politique, artistique ou religeuse, la maison est sacrifiée au palais et au temple, jugés plus prestigeux.” (Aurenche 1981, 3)

Historically, there have been only few studies in Ancient Near Eastern archaeology that have concentrated on domestic buildings. While we have extensive knowledge of temples and palaces, much less is known about houses and residential districts. This is a serious gap in the knowledge of Mesopotamian culture, since the house is one of the main spaces for social dynamics. Taking into consideration only the “public architecture,” we understand the big political and economic movements but we miss their impact on Mesopotamian society (Pollock 1999, 147–8). Empires come and go but houses and households represent continuity in a society, being the vital element containing the fundamental characteristics of a society, and one must therefore comprehend houses to understand a culture. Moreover, they reflect the social changes that are key factors in the development of complex political systems (MacEachern, Archer, and Garvin 1989, 526). Sanders (1990, 45) says that: “A building is a cultural unit of meaning before it is an object of practical function … the built environment plays a crucial role in providing cues for proper social behavior by encoding the world view and cultural values of the builders.” In the words of Tringham, who sees architecture as an arena for social action: “the social, political, organisational, and other behavioural and cultural factors are more directly influential in shaping architecture and the

4

Chapter One

use of space than is the effect of the physical environment” (Tringham 1995, 164). When we study houses and households we find information on the material structure, people, and ideology that inform the creation of the structure and the cooperation of the group as well as the socioeconomic activities that the households are engaged in (Sillar 2000, 2). For example, evidence of cereal grinding and textile production found in houses gives us valuable information about the economy and the role of women and children in Mesopotamian society. Moreover, Schloen recognises, in what he calls the “patrimonial household model” (see below), the basis of Ancient Near Eastern societies. He believes that the “familiar patriarchal household served as the universal paradigm for all social relationships, whether economic, political, or religious” (Schloen 2001, 54). For example, household-derived terms such as “father,” “son,” “brother,” “master,” and “servant” were used to describe political relationships in diplomatic correspondence (Schloen 2001, 256). In the archaeology of the Ancient Near East, many of the relevant publications that do address domestic buildings have focused on only one or two sites and, with some exceptions, do not provide a general picture of Mesopotamian dwellings throughout the millennia (see, for example, Brusasco 1999–2000; Chesson 1997; Stone 1987; Verhoeven 1999; Starzmann 2008). The study of the long-term socio-historical developments has seldom been undertaken in recent years by students of the Ancient Near Eastern civilisation. Near East specialists have tended to restrict their investigations to topics defined more narrowly in spatial and temporal terms. (Schloen 2001)

In the next paragraph, many studies will be discussed to provide a picture of a household’s research context. The purpose of this book is to analyse the characteristics of Mesopotamian houses to fill in aforementioned knowledge gaps. By covering multiple sites over an extensive region, it is possible to examine the range of housing variations and highlight specific characteristics to draw conclusions beyond the level of the mere documentation of the excavation reports (Margueron 1997, 45).

Theoretical Background

5

1.2 Research Context There have been few exhaustive studies of private houses in Mesopotamia. Publications on its domestic architecture usually consist of reports from excavations and remain at this stage; they tend to be limited to one site and generally lack the development of a longer-term architectural study. The three-volume study of Mesopotamian houses by Aurenche (1981) is an exception because he recognises the central importance of the study of houses in the context of archaeology. This study is principally organised by typology: he produces an excellent catalogue of building materials, building techniques, and types of houses, but he leaves little space for a cultural interpretation of the evidence. From the point of view of my research, this is a limitation because this work concentrates more on what can be known about the house as the main arena of social dynamics. Moreover, Aurenche’s study focuses on prehistory and considers Mesopotamian house culture only up to the fourth millennium BC. A number of recent works, which take into consideration a region or a long period of time and compare several sites, remain too focused on the classification of typology (Yon and Callot 1997; Battini-Villard 1999; Battini 2006). An article by McClellan (1997) analyses three sites from North Syria in the Late Bronze Age (Emar, Munbaqa, and Alalakh) and looks at variations in domestic architecture from a typological point of view. Towards the end of the paper there is a mention of more ideological interpretations, but it is far from exhaustive. An example of an interesting study based on only one site is the article by Vallet (1997) about Habuba Kabira, which gives a typology of houses from the Uruk Period with a study of the evolution of the dwellings. There are other relevant studies, such as Stone (1987) about Old Babylonian Nippur. The author makes a parallel with several medieval Islamic cities (Fez, Damascus, Aleppo, Cairo, Baghdad, Harat, Isfahan, and Bam). From the Isin-Larsa and Old Babylonian texts, she interprets that a quarter, or a neighbourhood, of the Mesopotamian city was, like in medieval Islamic cities, a social and geographical entity. Her hypothesis is that the population belonged to all classes and shared a common village origin or ethnic identity. There was a clientele organisation. Spatially, a quarter was organised with a main artery and several cul-de-sacs at its

6

Chapter One

sides. Stone estimates the population of a neighbourhood to be between five hundred and one thousand people (Stone 1987, 3). The limitation of this very good study, as Brusasco notes, is that it lacks a theoretical discussion of the relationship between house plan and social structure. Moreover, she concentrates on only one site and makes her assumptions about Mesopotamian society based only on the example of the city of Nippur (on the lack of theoretical discussion in comparative methodology see Schloen [2001, 3]). In 1996, Pollock published the results of her survey investigating the Uruk Mound of Abu Salabikh. Her team produced maps of the distribution of artefacts that were used to interpret economic production. The result was a widespread distribution suggesting that most artefacts, such as pottery and stone tools, were produced in houses. The results were very interestingly used to support the theory that in the Abu Salabikh of the fourth millennium the production was not centralised and controlled directly by the institutions (Pollock, Pope, and Coursey 1996). The PhD thesis by Chesson (1997) on urban households in the Early Bronze Age (EBA) communities of Syro-Palestine explores how microscale data resonates with macro-scale analyses. Chesson utilises the social anthropological House Society model, first developed by Lévi-Strauss (1979), to provide a link between the two levels of analysis. Chesson proposes that EBA Syro-Palestine society was hierarchical and differed from the Mesopotamian society characterised by an economic and political class system. Her study is based on the sites of Tell el-Handaquq South, Khirbet es-Zeraqun, and Arad. A useful concept from the same thesis, also employed by other authors (Chesson 1997, 110, 114, and 183), is the potential non-utilitarian use of artefacts, for example pottery. Kohlmeyer (1996) published the results of the excavation of the Late Uruk Period in Habuba Kabira South. In that excavation, several house plans were exposed in portions of residential neighbourhood. An analysis of the space use led to a hypothesis of multifunctional rooms in which several activities took place, such as eating, sleeping, indoor working, and cooking. Forest (1997) analyses one house from the Uruk Period in Jebel Aruda from an anthropological point of view, and hypothesises that there was a gender and age division of space and an evolution of the dwelling from the Obeid Period.

Theoretical Background

7

Wattenmaker (1998) published a study on Kurban Höyük, a thirdmillennium site in Turkey. She explored the social aspects of thirdmillennium society, asking questions, such as, “Why did craft specialisation intensify concurrently with complex, urban societies?” She sought answers to this by looking at non-elite residential areas to investigate beyond the assumption that specialisation is only related to the production of prestige goods for elites. This interesting book is again limited to one site only. In 1999, Allison edited a collection of articles from different geographical areas and periods, which is important in approaching the problems of the archaeology of households. From the point of view of architecture, for example, Allison addresses the problem that whoever designs a house may not be the one who inhabits it (Allison 1999, 4). As a consequence, we cannot assume that the archaeological remains are a reflection of the inhabitants. Following this logic, it would be possible to investigate the physical house only and not the social unit. This is an interesting point of view that can be taken into consideration, but it must not be brought to the extreme. It is useful to keep in mind that our perspective may be biased towards the builder’s point of view, but houses have some information to give us about the households, especially mudbrick houses that are very easily modified to accommodate the different needs of different generations. Moreover, even if we cannot retrieve detailed information about a single house owner, it is possible to have more general information about the cultural values of that specific society (Sanders 1990). In fact, if the interior details of a house can reflect the personal values of the occupant family, “the generalities and behavioural cues of the architectural organisation and forms reflect broader cultural conventions” (Sanders 1990, 46–7). In 2001, Schloen published a very interesting book about the concept and the symbolism of “The House of the Father” in the Levant. For the analysis, he looked at the archaeological and textual evidences from firstmillennium Ugarit. Before the analysis of the evidence, there is a very thorough discussion of interpretative theoretical paradigms to approach the history of the Ancient Near East. Schloen criticises what he calls the twosector model (see 1.5.1 below for a more-detailed explanation) adopted by Diaconff and Liverani and based on Marxist analysis (Schloen 2001, 187– 210). As an alternative interpretation, he proposes the Patrimonial Household Model (see 1.5.1 below for a more-detailed explanation) derived from Weber’s theories (Schloen 2001, 63). He applied this model to the interpretation of some well-known Near Eastern Bronze Age

8

Chapter One

societies supporting the interpretation with an accurate analysis of contemporary textual evidences (Schloen 2001, 255–313). In 2001 Pfälzner published his thesis on “Habilitation.” This work is organised by typologies with a classificatory purpose and built an excellent catalogue of a vast amount of domestic material. The author is also interested in understanding the functions of rooms. To achieve this, he utilises percentage statistics of artefacts (Pfälzner 2001, 62–3). Pfälzner also used the technique of building circulation charts to represent accessibility as in Hillier and Hanson (1984), although their work is not explicitly cited. The limitation of this study, as Bernbeck points out, is that Pfälzner’s methodology of excavation (“diachronische Flächengrabung”) seems to fail in identifying use phases. Only seven out of sixteen Bederi houses analysed display several use phases, and in regard to the ceramic analysis only four houses can be taken into consideration because the others do not have any vessels, or the vessels are present only on the latest phase (Bernbeck 2006, 125). To help in his interpretation, the author utilises ethnographic analogies with material from Syria and Africa. The analysis is conditioned by the assumption that the rooms were monofunctional. A recent method utilised to study houses is the micro-debris analysis, which is a growing field in archaeology and also has its representatives in Near Eastern archaeology. The approach focuses on the study of microartefacts. Rainville (2005) applies this approach for a detailed discussion of households from three Upper Mesopotamian sites in modern southeastern Turkey: the ancient cities of Kazane Höyük and Titris Höyük and the ancient village of Tilbes Höyük. These sites belong to the Early Bronze Age dating between 3100 and 2250 BC. The study investigates the variability in activity areas in houses in order to relate the architectural and artefactual evidence to the role of household labour in early Near Eastern settlements. Complementary to micro-archaeology is the micro-morphology approach that analyses the geological and chemical composition of sediment samples (the sedimentary matrix and the organic remains). Matthews was one of the first to apply this analysis to Near Eastern evidence (see, for example, Matthews et al. 1994, 171–212 and Matthews 2003, 377–88). In 1975, Matthews applied quantitative analysis during the excavations of the Sumerian site of Abu Salabikh. At that time, the systematic collection of animal bones and flotation from charred botanical remains was not usually

Theoretical Background

9

utilised in historical Near Eastern archaeology (Matthews et al. 1994, 172). Matthews also performed thin-section analysis on occupation deposits, finding information on the use of space see for example Matthews et al. 1994, 196). In 2008, Nishimura completed a PhD thesis on the layout of Titris Höyük in the third millennium, combining archaeological evidence with the magnetometry surveys conducted on almost half of the thirty-five hectares of the ancient settlement (Nishimura 2008, xvi–xvii). The increasing use of survey techniques allows for the analysis of urban layout on a large scale. It is often possible to identify city walls, residential districts, temples, and palaces and their spatial relationship rather than limiting research to the partially excavated areas. The above are some examples of publications about houses in Mesopotamia. To the best of my knowledge, there is no study on the general characteristics of Mesopotamian houses in the third millennium BC as observed by Margueron in his considerations about the state of the research (1996, 21–2). Moreover, the majority of the studies have not related household-level evidence to the emergence of urbanism. In this work, some sites of third-millennium Mesopotamian society will be considered and will be analysed as case studies. The third millennium in Mesopotamia is a critical period of early urbanisation. It is very important to look at houses to understand the culture of this period. The intent of this thesis is to identify common aspects and differences in Mesopotamian houses. Moreover, I intend to relate these characteristics to the socioeconomic history of the period to broaden the understanding of this significant period in Mesopotamian culture.

1.3 Research Issues and Questions To begin the analysis it is useful to make some research questions explicit. The intention of some of these questions is to identify variations in Mesopotamian houses where they can be related to geographical (differences among regions or sites), chronological, or social (related to different strata of the population) factors. These findings would in turn provide a basis for reflection on their difference or variation as a function

10

Chapter One

of the relationship between architecture and social behaviour in a complex society. It is possible to ask some questions about Mesopotamian houses, such as: (1) How large were Mesopotamian residential groups? (2) How was space used in Mesopotamian houses? (3) What evidence is there of variation in the wealth of Mesopotamian houses? (4) What is private and public for the occupants of Mesopotamian houses? In the following I will explain where these questions come from and why they are asked.

1.3.1 How large were Mesopotamian residential groups? One key aspect when analysing houses and households is knowing the number of components of a residential group. This basic information is not easy to find for ancient societies. If a minimum required amount of square metres per person must have existed, for poor household this could have been as small as the few square metres necessary to lay on the floor for sleeping. On the other hand, issues of wealth and prestige could have influence the maximum number of square metres available per person. The variation between those two limits could have been large. Textual evidence, where available, is not of much help because it mainly regards the contracts of sales of properties or inheritance involving only the individual taking part in the legal transaction and not all the components of a household that in some cases could comprehend servants and/or various kin (see paragraph 2.1.4 for this topic). Even evidence on the sizes of populations in ancient Mesopotamian towns is at the level of hypothesis and the numbers of inhabitants are estimated in range with large uncertainty, for example between 2,120 and 10,303 individuals proposed by Postgate for Sumerian cities (see paragraph 1.5.1). Schloen, in his book The House of the Father as Fact and Symbol (2001), suggests that Bronze and Iron Age Levant was characterised by “sharecropping, impartible inheritance, [and] complex-family households.” Schloen thinks that because the land available for agriculture was scarce the inheritance was impartible, and landless people survived as dependent

Theoretical Background

11

household workers in a complex family as servants or in some form of adoption (Schloen 2001, 119–20). One of the aims of this work is to see if Schloen’s model for Bronze and Iron Age Levant or other models derived from ethnographical and ethnoarchaeological data (see the chapter on methodology) can be applied to third-millennium Mesopotamia using the archaeological evidence from some case study residential districts.

1.3.2 How was space used in Mesopotamian houses? The interpretation of the use of space in Mesopotamian houses has always been difficult. It has already been pointed out that, in the history of excavations, residential buildings had too often been less attentively excavated then other buildings, and moreover that stratigraphy in small buildings is more delicate and there are more frequents repairs, subdivisions of rooms, and changes of use within the lifetime of a dwelling (Postgate 1994, 58–9). Postgate cites three studies as examples of the analysis of the use of space in Mesopotamian houses based on evidence collected in well-conducted excavation (Postgate 1994, 59). Two studies are from the 1980s: Henrickson’s analysis of third-millennium Khafajah (Henrickson 1981) and Stone’s analysis of second-millennium Nippur (Stone 1987). The third is Postgate team’s excavation at Abu Salabikh in the 1990s. The examples show how good studies produced results with too many uncertainties to allow for the confident attribution of function to rooms in houses (Postgate 1994, 59). Generally, for this analysis, built-in features are taken into consideration: some are evident indicators of room functions, while others are subject to interpretation. These elements are: hearths, ovens, tannours, benches, drains, basins, bins, jars embedded in the floor, and other details such as plastered floors, the presence of altars in the room, or tombs under the floor. This information helps in the definition of room functions, while some are self-evident—for example ovens and tannours indicating where cooking took place.

Chapter One

12

Hearths are usually identified as heating facilities and not cooking places, this interpretation leading to identifying rooms with hearths as “living rooms” by archaeologists (Postgate 1994, 59–60; Kramer 1979). Objects, on the other hand, are not of great help in identifying room function. Not many objects were left in houses no longer in use, and those found may have come from collapsed roofs (Postgate 1994, 60). In the cited examples of Khafajah and Nippur, the distribution of artefacts was not especially helpful in establishing room use (Henrickson 1981; Stone 1987). Analyses of the plan of buildings have been useful in understanding the use of rooms. The size, shape, and position of rooms relative to courtyards, other rooms, or outside helped archaeologists to make hypotheses (Matthews and Postgate 1987; Deblauwe 1994; Brusasco 1999–2000). In paragraph 1.4 below and in the methodological chapter, models of spatial analysis are analysed and some are tested in the case studies.

1.3.3 What evidence is there of variation in the wealth of Mesopotamian houses? Attempts have been made to collect evidence of variation in wealth in Mesopotamian houses. Usually, the first object of analysis is the size of dwellings. Kramer researched a modern Iranian village, showing a correlation between compound size and wealth: richer households, in her samples, possessed larger land. But there was not a correlation with wealth and the actual size of dwellings. In Kramer’s evidence, richer families owned more land but did not use more square metres of dwelling space per person (Kramer 1979, 152–4). Schloen pointed out how: archaeologists working in the Near East traditionally identified larger houses as those of the rich and smaller houses as those of the poor. But wealth is highly correlated to family size under pre-modern conditions, because richer householders can afford polygamous marriages or more numerous servants. (Schloen 2001, 181)

Luxury objects are another indication of wealth that may be found on floors and in tombs of houses. Moreover, the presence of seals and private archives of clay tablets are considered indicators of the social status of the inhabitants of a house that may be part of the administrative elite of the

Theoretical Background

13

city. Indeed, registration contracts and lists of servants and goods have been found in household archives.

1.3.4 What was private and public for the occupants of Mesopotamian houses? Traditionally, a dichotomy between “public” and “private” spheres has been seen in ancient society. The interpretative models are usually assumptions derived from Western paradigms that see men dominating the public sphere and women the private sphere. One of the consequences of this model is that “women’s work” is considered irrelevant to the nondomestic sphere (Rainville 2005, 5). This could lead to totally overlooking some segments of the Mesopotamian economy, such as craft production carried out in houses or food processing such as the grinding of cereals. Many archaeologists note the similarity between Mesopotamian houses and residential areas with Islamic houses and residential areas. The characteristics are seen in the research of privacy in the winding streets, narrow exterior doorways, and absence of exterior windows (Schloen 2001, 329). Then there is the aspect of private archives in houses that seems to comprise texts concerning private transactions and public administration. The concept of public administration seems to be different from the modern Western idea. For example, Schloen points out that seals are personal, related to a particular person and not a public role (Schloen 2001, 265). Researchers who wanted to analyse the concept of privacy using archaeological evidence are increasing using the tool of space syntax which allows for the quantitative description of patterns of spatial differentiation (Deblauwe 1994; Brusasco 1999–2000; Starzmann 2008). This model is analysed in detail in the methodological chapter, and is applied to the case studies of this theses (see chapters three, four, and five). Archaeological evidence for the analysis of households in thirdmillennium Mesopotamia is principally represented by the ground floors of houses. For this reason, “spatial studies” are more and more applied as models for interpretation and represent one of the best tools available. The next section is an overview of the publications about spatial studies.

14

Chapter One

1.4 Spatial Studies Traditionally, the “spatial analysis” of houses has been mainly based on the distribution and association of household artefacts because the objects have always been considered the most important source of information (Steadman 1996, 63). Recently, research has sought a different point of view to broaden the understanding of intra-site sociocultural relationships by focusing on the container of objects as a main source of information. In this way, the meaning of architectural structures has been emphasised, promoting the study of the spatial relationship of built-in features within rooms, the relationship of one room to another, and the relation of the entire domestic unit to others in the same settlement (Steadman 1996, 64). One way to interpret the meaning of an architectural structure is to look at its symbolic value using a theoretical framework derived from linguistic theory and cultural geography (Steadman 1996, 65). Another approach to such issues is to focus on parallel opposition. For example, within a house it is possible to contrast dark with light, female with male, internal with external, and clean with dirty, applying a variety of analytical perspectives to the same space (Lawrence 1990, 76; Steadman 1996, 65). When applying such a schema to the Mesopotamian world we can remember that the interior/exterior opposition is strongly present within this civilisation in the concept of the city: the interior, the city, represents a zone of security in opposition to the hostile forces of the outside world (Castel and Charpin 1997, 251). Perhaps this opposition can be translated to the single house (interior) in opposition to the exterior. Among the “hostile forces” was the weather: the strong sun, wind, dust, and rain. The climate in Ancient Mesopotamia was characterised by very hot summers and cold winters with rain and snow in the northern regions. It was similar to the modern climate in the area, but wetter until the second part of the third millennium BC (Altaweel, personal communication). There are also cultural characteristics that can be seen as potentially hostile elements to the family. A common characteristic of Mesopotamian houses is being quite closed in with few or no windows (Castel and Charpin 1997, 252) and almost no decoration on the outside in contrast to a decorated interior: the front of the house and the other sides present no features of interest, nothing but blank walls broken by the entrance-door alone, thus

Theoretical Background

15

emphasising the introverted character of the house that probably means a need of privacy for the household as a whole. (Brusasco 2007, 57)

The interplay between material culture and architectural pattern can also be analysed from the point of view of semiotics, which recognises in buildings signs and symbols of a system of non-verbal communication. The focus of this type of analysis is accessibility, the interrelationship between spaces, and the social meaning of space organisation (Steadman 1996, 66). A very interesting study by Hillier and Hanson starts from the premise that spatial organisation is a product of social structure (Hillier and Hanson 1984, x): “The ordering of space in buildings is really about the ordering of relations between people” (Hillier and Hanson 1984, 2). The authors take into consideration not the physical or functional nature of individual spaces, but the system of spatial relations that constitute a building (Hillier and Hanson 1984, 3). The problem with Hillier and Hanson’s approach is that it is built from analysing modern Western societies, and the social interpretation of the results of the analysis are not directly relevant to traditional ancient societies. This permeability approach is increasingly applied to the study of the use of space in Mesopotamian houses (Deblauwe 1994; Starzmann 2008). A study by Brusasco utilised the concept of permeability and control to assign a degree of accessibility to each room in a house in order to identify social segregation (1999–2000, 9–10). Moreover, he takes into account the opposition of gender-specific versus non-gender-specific loci, age-specific versus non-age-specific loci, and multipurpose versus mono-functional loci (Brusasco 1999–2000, 91–2). In chapters two, three, four, and five, permeability charts based on Hillier and Hanson’s theory have been applied to the case studies in order to look at possible variations in Mesopotamian houses. Psychological approaches see spatial behaviour as one of the principal ways of non-verbal communication because human behaviour influences the organisation of the built environment. Spatial detail may contain clues to social information such as personal values, experiences, aspirations, and sociocultural tendencies (Sanders 1990, 44–7; Lawrence 1990, 75; Brusasco 2007, 23).

16

Chapter One

The aforementioned approaches have in common the attempt to broaden the understanding of interaction between human behaviour, architecture, and spatial order; the works of Kent and Rapoport combine elements of such different approaches with the household archaeology (Steadman 1996, 68). Kent builds a cross-cultural model that sets up a direct correlation between sociopolitical complexity and increasing architectural segmentation (Kent 1990, 5). She also analyses the parallel increase in the ratio of genderspecific to non-gender-specific activity areas as well as the ratio of function-restricted to multipurpose areas (Kent 1990, 5). In order to do this, Kent identifies five categories of increasing sociopolitical complexity (Kent 1990, 128–48). The attempt is to look at universal characteristics in order to avoid the controversial use of definitions such as “chiefdom.” The result has become a matter of controversy, with scholars disagreeing with her classification (Whitelaw, personal communication; see also Brusasco 1999, 93). Indeed, as Kent herself acknowledges, not every society can be accommodated within one of the five categories (Kent 1990, 149). It seems that it is possible, however, to separate the debated classification from Kent’s intuited relationship between social complexity and architectural segmentation and from her study of function in relation to gender-specific spaces. Rapoport researches Environment-Behaviour studies (EBS). In a 2006 article he explains his point of view on the relationship between archaeology and EBS. He outlines the overlapping of interest in the two disciplines, for example in the emphasis on material culture. He noticed the growing interest of archaeologists in everyday life and environment, especially dwellings and settlements. The consequence is “a greater realisation of the significance and importance of space as opposite to buildings and artefacts” (Rapoport 2006). In 1990, Rapoport summarised his approach to the link between human behaviour and the built environment. For Rapoport, the relation between activities and architecture is mediated by culture. Architecture affects behaviour and activities tend to shape architecture (Rapoport 1990, 11). He insists on the importance of considering systems within systems of settings. The observation that activities are organised not only in space but also in time is useful (Rapoport 1990, 18). Too often, we try to identify “the room” for one specific activity, probably influenced by our experience with Western, twenty-first century houses. Traditional societies

Theoretical Background

17

tend to have multipurpose rooms in which a broad range of activities took place (Hansen 1961; Watson 1979; Kramer 1982b; Aurenche et al. 1997). Sanders notes that, “Territorial markers play a major role in the regulation of expected behavior” (1990), and Rapoport focuses attention not only on “fixed” elements in houses (for example, benches and fire places) but also “semi-fixed” features. In this way, he highlights the fact that activities in a room can change with the replacement of a tool or a piece of furniture (Rapoport 1990). His intuitions can be very useful for understanding the physical layout of architectural remains and to attempt to “read” the societies that produce them. Every person needs a certain degree of privacy, even inside a “private” environment such as one’s house. In the case of contemporary Western society there are personal rooms such studios or bedrooms. In the case of traditional societies, this privacy can be present, for example, in different areas of a multipurpose room. In fact, there are many ways to divide an area into different spaces, for example with semi-fixed features like furniture, other portable items (Steadman 1996, 67), curtains (Aurenche 1981, 211), or without any material object. A division can be abstract or temporary for a number of hours of the day. In the archaeological record it is possible to identify visible boundaries such as walls, but there are often invisible boundaries composed of perishable materials, for example (Steadman 1996, 67).

1.5 House, Household and Society in Upper and Lower Mesopotamia The scope of this research is “Greater Mesopotamia,” a term that, in the current scholarly practice, designates the region that includes the Mesopotamian alluvial plain in the south of modern Iraq and the northern area comprising Upper Syria, southeastern Turkey, northern Iraq, and part of the Iranian plateau. From a chronological point of view, the mid to late third millennium BC is taken into consideration. As was said before, this is a critical period in the early urbanisation of the region. The evidence base from this time is rich. We talk about Greater Mesopotamia because the area during the third millennium and after, with certain sub-regional differences, has general cultural characteristics that are usually ascribed to a culture.

18

Chapter One While Upper Mesopotamia should be considered separately from Lower Mesopotamia (due to distinct climatic, political, religious, and most likely, social traditions), the inhabitants of both regions did share certain types of material culture, including cylinder seals, administrative iconography, cuneiform writing, and ceramic styles. (Rainville 2005, 12)

Frankfort developed the Early Dynastic chronology on the basis of his excavations in the Diyala Region (Frankfort 1934). Later historians adopted the same periodisation (Liverani 1988, 164–6). In this chronology, Early Dynastic I and II overlap with the Early Bronze Age. In terms of absolute chronology, different dates are proposed by different researchers: ca. 2900–2750 BC (Liverani 1988, 164–6) or ca. 3000–2750 BC (Rainville 2005, 121) for Early Dynastic I; ca. 2750–2600 BC (Liverani 1988, 164–6) or ca. 2750–2650 BC (Rainville 2005, 121) for Early Dynastic II. In the south, the Middle Bronze Age overlaps with the Early Dynastic III: Early Dynastic IIIa ca. 2600-2450 BC and IIIb ca. 2450-2350 BC (Liverani 1988, 164–6) or ca. 2650–2350 BC (Rainville 2005, 121) for the whole Early Dynastic III. To establish an absolute chronology for Ancient Mesopotamia, it is necessary to take into consideration many factors that can influence the accuracy (see for example Wright 1980). Wright cites some Carbon 14 analysis from six Mesopotamian sites as support of a relatively high chronology. Among them there is Abu Salabikh, where an Early Dynastic IIIa building has been dated with a good probability between 2520 and 2550 BC (Wright 1980, 95). More recent articles tend towards a middle or middle/high chronology (Manning et al. 2001). In the following chapters houses from seven cities are analysed: Tell Melebiya, Titris Höyük, and Tell Taya in northern Mesopotamia; Khafajah and Tell Asmar in central Mesopotamia; and Tell Abu Salabikh and Shuruppak in southern Mesopotamia. The selected sites contain houses dating to the Early Dynastic. The Early Dynastic is a very interesting period in Mesopotamian history because it is the time during which the majority of the specific characteristics of this culture developed. The Old Babylonian civilisation of the second millennium BC has deep roots in the third millennium BC (Postgate 1992, xxi).

Theoretical Background

19

It is useful to describe here the historical and social situation of the region during this period.

1.5.1 Lower Mesopotamia The Early Dynastic I was a time of recession in the history of Mesopotamia with an almost total interruption of commercial relationships between Lower Mesopotamia and its neighbours. The Early Dynastic II and the Early Dynastic IIIa and IIIb had homogeneous development. There was an increase in the archaeological documentation and the textual evidence from these periods. Mesopotamia at that time was politically organised in a number of competitive city states similar in dimension (Postgate 1992, 26). Mesopotamia’s alluvial plain registered a growth in population during Early Dynastic II and III. Canalisation upstream affected cities downstream leading to competition for water resources. Over a long time, this led to a progressive shift of power and influence towards the north (Liverani 1988, 164–6). A change in settlement patterns was also registered all over Sumer with the reduction in the number of villages and towns and a parallel growth in the size of a few. Chapter five analyses the city of Fara, ancient Shuruppak, which reached extended to fifteen hectares during the Early Dynastic II and IIIa (Martin 1988, 126). Liverani (1988) takes a Marxist approach and recognises two main economic systems of production in the Ancient Near East: the “Palatial” system and the “Domestic” system. In his point of view, the first is a consequence of the urban revolution, with the centralisation of the means of production in the hands of big palatial and templar organisations, with the producers living in servile conditions. He supposes a centripetal flow of goods and a redistributive, highly hierarchic system. He thinks that the second system was a residue of a Neolithic way of production in which the work forces possessed the means of production, with a web of exchanges and a system of reciprocities. In this case, specialisation was not structured by a rigid hierarchy because any specialisation involved was not the product of an overarching hierarchy, but rather was left to the individual artisan or group of producers to evolve their own divisions of labour.

20

Chapter One

In Liverani’s model, the two economic systems were in a hegemonysubordination relationship. The “Palatial” system could not exist without the “Domestic” system. The “Palatial” system of the Early Dynastic utilised the pre-existing (from the Neolithic) “Domestic” system and reorganised it for its own purposes (Liverani 1988, 52). There was also a geographical difference between north and south in Mesopotamia’s alluvial plain, with the stronger presence of a templar economic system in the south and a more “free” population in the north. Another difference between the north and the south of the alluvial plain was the easier control of the water in the north and a tendency to swamping in the south. In Early Dynastic city states, political power was represented by the palace and religious power by the temples. Temples had their own economic activities that were a part of the overall city-state organisation. It is interesting to note that the words “temple,” “palace,” are under the same linguistic category of “house” (Sumerian é, Akkadian bitum). The house was both the basic productive and administrative units. Private houses were residences for people and the place of their economic activity: likewise, the temple (the house of a certain god/goddess) was the property and residence of the god/goddess and the place of the economic activities done in his or her name. There is also an architectural resemblance between temples and houses that may reflect the intention of building temples as houses of gods (Matthews 2003a, 168). The palace was simply a big house (Sumerian é-gal, Akkadian ekallum) that reproduced a house structure on a big scale with a relationship of dependence with the “small” houses and temples (Liverani 1988, 170–2). Schloen (2001) defines the Near Eastern system as being based on a “patrimonial household model.” He analyses Bronze Age Canaan society and believes that the Near Eastern neighbours shared a similar social organisation (Schloen 2001, 53). In a patrimonial regime, the entire social order is viewed as an extension of the ruler’s household, and ultimately of the god’s household. The social order consists of a hierarchy of sub-households linked by personal ties at each level between individual masters and slaves or fathers and sons. There is no global distinction between the private and the public sectors of society, because governmental administration is effected through personal relationships on the household model rather than through an impersonal bureaucracy. (Schloen 2001, 51)

Theoretical Background

21

In light of this interpretation, the term “state” also has a different meaning than the modern Western interpretation. The Ancient Near Eastern “state” was probably view as the personal possession of the ruler—his “household” (Schloen 2001, 72). An article by Postgate (1994) gives an example of the use of space in a small third millennium city in southern Mesopotamia. He estimated a density of about twenty-six houses per hectare of urban space, with ten percent of each hectare allotted for public use (i.e. streets, public open spaces). The assessment of the total population for a small city during the Early Dynastic is a figure between 2,120 and 10,303 inhabitants. Textual evidence, for example from the archives of Fara, shows private transactions between people buying and selling private properties (Liverani 1988, 176). As described by Liverani, the transactions suggest a change of use for the house brought about by social change: the sellers of the property were several in number and were reimbursed for the sale according to their position in a family hierarchy; reflecting that the property had previously been held in common by the family. This in turn reflects the traditional use of the dwelling. It often happened that the new owner was a single person suggesting that property had changed status to that of a commodity or asset, unrelated to either family or traditional use (Liverani 1988, 176). The widespread interpretative model of Mesopotamian society during the Early Dynastic sees an increase in the workforce within the palace and the temples, with a growing number of public servants, scribes, businessmen, and specialised craftsmen that should have become a dominant elite. The consequence was a general trend towards stratification in the population, with field workers getting poorer and the specialised strata getting richer (Liverani 1988, 176–8). However, Liverani also recognises that even with an increase in craft activities and commercial activity, the economic base of Early Dynastic civilisation remained agricultural and pastoral (Liverani 1988, 178). Pollock seems to apply this model, considering that a population concentration in towns in third-millennium Mesopotamia caused a reorganisation of the economy (Pollock 1999, 117). As an example, she states there were large concentrations of the workforce in two main sectors of the economy in the cities. The first was the grinding of cereals for the production of flour. This production employed people using simple

22

Chapter One

Neolithic-style grinding stones. This activity had been carried out at the family level in previous periods, but in the Early Dynastic a large number of women were increasingly employed, working for big redistributive organisations (Pollock 1999). The second was the textile industry, as there was a large quantity of wool coming into the cities to be transformed in textiles, both for internal use and trade and export. Fabric was a typical product for export. The workforce was composed of women and children because these were workforce sectors that required little technical skills, in contrast to other sectors such as metallurgy and the processing of precious stones that required highly specialised elites, usually men (Liverani 1988, 182–3). Schloen criticised this model of the interpretation of Ancient Near Eastern societies. His hermeneutic interpretation of a society based on patrimonialism exaggerated the contraposition between rural and urban differences. He believes that most inhabitants of cities were part-time specialists, and the majority of households were self-sufficient in basic commodities (Schloen 2001, 196).

1.5.2 Upper Mesopotamia After the decline of the Uruk settlements at the end of the fourth millennium BC there was a hiatus in the urbanisation of Mesopotamia, more evident in Upper Mesopotamia than in Lower Mesopotamia. The development began to increase again during the Early Dynastic II and culminated in the Early Dynastic III (the three phases of the Early Dynastic roughly correspond to the Early Bronze Age). Then, during the Middle Bronze Age, the settlement’s occupation declined, reaching a low point in the Late Bronze Age. In this period in Upper Mesopotamia there were cities like Mari on the middle Euphrates and Assur on the middle Tigris with cultures directly influenced by the Sumerian culture and a series of settlements in the area comprised by Upper Syria, the Balikh and Khabur valleys, and Assyria, which had their own political and cultural autonomy (Postgate 1992, xxii). This area is characterised by sufficient rainfall for the practice of “dry” agriculture (that is, without the need for canalisation to irrigate fields) (Liverani 1988, 201–2). The economy in this area was predominantly based on dry farming (primarily cereals) and goat/sheep breeding (Wilkinson 1994, 485). In

Theoretical Background

23

addition, horticulture and arboriculture were practised, supplemented by hunting and fishing along the rivers (Rainville 2005, 12). From an environmental point of view there is the presence of timber and stone in northern Mesopotamia while both are rare in the south, which is rich in palms and reeds. The raw materials influenced the buildings, as southern structures were made of brick and reeds. Stone footing with a mudbrick superstructure and mud roofs supported by wooden beams were common in the north (Rainville 2005, 12). Upper Mesopotamia has suffered from a “south-centric” point of view. Many cities in Lower Mesopotamia are well known for their archaeological remains and written testimonies. These texts give us a picture of Lower Mesopotamia as the centre of the “civilised world” and the Upper lands as a “periphery,” from which, for example, raw materials flowed towards a powerful centre. This biased perspective, together with a delay in the archaeological exploration of Upper Mesopotamia, also affected scholars up until 1978, when Seyton Lloyd published a book on the archaeology of Mesopotamia with a chapter called “Illiterate People of Northern Mesopotamia” (Wattenmaker 1998, xii). Explorations in Upper Mesopotamian cities, for example Tepe Gawra, Tell Taya, Tell Khuera, Tell Brak, Tell Mozan, and many others (some of which are still in progress), are correcting this point of view (Liverani 1988, 202–4; Rainville 2005, 13; Butterlin 2009, 1). Mesopotamian culture is no longer considered a product of Sumerian civilisation (Butterlin 2009, 1). An article by Wilkinson (1994) analyses settlement patterns of the Early Bronze Age in the Jazira region of Syria, Iraq, and Turkey. He notes that settlements in this zone of “rain-fed farming,” as he calls it, tended not to exceed one hundred hectares. He estimates that, in such an area, it would have been possible to accommodate between ten and twenty thousand people with a population density of one hundred to one hundred and fifty persons per hectare. He further assumes that half of the population practised agriculture (Wilkinson 1994), while Liverani, in his comments on Wilkinson’s article, estimated that for the Near East in the Bronze Age in general around eighty percent of the active population would have been engaged in food production and twenty percent in non-food related activities (Wilkinson 1994, 509).

24

Chapter One

Bronze Age urban sites in the south Mesopotamian plains, on the other hand, could have reached areas of more than four hundred hectare. Wilkinson suggested that the limitation in settlement sizes in Upper Mesopotamia, which can be ascribed to environmental causes, could have constrained the economic and political power of the northern cites. Nevertheless, he thinks that “the productivity of Upper Mesopotamia may have been equal to or even greater than that of the irrigated lowlands, despite the letter’s traditional reputation as granary of the Ancient Near East” (Wilkinson 1994, 485).

CHAPTER TWO METHODOLOGY

In chapter one, different trends in spatial studies were discussed, leading to different methods of spatial analysis. In this chapter, Hillier and Hanson’s space syntax approach will be explained in detail, together with the ground-floor analysis that will be applied in the following chapters that analyse the evidence from the seven sites selected. Ethnoarchaeological information has been taken into consideration in order to help generate hypotheses for the use of space in Mesopotamian houses. These approaches are particularly suitable for this study, which is principally based on published excavation reports with the relative plans of buildings. In fact, ground plan analysis needs the availability of house plans and all possible observations about the remains found and recorded in the excavation reports, while space syntax needs very accurate plans of dwellings. Because of its numerical approach, it is suitable for a study of many archaeological sites with several houses each. Moreover, in order to address the research questions posed in chapter one, these methods are suitable for looking for variations among houses with the aim of interpreting them and broadening the understanding of Mesopotamian society. A challenge to analysing Mesopotamian houses is the nature of the evidence. Some of the existing evidence is badly or partially published, as often only the plan of houses is available and the phasing is not clear because of poor excavation. As has been stated, the attention of archaeologists has been traditionally focused on palaces and temples. As a result, domestic areas have often been inaccurately excavated or entirely overlooked (Battini-Villard 1999). In general, there are few houses in each site for which we know the complete layout. Without this information, it is almost impossible to understand the internal circulation at the ground floor. On the other hand, there are some well-excavated dwellings in sites such as Assur, Nippur, Larsa, and Sippar.

26

Chapter Two

Then there is the additional problem of a possible storey above the ground floor that is well documented by Margueron (1997). Scholars have different opinions about the question. Margueron, for example, thinks that a second storey was always present. If this proves to be the case, it will have a bearing on spatial analysis because we do not know the complete layout of a house. Other scholars disagree with Margueron, such as Lebeau in the case of Melebiya in chapter three. Kramer, because of her ethnographic investigation of an Iranian village, considers that every household would ideally have a second storey, but because of economic reasons this was the case for less than half of the houses in the village she studied. Kramer noted that two-storey structures generally require walls one metre thick at ground level, while single-storey houses have considerably narrower walls of around half a metre (Kramer 1982a, 99). Aurenche’s ethnographic records support Kramer’s argument. The result of his analysis is that the richest households have a second storey (Aurenche 1996, 4). From the issues mentioned above, it is clear that all evidence has limitations. Often, it is possible to analyse only a few houses per site. It is then necessary to use as many methods as possible to retrieve as much information as possible from the evidence available. This variety in methodology is reflected in the number of methods applied in this thesis, which range from the classical observation of the ground plan and the typology of houses to the flow charts that analyse the degree of permeability, to comparisons with ethnographic information.

2.1 A Survey of Some Methodologies in Houses and Household Analysis 2.1.1 Typology There are several methodologies that it is possible to apply to the study of houses. One technique that broadly applied to the analysis of buildings is the comparison of typologies. For example, houses are classified on the basis of shape, dimensions, or features. However, a complication in the application of these typologies to analyse evidence is represented by the vast number of typologies used by archaeologists. Many of them create a personal classification of houses for their case studies (Aurenche 1981; McClellan 1997; Braemer 1997; Vallet 1997; Foucault-Forest 1997; Yon and Callot 1997). The majority of typologies are based on variations of the

Methodology

27

ground-floor plan and few take into consideration hypotheses on the volumes of houses. For example, Battini-Villard’s typology is especially clear because it is simple, distinguishing between three distinct groups of houses, and is suitable for wide application. Battini-Villard organised her typology after a careful analysis of the classifications of many authors (Battini-Villard 1999, 345–50). The basis of her typology is the circulation of the inhabitants inside a house. After analysing 131 Old Babylonian houses (the total number of Old Babylonian houses with complete ground floors that had been exposed at the time of her work), Battini-Villard divided them in three types. Type I: houses “à espace central” (with a central space), Type II: houses “sans espace central” (without a central space), and Type III: houses “à salle central” (with a central room), which are in the minority when compared with the houses fitting into types I and II. In the first type, the circulation is centripetal complex, the second is linear, and the third is centripetal simple (Battini-Villard 1999, 348). Other classifications have so many types of houses that they stop being useful. The limitation of many typologies is that they take into consideration only the classification of the physical aspect of houses, lacking a discussion of the relationship between house plan and social aspects.

2.1.2 Features Another very widespread method of analysis is to examine the characteristics of the ground floor, taking into consideration all the physical aspects of the structure and the preserved fixed elements (such as fire places, benches), semi-fixed elements and objects (such as furniture, pottery), and other types of findings (such as ashes that can indicate a fireplace when other structures are lacking) (Jameson 1990; Meskell 2002). This method might lead to the formulation of a hypothesis about the possible uses of spaces. For example, Henrickson, in her study of Tell Asmar and Khafajah domestic areas, organised the evidence of each coherent house, taking into consideration five aspects: • •

whether it was a front, centre, or back room within the house whether it was an inside room with no possible window or an outside room

Chapter Two

28

• • •

whether, compared to the other rooms in that house, it was a small, medium, or large room how many doorways the room possessed what features and facilities were present in the room

Some aspects of Henrickson’s method are very similar to the method that has been applied in this work. The size of rooms has been taken into consideration in both methods, as has the presence of features and facilities inside the rooms. On the other hand, the creation of flowcharts to represent the degree of permeability of rooms, as explained below, is a more efficient method of analysis then Henrickson’s other three points. In a synthetic way, it allows for a visualisation of the access to a house, and the position of every room and doorway among each other and in relation to the outside. It also makes it possible to quantify the degree of permeability. In fortunate situations, specific analyses have been carried out, for example a micro-analysis of a portion of the floor in order to identify the geological and chemical composition of sediment samples (the sedimentary matrix and the organic remains). With this analysis, it is possible to understand the use of a surface, if animals have been kept in that space, or even if a carpet was lying on the floor. These clues lead to more accurate hypotheses on the use of the space. Matthews was one of the first to apply the micro-morphology approach to Near Eastern evidence, but this type of analysis is very rare and was not present in this theses for the majority of the case studies. A complementary approach is micro-archaeology, which investigates in detail activity areas in houses. Rainville (2005) used this approach when she studied the architectural and artefactual evidence from houses in three Upper Mesopotamian sites in modern southeastern Turkey: the ancient cities of Kazane Höyük and Titris Höyük, and the ancient village of Tilbes Höyük. Rainville attempted to investigate in depth the variability in activity areas in houses in order to relate activities to the role of household labour in early Near Eastern settlements. She compared the micro-debris analysis with the macro artefact distribution and found that they did not always match, confirmation of the fact that it is not wise to assume that an object found in a place was actually used in that place. Many events could have occurred during the abandonment of a building or after, resulting in the displacement of an

Methodology

29

artefact. For this reason, any interpretation based on the distribution of movable artefacts remain hypothetical and cannot be conclusive. Unfortunately, the results of micro-debris analysis are available for very few houses. It was possible to use them in this thesis, but for the majority of case studies they were not available.

2.1.3 Architectural Reconstruction When possible, it is useful to take into consideration architectural reconstruction. Very rarely are archaeologists able to propose a reconstruction of the volume of a building or hypothesise about the height of walls. However, attempts at architectural reconstructions are very useful because otherwise one can forget that we are dealing with threedimensional buildings, and we resonate in bi-dimensional terms, which can lead to serious misinterpretations. Ideally, archaeologists should propose various scenarios and several possible reconstructions, as Margueron did in his article about Syrian houses (Margueron 1997). As gender archaeology suggests, the intrinsic ambiguity of archaeological records calls for multiple interpretations of evidence (Conkey and Gero 1997, 424–5, 429; Tringham 1995, 81, 97). By using multiple interpretations, one can avoid becoming too attached to one interpretation at the risk of not being open to revisions when new evidence arises.

2.1.4 Texts Textual evidence can been used as a source of information about household composition. Although the majority of written texts from the third millennium BC have been found associated with large institutions such as temples or palaces (Zimansky 2005, 317), archives have also been found in private houses. From contracts and wills we have information about properties and inheritances and a glimpse of the social role of women who could, for example, own properties. There are even cases of female scribes (Lion and Robson 2005). Tentative efforts have been made to relate the household composition emerging from a private archive to the layout of the house in which the archive was found. So far, they have not been successful (Brusasco [1999–2000] tried with the well-preserved case of Old Babylonian Ur). A detailed explanation of the difficulties with this type of analysis is given in Larsen (2007), in which private Old Assyrian archives from houses in the site of Kanesh where examined. The absence

30

Chapter Two

of a system of surnames and the sparing use of patronymics make it very difficult to reconstruct family patterns (Larsen 2007, 95). Moreover, not every member of a household is named in texts, but only the individuals taking part in legal transactions such as the previously cited inheritances, sales of properties, etc. As a result, very rarely is there a complete account of a household composition that can be compared usefully with the layout of a house, and often the remains of the house in which the archives have been found are not complete for the phase in which the generation described in the texts was living in the building (Brusasco 1999–2000; Larsen 2007, 95). For the reasons described above, this project developed a method of analysis based entirely on archaeological evidence.

2.2 Ground Plan Analysis For this project analysing the ground plans from excavation reports, a database with information regarding the sites in consideration has been compiled. As already mentioned, in the best cases we have the layout of the ground floor to analyse. The collection of as much useful information as possible from the ground floor of the house under consideration has been attempted, including the total surface of the ground floor in square metres, the number of rooms, and the presence or absence of an internal open courtyard. Then, a second database was compiled for each house with information about the surface of each room, the presence in each room of fixed features such as hearths, domestic ovens (tannours), benches, drains, basins, bins, and jars embedded in the floor, and other details such as plastered floors, the presence of altars in the room, or tombs under the floor. This information has been used to build bar charts in order to compare the surface of each room (see Fig. 3.5). Each blue bar corresponds to one room, with the surface value indicated on the y axis. The detail of relevant fixed features overlaps the bar charts to show the correlation between room size and function. For example, in Fig. 3.5 the presence of a hearth has been indicated with a red square in room 15, and the presence of benches with yellow triangles in rooms 16, 5, 18, 9, 3 and 4. This information helps in the definition of room functions. Some are selfevident, e.g. ovens and tannours indicate where cooking took place. A large number of them lead to questions about the size of the household and the possible activities taking place.

Methodology

31

The second database takes into account the number of hearths that are usually identified as heating facilities and not cooking places, this interpretation leading to identifying rooms with hearths as “living rooms” by archaeologists (Postgate 1994, 59–60). In her ethnographic study, Kramer suggests that the presence of a hearth in a room identifies that space as a living room in which the family gathered and slept. In her body of evidence, storage rooms and alleys generally do not have hearths (Kramer 1982b, 104). By analysing the sleeping patterns and the demographic data, Kramer outlined a relationship between the number of hearths in a house and the number of nuclear families in a household (Kramer 1982b, 119). This information has been used to do estimations of other archaeological sites (Brusasco 1999–2000 for Ur). The problem in applying this analogy to other sites is that Kramer studied a rural site in which the majority of the households were nuclear families and the majority of the dwellings had only one living room. It is better to avoid assuming the same correspondence (one hearth = one living room = one nuclear family) for towns with larger houses that have several living rooms, especially because other ethnographic information reported extended families sleeping and eating together in only one living room (see later in this chapter). Nevertheless, the presence of hearths or tannours is a fundamental indicator of the use of space. Other important features, such as several basins or grinding stones, can indicate the presence of food processing and/or craft activities. A room with bitumen-plastered floor with drains indicates some kind of washing place, while the presence of grain-bins indicates a storage function (Postgate 1994, 59–60).

2.3 Space Syntax Where possible, an analysis of the degree of circulation based on Hillier and Hanson’s social theory for architecture has been made. Their method started from the assumption that: “spatial organisation is a function of the form of social solidarity in that society” (Hillier and Hanson 1984, 143). They create syntactic parameters in relation with social variables, arguing that variations in the syntactic properties of architectural forms reflect sociopolitical differences. This method has been increasingly applied in the archaeological analysis of houses (Dawson 2002; Deblauwe 1994;

32

Chapter Two

Brusasco 1999–2000; Ben-Shlomo and Garfinkel 2009; Starzmann 2008). However, the interpretation of the results can be problematic because “space syntax assumes that relationship between space translates directly into relationship between people” (Dawson 2002, 466). This theory was created to analyse modern Western societies, and the social interpretation of the results of the analysis is not directly relevant to traditional ancient societies. Considering only the graphical representation of buildings has been judged to be too deterministic (Lawrence 1990, 75). Moreover, Lawrence pointed out that: they give no explanation of why house form has changed in a specific context through the passage of time; or of why different house forms are provided to serve the same or similar domestic function in the same and different societies; or of why there are individual differences within a general socio-cultural pattern. (Lawrence 1990, 74)

One of Brusasco’s problems in applying the space syntax approach to Ur houses was the difficulty in matching archaeological evidence with textual evidence. In Ur, there are many private archives in houses, but they are not always from the same period as the archaeological remains of the building itself. Nevertheless, Brusasco believes that applying Hillier and Hanson’s ideas in combination with an analysis of textual evidence could provide a good model for the social analysis of households that can also be used as a method to analyse sites lacking textual evidence. The textual analysis should firmly ground the method in Mesopotamian culture, avoiding the problem of referring to only Western modern societies, as in the original Hillier and Hanson study. In addition, a set of techniques of analysis that highlights degrees of accessibility has its own value. It allowed for the possibility of comparison to search for variability among houses. In general, there is a value in outlining the general characteristics of a culture where it is not possible to define the details. Since the publication of the 1984 book The Social Logic of Space, the theory and method of space syntax have been broadened to apply to a wider range of building and settlement types and for use with computer software for a more sophisticated quantitative analysis (Hanson 1994; 1998; Hillier 1998; Peponis et al. 1997; 1998; Batty and Rana 2002). However, the new developments do not often add useful information about

Methodology

33

the basic spatial analysis of houses, especially in the archaeological context. For example, Batty and Rana (2002) concentrate on generating space syntax maps of settlements with an unique procedure to be implemented in GIS software to analyse the data. From the point of view of archaeological studies, those developments are relatively less interesting. More interesting is the application of the basic principle of the spatial analysis of houses by archaeologists who studied different periods and cultures to look at how they applied the theory and method to the subject of their studies. For example, Dawson (2002) applies the space syntax analysis to the study of Inuit snow houses. Sketches of three snow houses from ethnographic and ethnohistoric sources are used to explore whether social information is present in their spatial layouts. The three houses belong to three different Inuit groups (Copper Inuit, Netsilik Inuit, and Iglulik Inuit). Before proceeding with the spatial analysis, Dawson defines the characteristics of Inuit social organisation. These three groups occupy different regions of the Canadian Arctic. Dawson utilised space syntax to verify the assumption that social integration and cohesion intensify in Inuit culture as one moves from west to east in the Canadian Arctic. With previous identification of the characteristic features of Inuit social groups, Dawson was able to apply a Western architectural tool such as space syntax to a non-Western society, overcoming the major problem of the Hillier and Hanson approach in archaeology. A criticism of Dawson’s operation is that in order to make a valid argument about social relationships from archaeological records, one almost needs to know the answer to the question before it is asked. However, it is still a useful tool to analyse variations from different points of view and to create numerical parameters applicable in the further exploration of houses from the same culture. Hillier and Hanson built flow charts in which each room of a building is assigned an element (represented here with a square shape with a number at the centre to identify the room). Each door is represented with a connection line between two rooms (see for example 3.4[b]). As stated in chapter one, this method can be applied only where the ground floor of a building is complete and the position of the walls and the doors is known. To deepen the analysis, it is possible to apply Hillier and Hanson’s numerical method to establish the relative asymmetry (RA) of a house. This is calculated by taking into consideration the accessibility, or depth,

34

Chapter Two

value of each room of a house according to how many rooms it is away from the entrance. We can then define a mean depth, which is the average of the rooms’ depth, in other words the sum of the depth of each room in a house divided by the number of rooms (where k is the total number of rooms) and a RA (RA=2(MDí1)/(kí2)) with 0”RA”1. For a more-detailed explanation, see Hillier and Hanson (1984). Values obtained with these calculations are topological comparative tools to measure the property of space. This analysis gives information, such as the depth of a single room, which is directly related to its relative privacy, the average depth of a single house, and the RA in relation to the accessibility value of a house. Hillier and Hanson correlated these spatial values with social meaning: a low accessibility value (RA) should indicate a non-hierarchical household where the social categories of men, women, servants, or age groups are integrated; a higher accessibility should indicate segregation among social categories. The degree of social control in a house should be given by the number of ways to access a single room. If there is only one way to access a room there is a high level of control over it. More access ways indicate a lower possibility of control. When there are multiple ways to access a room it is clearly evident from the presence of “rings” in the flow charts (see, for example 5-9[b]) (Hillier and Hanson 1984; Brusasco 1999–2000). Flowcharts based on Hillier and Hanson’s method were used to look for variation in Mesopotamian houses, but not the relative asymmetry calculation.

2.4 Ethnography and Ethnoarchaeology Archaeologists often use ethnographic parallels to interpret their evidence. During the 1960s, with theoretical archaeology flourishing, an explicit archaeological ethnography started (Stiles 1977, 90). Verhoeven defined ethnoarcheology as the “study of contemporary cultures in order to obtain data that can be used to aid archaeological interpretation” (Verhoeven 2005, 251). In the 1970s, ethnoarchaeology became a sub-discipline of its own. When first used, ethnographic analogy was applied only to prehistoric evidence (Stiles 1977, 90). More recently, this method has been frequently used in the analysis of historical periods.

Methodology

35

This practice has been the object of criticism by some scholars; for example, in a recent article, Fahlander emphasises that cross-cultural analogies have sometimes been pushed too far, going from the comparison of similar objects to the interpretation of social behaviours and ideologies (Fahlander 2004). He points out a real danger, especially when the analogy is made between modern contemporary cultures and a past culture belonging to a different geographical region, for example when anthropological data from a modern African population is used to interpret prehistoric European culture (Fahlander 2004). Ethnoarchaeologists themselves admit that sometimes comparisons have been carried on too far and that the discipline needs a systematic theoretical reflection (Bazzana and Delaigue 1995, 8–9). Hodder and other post-processualists suggest that analogies are more probable when there is a closeness in time, space, or level of technology (Hodder 1982). This comparison takes the name of “Direct Historical Analogy” (Verhoeven 2005, 252). Still, Fahlander warns against the danger of assuming the existence of a continuously unbroken tradition (Fahlander 2004, 192) and no interference from other cultures. In Mesopotamia, for example, we can find a cultural and geographical continuity (Watson 1979; Kramer 1982b; Aurenche et al. 1997; Brusasco 1999–2000, 62; Dr. A. al-Hussainy personal communication), although it is necessary to take into account the influence of the posterior Islamic culture. While specific ethnographic studies are rare, it is more often possible to find historical information from later periods. Brusasco, for example, defines the resemblance between Old Babylonian houses at Ur and Islamic buildings as “more than striking” (Brusasco 2007, 112). Most of the Mesopotamian houses—like the Islamic ones—have in fact an introverted form, conceived with an inward-looking plan, with overt emphasis on the decoration of the interior elements, such as the court facade, whereas the external facade is usually a blank wall … Hence in either building walls ought to be built to a height that ensures the privacy of the domestic interior so that intruders are discouraged. (Brusasco 2007, 115)

These, however, are still characteristics of many ancient societies. Other features more typical of the Near East are the presence of tannours, which

36

Chapter Two

are still used in south Iraq, and the serdab, which is an underground room that is still excavated in south Iraq under the floor for the family to keep cool during the hot summer days. Only one hundred years ago, some families use to bury their deceased in houses under the floor of the room assigned for prayer (Dr. A. al-Hussainy personal communication). Brusasco believes that contemporary behaviour that produces a specific material culture can be compared with hypothetical ancient behaviours that produced similar structures found by archaeologists (Brusasco 1999– 2000, 61). If it is dangerous to assume that cultures that are similar in some aspect are entirely similar, Verhoeven regards ethnographic analogies as “media for thought rather than … models to be either fitted to or tested against archaeological data … in the attempt to slough off ethnocentric presuppositions” (Verhoeven 2005, 258). In other words, ethnographic data “should not be regarded as parallels, but as examples” (Verhoeven 2005, 264). The following paragraphs analyse the ethnoarchaeological studies of Watson and Kramer in western Iran, Horne’s studies in northeastern Iran, Seeden’s observations in Syrian villages, Biewrs’s analysis of Jordanian villages, the ethnoarchaeological account of Aurenche in Turkey, and Hansen’s research on Kurdish ethnography. It is important to highlight that those are ethnographic models that may or may not match the Mesopotamian.

2.4.1 Kramer’s Village Ethnoarchaeology In 1975, Kramer conducted ethnographic fieldwork in a village with the pseudonymous of Alibad near Kangavar in central western Iran (Kramer 1982b, 10). At the time, the village was inhabited by around four hundred Kurdish-speaking people and was relatively self-sufficient. They utilised wheat, barley, sheep, and goats as resources, as has been the case in that area for over seven thousand years (Kramer 1982b, 12). Alibad village shares many similarities with the Kurdish villages studied by Hansen and Watson (see below). Households were characterised by the preference for village endogamy and patri-virilocal residence in extended families (Kramer 1982b, 21). Household size ranged from one to fifteen people, but the median was six people per house (Kramer 1982b, 24).

Methodo ology

37

us houses Alleys two to three metrres wide sepaarated blocks of contiguou ul-de-sacs with gardenns and some shops, and some streets ended in cu (Kramer 1982b, 85, 88).. Dwellings were w built witth mud and, as in the other Iraniann villages, weere generally organised o arouund a courtyaard. There were stabless and storeroooms and living g rooms that w were used to eat, e sleep, and entertaiin, although thhose activities could take place in courrtyards or roofs and ffoyers accordding to the seeason. Cookinng facilities could be present in thhe courtyard or o be organiseed in a specifi fic room, and clay bins were sometiimes used for storage

(a)

(b) wchart. Fig. 2.1. ((a) Ground plann of house (Kraamer 1982b, Figg. 4.6); (b) Flow

Kramer notes that livving rooms weere the most ppublic spaces (Kramer 1982b, 10).

38

Chapter Two

(a))

(b) Fig. 2.2. ((a) Ground plann of house (Kraamer 1982b, Figg. 4.7); (b) Flow wchart.

Stables could be undergrround or at ground g level (Kramer 1982b, 107). Latrines werre present in some compou unds, while otther household ds shared such facilitiees or used their gardens (Krramer 1982b, 111).

Methodo ology

(a)

(b) Fig. 2.3. (aa) Ground plann of house (Kram mer 1982b, Fig . 4.18); (b) Flow wchart.

39

40

Chapter Two

There were five shops in Alibad in the form of a room accessed from the ally, attached to a house compound. The shops were empty of features on the inside (Kramer 1982b, 113). Generally, Alibad does not appear to have been socioeconomically complex, however there did exist a comparative economic variation (Kramer 1982b, 80). Larger compounds and larger household sizes are associated with relatively high economic rank in Alibad and elsewhere (Kramer 1982b, 55, 70). Some portable objects and fixed features reflect variations in economic status in Alibad. For example, a specialist had to be paid to install a well or a latrine (Kramer 1982b, 70). However, Kramer suggests that “no single feature of material culture is a definitive or invariably useful measure of variation in households’ rank. Rather a number of classes of evidence might be used in combination” (Kramer 1982b, 83). In order to interpret variations in Mesopotamian households it is useful to note Kramer’s observation that: Although the picture is complex, Alibad’s wealthier households tend to be larger and to have more coresiding nuclear families, and therefore more dwelling space. They also tend to reside in larger compounds than households of lower economic rank. (Kramer 1982b, 263)

2.4.2 Archaeological Ethnography in Western Iran From September 1959 to June 1960, Patty Jo Watson carried out ethnographic fieldwork while she was part of the Iranian Prehistoric Project (Watson 1979, xi). She collected information from three Iranian villages: Hasanabad, Shirdasht, and Ain Ali. Watson, being an archaeologist, gave information about details that are very relevant for interpreting archaeological data. She produced plans of the three villages, plans of all Hasanabad houses, and tables with demographic information on the population. The three villages, at the time of Watson’s investigation, had around one to two hundred inhabitants located in the Zagros Mountains of western Iran (Watson 1979, 13). Households were generally composed of a nuclear family, occasionally including a grandfather, a grandmother, or an unmarried brother. There were only four cases of extended households (Watson 1979, 223).

Methodo ology

41

nto bricks Houses weree built of stonne and straw-teempered mudd not shaped in but simply ppiled up to forrm walls (see Fig. F 2.4) (Wattson 1979, 36)). One family lived in a bllack tent, whiile other famiilies had tents in their courtyards tthat functioneed as living rooms (Watsoon 1979, 241)). Interior and exteriorr walls were plastered with midchaff (Watson 197 79, 120). Windows w were usually very v small, and the occasioonal large win ndow was sealed durinng winter becaause the walls were generallly half a metrre wide or more, the seealed window w becoming a storage nichee. Ventilation in rooms were also a co was also proovided by smaall holes in thee roof, which w onvenient means of coommunication with people on o the roof duuring the warm m season. Doorways w were usually loow, around 1.5 metres highh, and one mettre or less wide (Watsoon 1979, 1222). Floors werre covered wiith carpets an nd people took off thheir shoes uppon entering (Watson 19779, 182). Roofs were constructed of poplar beaams covered with w reed matts or thin plan nks and a layer of straw w-tempered mud m (Watson 1979, 283).

Fig. 22.4. Walls of piled-up straw-teempered mud (W Watson 1979, 37)

Houses gennerally had a walled courty yard in front and a foyer called an aywan connnecting the coourtyard with h a living rooom. Opening into the courtyard w were stables that could have been onn the ground level or undergroundd (Watson 19979, 126); th here were offten dung-fueel storage chambers caalled tapkadann as well as sttraw storage ro rooms called keyan k and small roomss for firewoodd called hizmda an (Watson 19979, 36).

42

Chapter Two

Wealthy fam milies sometim mes had two living rooms and in such cases c one was maintaiined for guessts. The secon nd living roo m was on th he ground floor or buillt as a second--story addition n (Watson 19779, 126). Kitchens annd stables couuld be subterraanean (Watsoon 1979, 183). Ain Ali features ram mps of unshapped limestonee rocks and ppacked earth to allow access to thee roofs whichh did not exceed three metrres high (Watsson 1979, 284). Every livingg room had a hearth for cooking, c heatting, and ligh ht if there were no keroosene lamps (Watson ( 1979, 122–3).

Fig.. 2.5. Hasanabaad hearth plan an nd section (Waatson 1979, 124)

The older w women usuallyy collected an nimal dung, w which was miixed with water and ussed as fuel in households. Almost eveery living rooom had nich hes for storaage. Items were w also suspended ffrom woodenn pegs or iro on nails. If tthere were lo ow stone benches theey were usedd as storage spaces s (Watsson 1979, 183 3). Grain storage pits often occupiied a corner of the living rooms (see Fig. F 2.6). Watson repoorted that thee niches weree nearly identtical to those found in

Methodo ology

43

early villagee sites, such as Jarmo in northern Iraqq or Fayum in Egypt (Watson 19779, 124–6).

Fig. 2.6. Hassanabad storagee pit (Watson 19979, 126)

Cooking faccilities could also a be presen nt outdoors in the courtyard d (Watson 1979, 156). Watson conssidered the general arrangem ment of the household h complex andd numerous other o materiall culture featuures as similaar to their ancient counnterparts. In particular, p roo oms grouped aaround a courrtyard are present at prrehistoric sitees such as Jeriicho, Hassunaa, and Tell ess-Sawwan (Watson 19779, 160). Watson alsoo gives a detaailed descriptiion of the dom mestic equipm ment. For example, am mong the utennsils there waas the saj, an iron plate forr cooking bread (Watsson 1979, 161)).

44

Chapter Two

Fig. 2.7. H Hasanabad metal plate (saj) fo or cooking breadd (Watson 1979 9, 162)

Every housee had one or more m kanu plaaced in the livving room or the t foyer. These were chests or binss the women made m by mixinng mud with chaff and used as conttainers for flour, grain, or other o foods (F Fig. 2.8) (Watsson 1979, 162).

Methodo ology

45

Fig. 2.8. H Hasanabad flour storage chest (kenu) with openning stuffed wiith a rag (Watson 19 979, 167)

Houses werre generally empty, the only o piece oof furniture being b the wooden cheest that each woman receiived when shhe married. Sh he would keep the keeys to it (Waatson 1979, 171). Occasionnally, in livin ng rooms there were w wooden benchhes used as a storage placees for bedding g (Watson 1979, 172). Watson founnd examples that were “vvirtually identtical with prehistoric oones excavateed at Jarmo and Karim S Shahir” (Watsson 1979, 284).

46

Chapter Two

(a)

(b) Fig. 2.9. (a) Ground plan p of house (W Watson 1979, 1 29); (b) Flowch hart

Methodo ology

(a)

(b) Fig. 2.110. (a) Ground plan p of house (W Watson 1979, 1 33); (b) Flowcchart

47

48

Chapter Two

(a)

(b) Fig. 2.111. (a) Ground plan p of house (W Watson 1979, 1139); (b) Flowchart.

Methodo ology

(a)

(b) Fig. 2.112: (a) Ground plan p of house (Watson 1979, 1142); (b) Flowcchart

49

50

Chapter Two

(a)

(b) Fig. 2.113. (a) Ground plan p of house (W Watson 1979, 1145); (b) Flowcchart

Methodology

51

In Hasanabad, for example, the ratio between space and people for an average-roofed dwelling was an area of 7.3 m2 per person, stable excluded (Watson 1979, 291). There was an average of 3.2 people per living room (Watson 1979, 296). Evidence of wealth was in the size of the houses and their aspect. More affluent families had bigger, better-kept houses, sometimes with two living rooms (Watson 1979, 296). Watson summarises the general characteristics of the different rooms: foyers were the smallest rooms; living rooms always had a hearth, wall niches, pegs for storage, household equipment, furniture, and bedding; storage rooms were featureless or contained unused equipment; and stables generally had mud mangers (Watson 1979, 295). From the social point on view, Watson notices that social solidarity and collaboration took place only among the families of brothers and people of the same economic conditions (Watson 1979, 230).

2.4.3 Horne’s Study of Village Spaces Horne had the chance to study the village of Baghestan in the Khar/Turan area of northeastern Iran during three summers between 1976 and 1978 (Horne 1994, 1–11). The village shared similar characteristics with the Iranian villages studied by Kramer and Watson: small sizes, agricultural and pastoral economies, Islamic religion, and Iranian culture (Horne 1994, 3). Horne argues that the pastoral-agricultural technology appears to have been part of Near Eastern settlements for at least eight thousand years (Horne 1994, 8). Baghestan was at the time of Horne’s research a small village of one hundred and fifty residents (Horne 1994, 9). The climate in the region is arid, adequate for a pastoral economy but without sufficient precipitation for dry farming (Horne 1994, 14). At the time of the study, the majority of the inhabitants were herders or farmers who owned the land, animals, and tools (Horne 1994, 35). Both men and women worked in the fields and with the animals, while a small percentage were part-time specialists of craft activities (Horne 1994, 35– 6). Horne visited the thirty-seven villages of the Tauran plain and calculated that half of the buildings were occupied by humans and the other half by animals and agricultural equipment (Horne 1994, 58–9). Buildings in Baghestan were made of individual rooms opening directly onto courtyards or alleys (Horne 1994, 90).

52

Chapter Two

Horne classified four basic types of room: (1) Living rooms used for eating, sleeping, entertaining, tea-making, some cooking, and miscellaneous non-commercial craft activities (2) A household storeroom for cooking, food storage, domestic equipment, and clothing (3) A storeroom for agricultural equipment, firewood, or straw (4) A stable for animals and agricultural equipment A few houses had an additional small kitchen-storeroom (Horne 1994, 90). A large percentage of house space was used as a workplace and another of the most important functions of a building was that of shelter for people, animals, activities, and equipment (Horne 1994, 96). The door was often the only effective light source (Horne 1994, 99). Horne understood that the ideal house in Baghestan should be composed of: an evan, a family living room, a guest room, a storeroom-kitchen, another storeroom, an animal house, a tannour, a privy, and a shower (Horne 1994, 150). Often, houses were built gradually, starting with the living room and finishing with the courtyard wall and the gate. The order gives a sense of priority of the different spaces (Horne 1994, 150). Horne also noted that the minimum acceptable leaving quarters were composed of a living room, a storeroom-kitchen, and a forecourt (Horne 1994, 151). The measures of wealth were land and water, the number of animals, the number of rugs, the total number of rooms in a house, and the size of the courtyard (Horne 1994, 152–60). The size of a house was not related to the number of children and a living room area was not in correlation with the wealth of the household (Horne 1994, 158–9). Living rooms in different houses were in fact similar in size because too-large rooms were too cold in winter (Horne 1994, 160). Throughout most of the rural Near East rooms are multifunctional, but it is feasible to assign a general descriptive type (Horne 1994, 177): “at the very least, being able to identify the core dwelling unit—the living room— is an important step in identifying and characterizing households” (Horne 1994, 176).

Methodology

53

Remodelling to change a room’s function is a common practice (Horne 1994, 178). It is interesting to note that the use of space was subject to seasonal changes; for example, during the summer the majority of activities, including sleeping, took place outside and heaters were removed when outside hearths were used for cooking (Horne 1994, 98). During winters, draughty windows and roof vents ware bricked up and plastered (Horne 1994, 98). Living rooms could be seasonally alternated with storerooms (Horne 1994, 159). For the purpose of this project, it is also useful to look at the concept of privacy. For example, Horne reported that: “All space is shared in one way or another,” and that usually no one lives alone (Horne 1994, 92). Also: Personal or individual privacy in the sense of solitude is neither desired nor approved of in Baghestan. Wanting to be alone is regarded with suspicion. No matter its size, a family is not usually divided spatially for eating or sleeping, even when enough rooms are available. Access to certain areas, however, is restricted by gender or social relationship. Privacy can also be achieved without architectural barriers. Without leaving the room, a woman who wishes to sleep can remove herself from the presence of others by covering her face and body with the chador (long veil). (Horne 1994, 119)

Horne also notes that all the members are sleeping, eating, and entertaining in one living room (Horne 1994, 159), and that it has been reported that the number of living rooms in a house is not related to the number of children in the household. There is also the fact that courtyards are the most public places (Horne 1994, 159) but that it is not important to enclose them with walls (Horne 1994, 150). It was mentioned above that building walls and a gate for the courtyard is at the bottom of the list of priorities.

54

Chapter Two

(a)

(b) Fig. 2.14. (a) (Horne 1994, 93); 9 (b) (Horne 1994, 205)

Methodo ology

(a)

(b) Fiig. 2.15. (a) Groound plan (Horn ne 1994, 204); (b) Flowchart

55

56

Chapter Two

2.4.4 Seeden’s Analysis of Syrian Villages Helga Seeden made ethnoarchaeological observations in Syrian villages between 1980 and 1985 in the area where there have been villages since the Halafian culture seven thousand years ago (Seeden 1985, 289–303). Seeden claims that “basic similarities in living standards and food processing … are apparent. Today’s villages yield evidences of continuities in architectural methods and food production …” (Seeden 1985, 291) In Seeden’s work, the description of the inbuilt domestic structures and features is interesting; for example, she reports that courtyard work platforms are usually built in front of kitchens or that stone benches were built in cool areas of rooms or courtyards to work, rest, or sleep (Seeden 1985, 294). On the other hand, the movable items present in living/sleeping rooms were perishable objects such as mats, baskets, rugs mattresses, and cushions (Seeden 1985, 296). Stone niches or low wooden benches were storage places for bedding (Seeden 1985, 297). Seeden notes that, in contemporary houses, baskets, trays, craters, etc. were distributed over courtyards and terraces alike (Seeden 1985, 297).

2.4.5 Villages in Jordan Biewers analyses three Jordanian villages: Smakieh, ’Aima, and Khirbet (Biewers 1995, 30). The village of Khirbet was occupied from 4000 BC by four tribes of Nabataean Muslims. The buildings in the village were of dry stone and the development started in 1934 (Biewers 1995, 31–3). The village of ’Aima is in a region where dry agriculture is possible and the development started in the sixteenth century (Biewers 1995, 35). Generally, there are houses of one room in the village (Biewers 1995, 39). The village of Smakieh was founded in 1908 and inhabited by two Christian tribes (Biewers 1995, 35). The space of the village was divided into thirty identical parts each assigned to a clan and the settlement in respect of the tribal structure.

Methodology

57

The dwellings in Smakieh are composed of several rooms, and belong to vernacular architecture because their structure is not built with the help of an architect (Biewers 1995, 43).

2.4.6 Cafer Höyük In 1997, Aurenche conducted ethnoarchaeological research in the Kurdish village of Cafer Höyük (Aurenche et al. 1997). Although the geographical area is different, Aurenche describes a culture very similar to Kurdish village culture in western Iran and northern Iraq, with the same economic system based on animal breeding and agriculture, and a similar way of building houses of clay with some wood and stone when available (Aurenche et al. 1997, 61–79). Rooms are no longer than 3.50 metres because this is the maximum beam length, with an average of 16 to 18 metres2 (Aurenche et al. 1997, 81, 93). Rooms are described as being plastered in white or sometimes blue with no decorations, with only carpets or mats in the reception and living rooms of the richest households (Aurenche et al. 1997, 82). In Cafer Höyük houses, storage in living rooms is provided by niches that are sometimes closed with a wooden door and benches, and the latest can also be utilised to sit on according to Aurenche (Aurenche et al. 1997, 87–9). Aurenche reported a use of space similar to the previous descriptions: a life that takes place on the floor, the same space utilised for sleeping and eating. He noticed that those activities happened in the room closest to the external door to access the light. If the cooking facilities were in a separate room, this was closest to the external door, and sleeping took place in a more internal room (Aurenche et al. 1997, 95). Storage rooms were farther from the entrance. Animals shared were kept in a room far from the entrance or on the ground floor in the rare case of a double-storey dwelling (Aurenche et al. 1997, 97, 106). Courtyards are not common but many activities took place outdoors in the areas between houses (Aurenche et al. 1997, 97–8). Aurenche recognises three basic functions of houses: human shelter, animal shelter, and storage.

2.4.7 The Kurdish Woman’s Life In 1957, Henny Hansen spent four months among the Kurds in northern Iraq in villages and a town in the Little/Lesser Zab river area to investigate Kurdish ethnography and study women’s cultural patterns. Hansen, as a

58

Chapter Two

woman, hadd the advantaage of access to areas foorbidden to men m in a Muslim envvironment (Haansen 1961). Hansen is off the opinion that, t “the Kurrds possess a cculture layer older o than Islam, which become eviident at the most m importannt events in liife: birth, marriage andd burial” (Hannsen 1961, 132). Reading Haansen’s accouunt, it appearss that many aspects of th he Kurd’s material cullture are simillar to those off the populatiions living in the same area five millennia earlierr. The Kurds in 1957 were living l in houses built of muud or baked brricks with the occasionnal use of stoone in areas where w it was aaccessible. Th he houses were inwardd-facing with few openingss towards the outside in thee form of doors, and ooften an absennce of window ws or the preseence of small windows placed in thhe upper part of the walls. The rooms w were sparsely furnished (Fig. 2.19). The furniturre consisted of beds (gennerally in the form of mattresses), benches, sm mall tables, and occasionally ly wooden ch hests. The majority of tthe objects present in housees were placedd in niches in the walls or hung on walls with naails. There waas a wide usee of carpets during d the cold season,, while duringg the warm seaason the clay of the floors was w bare. The flat rooffs were reacheed with woodeen ladders (Fiig. 2.19).

Fig. 2.19. Wooden laddeer leading to roo of and inside off a house in Top pzawa (H Hansen 1961, 35 5, Figs. 25–26)

Grain was sstored at hom me in clay silo os placed in sstorerooms. The T wheat was crushedd by water miills or in a rev volving grindder that looked d like the

Methodo ology

59

grinding stoones found in archaeologiccal excavationns in sites datting from the Neolithicc onwards. Baking tookk place in what Hansen describes as “a jar-shaped eaarth oven, tandur, or onn an iron sheeet, tawa or sajj which is not a fixed installlation but a loose utensil” (Hansen 1961, 1 43–5). The list of material charracteristics off Kurdish lifee recorded by y Hansen correspondss with the material rem mains retrieveed in archaaeological excavations of sites in Mesopotamia, M to t the point w where the abo ove words can be usedd to describe dwellings frrom those sittes dating to the third millennium BC. The preccious ethnograaphic record of the use off material objects and the way of liffe that used to o take place inn those housess can help archaeologissts to formullate hypotheses of the pa st use of objjects and dwellings foound in archaeeological contexts. Hansen anallysed four houuseholds from four differentt environments. Village aristocracy iinhabited the first f household d, which contaained thirteen to fifteen people depennding on the season. The second s belongged to a family y of three ordinary peaasants. The thiird household was in an urbaan context inh habited by a merchant aand his familyy, with a totall of fourteen ppeople. The fo ourth was prepared forr an urban educcated couple engaged e to be m married.

Fiig. 2.20. Sheikhh Taifur’s housee (Hansen 19611, 29, Fig. 21)

60

Chapter Two

(a)

(b) Fig. 2.21. (a) Ground plan of Sheikh Taifur’s house (Hansen 1961, 33, Fig. 24); (b) Flowchart

Methodology

61

Figure 2.22. Bird’s-eye view of Sheikh Taifur’s house (Hansen 1961, 30, Fig. 22)

From Hansen’s recording of the use of space in those four houses, common patterns appear despite the significant difference in house size, the number of inhabitants, and the social stratum to which those inhabitants belonged. There is an extreme flexibility in the use of rooms within a twenty-fourhour period as well as a seasonal flexibility.

62

Chapter Two

Fig. 2.233. Elevation of Sheikh Taifur’s house (Hanseen 1961, 31, Fig g. 23)

Methodo ology

63

(a)

(b) Fig. 2.24. (aa) Ground plan of o a family hou use in Sulaimanii (Hansen 1961, 40, Fig. 30); (b) Flo owchart

Sleeping andd the consum mption of food could occur in the same sp pace. For the two largger households of thirteen to t fifteen peopple, Hansen comments c that: “The oonly possible solution of the problem of housing so o large a number of ffamily membeers in a home with relativeely so few roo oms, is to use them in common bothh for meals an nd for sleepinng” (Hansen 1961, 1 42). During the warm season, sleeping too ok place on rooofs, while during d the cold seasonn mattresses and a bedding were rolled oout in the areea of the house in whhich one hadd decided to sleep—only s sstorage roomss and the guest room w were not occuupied by the family. f All thee family memb bers slept together wiithout distingguishing betw ween nuclearr families orr gender. Privacy wass not generallly sought, and d people sleppt in their day y clothes. Occasionallyy, a married couple c retired to a different room for priv vacy.

64

Chapter Two

(a)

(b) Fig. 2.25. ((a) Ground plann of a house for a newly-marrieed couple in Su ulaimani (Hansenn 1961, 42, Fig. 31); (b) Flowcchart There waas no fixed rule as to which roo om or part of thhe house was to o act as dining room or bedroom m. The spot wh here the meal ttray—one for several s persons, w who continuallyy changed theirr grouping—waas placed on thee floor was the ddining room foor the short spaace of time it took to consum me the meal. (Haansen 1961, 35))

The preparaation of food took t place in the courtyardd and in the heywan h (a three-walledd room with one side opeen to the inteernal courtyarrd), while cooking andd baking faccilities took place p in a rooom reserved d for this purpose in tthe two biggeer houses, in the t courtyard in the schoollteacher’s house, and iin the loggia in the peasan nt house. The meat was preepared on the clay flooor of the dwelllings, and vin ne leaves werre folded arou und lumps of rice and pplaced in the cooking pot on the floor. In the Sheikh h’s house, meat and riice where rinssed in the basin in the garrden where th he dishes were washedd after meals.. “What we un nderstand as kkitchen work was thus distributed oover several places p and no ot assembled iin one room or at one spot” (Hanseen 1961, 34).

Methodology

65

The guest room was only used when entertaining guests on special occasions. This room was an essential part of a Kurdish house, and was always present even in small houses such as a schoolteachers’. Only the poor peasant’s dwellings did not have such a room. The guest room was necessary to maintain the Islamic custom of entertaining male guests in a room near the entrance to avoid any sight of the women of the house. The house was built taking into account this necessity to the point that family space was sacrificed. A storeroom was a constant presence in any dwelling. The four examples described by Hansen all have at least one storeroom. Also, in the tiny peasant house the only room was assigned for storage purposes during the summer. An open space was a constant presence. This took the form of an internal courtyard for the bigger houses, or a loggia in the front in the case of the poorest dwellings. All four houses described also had a latrine, which was a dedicated room in the three biggest houses and in the goat stables in the peasant dwelling. Running water was available only in the Sheikh house and the merchant house, which were also equipped with bathrooms. The inhabitants of the two smallest dwellings shared water facilities with neighbouring households. In Hasanabad were the remains of two old mills that used to be employed to ground cereals (Watson 1979, 36). After considering the evidence from the modern settlements across the Near East it is possible to make some observations. There are differences in the archaeological data from the ancient sites taken into consideration for this thesis. The first difference is that the majority of the ethnographic evidence exposed here is comprised of studies of rural settlements, while the seven case studies are of ancient cities, as is the majority of the evidence of the third-millennium dwellings that have been excavated. In this kind of rural settlements, the majority of the households sustained themselves with the products of the land. Houses were designed to store agricultural products, to care for animals, and to transform the agricultural and animal products. In third-millennium Mesopotamian urban settlements, however, there was a higher degree of social stratification. This implies differentiation in the work. Households were not self-sufficient. People purchased many products while they engaged in specialised activities, for example crafts, or they may have been part of the palace or temple hierarchical system, or land owners

66

Chapter Two

(Liverani 1988, 182–3; Postgate 1992; Pollock 1999, 117). As a consequence, the layout of their dwellings was different. In fact, Watson, Kramer, and Aurenche use the ethnographic data to study prehistoric societies that were less complex and more similar to those modern villages. Hansen’s study, on the other hand, although not carried out with archaeology in mind, also describes houses and households belonging to a more specialised environment. She can therefore provide a different model to be tested against the archaeological evidence of a society more complex than the prehistoric one. Moreover, she gives more details about the use of space in everyday life. For example, where the preparation of specific foods takes place or how sleeping and eating patterns are constantly changing. We have described a lifestyle conducted with a material culture similar to the one dated to the third millennium BC, retrieved from excavations of sites in the same area. If the life in households in the third millennium BC was as flexible as the Kurdish one, it does not make sense to try to identify a kitchen, a dining room, or a bedroom from the house plans found in the archaeological excavations. If the use of space in the third-millennium Mesopotamian houses was similar to the Kurdish model of life we can look for: (1) One or more storage rooms that might be in an area relatively less accessible compared to other rooms and that may or may not have windows (2) Cooking facilities in a dedicated room in an open space or a multipurpose room (depending on the size of the house) (3) Spaces different from those previously listed for the members of the household to sleep, eat, and work. During the warm season, this space could be an outdoor area such as a courtyard or the dwelling’s roof. When it was indoors, the majority of the literature calls it a living room (4) A latrine or bathroom facilities may be present It appears that the minimum requirement is one room for people and one for storage. Additionally, an open space was a constant. It took the form of an internal courtyard or a loggia in the front of the dwelling or the space in-between houses. The majority of the household activities, such as eating, sleeping, or the preparation of food, took place in the outdoor space and on the flat

Methodology

67

roofs. Those outdoors spaces were an extension of the house and must be considered as part of the living accommodation. Kramer’s study reveals an average of ten metres2 of roofed area per person for an unfortified rural village, the same figure proposed by Naroll’s crosscultural estimations (Naroll 1962). From Watson, a figure of 7.3 metres2 of roofed dwelling area per person is estimated (Watson 1979, 291). This kind of estimation is very approximate because it is based on assumptions such as a population of between four and eight on average in a household, and fifty percent of the ground floor of roof surface available, as well as the fact that it is not based on solid data (Schloen 2001, 168– 71). This number can be compared with Stone’s estimation of five to six metres2 per person based on her analysis of the textual and archaeological data from Old Babylonian Nippur (Stone 1981, 32). Schloen considers this figure too low and proposes eight to ten metres2 per persons also using the estimation of 6.7 per person per household based on demographic Mediterranean data (Schloen 2001, 168–75). Postgate in 1994 attempted quantitative estimates of persons per square metre and persons per household on the basis of the evidences of Abu Salabikh, trying to avoid ethnographic parallels. He worked with the figure of between four and seven metres2 per person, which implies an extremely wide margin of error, arriving at the number of “between 248 and 1205 Sumerians per hectare” (Postgate 1994, 51–64). Generally speaking, is better to have a wide margin of error since it accounts for more uncertainty. From Hansen’s account it appears that Kurdish culture did not know privacy as understood by the contemporary Western world. Hansen describes sleeping as “a family affair” (Hansen 1961, 41), meals where prepared by the women co-operatively, and even latrine facilities were used by many women at the same time. During the warm season, when sleeping took place on the roofs, the usual seclusion of the household women was impossible; in towns and often in villages, houses were attached one to another, and the roofs were separated only by mats. Hansen noted the striking contrast between the complete seclusion of the ground floors of houses with the lack of privacy on the roofs.

68

Chapter Two

The only private space for a Kurdish women in the 1950s was not a room but a chest, which often came as part of a bride’s dowry; she kept the key attached to her clothes. Men had even less privacy. They spent the majority of time outside of their houses in the streets, in coffee shops, or in their work places (fields if peasants, souk if merchants, etc.). They came home only for meals and sleeping. A son was given a room of his own for the first time when he married. He shared it with his bride until the birth of the first child, then, according to Hansen, the couple integrated the family pattern of use of space: the woman took her place among the other women, the man with the other men. If the same behaviour took place in ancient Mesopotamian society, it might challenge the widespread concept in the literature of the correspondence between the number of living rooms and the number of nuclear families. The presence of a guest room immediately accessible from the outside, allowing women to remain unseen by male visitors, seems to be a very Islamic characteristic. The information gathered from the ethnographic data will be used in the following chapter to build interpretative models of the use of space in third-millennium Mesopotamia.

CHAPTER THREE NORTH REGION CASE STUDY: TELL MELEBIYA, TITRIS HÖYÜK AND TELL TAYA

The methodology discussed in chapter two is applied here to three midlate third-millennium sites where houses have been excavated: Tell Melebiya, Titris Höyük, and Tell Taya. These three cities have been chosen because they are among the best remains of domestic architecture in the region, and in particular are examples of complete house plans from the same period (the second half of the second millennium BC). In this sense, the evidence is comparable. As stated in chapter two, there are a number of limitations in studying houses from this period in Mesopotamia given the amount of evidence suitable for analysis. Some issues are: the uncertainty of the existence of an upper storey and the necessity for a complete ground floor plan to carry out the spatial analysis, which is rare because of difficulties in preservation and sometimes poor excavation, or little interest in domestic quarters by excavators. These problems reduce the amount of evidence available for study often to only a few units per site. The first site taken into consideration is Tell Melebiya, where there are eight houses that present a complete plan of the ground floor. The majority of them do not present differentiation in the phase of construction, which questions the accuracy of the excavation (Lebeau 1993). In Titris Höyük, several thousand square metres of domestic buildings have been uncovered, but only four present an almost complete plan suitable for analysis. The houses seem well excavated and information from a micro-debris based functional analysis is present in the report for one house (Matney, Algaze, and Rosen 1999; Algaze et al. 2001). The third site is Tell Taya, in which only public buildings and incomplete private houses have been excavated. However, the site has a very large

70

Chapter Three

quantity of surface remains that have been recorded to prepare a plan of the ancient town including the layout of houses and streets. Nine houses have been selected for analysis in section 3.3 among the ones with more clearly visible walls (Reade 1968; 1971; 1973). As has been said in chapter one, the north region of Greater Mesopotamia had very cold, wet winters and the rainfall allowed agriculture without the need for irrigation. From the point of view of the building material, it is important to point out the presence of timber and stone together with the clay bricks utilised in the rest of Mesopotamia.

3.1 Tell Melebiya Tell Melebiya is an Early Dynastic III (ca. 2500–2300 BC) site located in the Khabur region south of Hassake in northeastern Syria (Fig. 3.1). A portion of the site has been eroded by the river that was flowing nearby, damaging a considerable part of the tell (Fig. 3.2). The site was excavated by a mainly Belgian team under the direction of M. Lebeau between 1984 and 1988. The emergency excavation was due to the construction of a series of dams and the consequent flooding of the region (Lebeau 1993, 13).

North Region Case Study: S Tell Meleebiya, Titris Hööyük and Tell Taaya

Figg. 3.1. Third-milllennium Upperr Mesopotamiaa (Lebeau 1993))

71

72

Chapter Three T

Fig. 3.2. Tell Melebiya (Lebeau 19996)

North Region Case Study: Tell Melebiya, Titris Höyük and Tell Taya

73

The team exposed the remains of an Early Dynastic provincial town characterised by an intense urbanisation with coherent blocks of dwellings crossed by narrow streets (Lebeau 1993, 48). The tell was probably rounded in shape with radiant urbanisation. The town was built on terraces and as a consequence the streets had slopes that efficiently drained rain and waste water. The archaeological investigations carried out at the site exposed residential areas while no administrative or religious buildings were found in the excavated part of the tell. Infant burials were found in some houses, yet no adult burials or cemeteries were exposed (Lebeau 1993, 41–6). For this analysis, only the eight houses with a nearly complete plan have been considered. As shown on the plans below, these houses have a nearly complete ground floor layout with occasional uncertainty regarding some door positions (see Figs. 3.4[b], 3.6[b], 3.7[b], 3.8[b], 3.10[b], 3.11[b], 3.12[b], 3.13[b]).

74

Chapter Three T

Fig. 3.3. Tell Melebbiya: excavated area Chantier B (Lebeau 1996 6)

For each off the eight houuses chosen, a flowchart baased on the Hillier H and Hanson metthod (see chappter two) has been drawn. Moreover, a bar chart has been creeated in orderr to compare the surface arreas of each room r in a house and to represennt the featurees found inn each room m by the archaeologisst. For practiccal reasons, a number was arbitrarily assigned to each room.

North Region Case Study: S Tell Meleebiya, Titris Hööyük and Tell Taaya

3.1.1 Hou use B1

(a)

(b) Fig. 3.4. (a) Plan of Houuse B1 (Lebeau u 1993); (b) Flow owchart of Housse B1.

75

Chapter Three

76

Fig. 3.4 shows that House B1 in Tell Melebiya was the largest building excavated at the site (203.6 metres2). It was composed of eighteen spaces with two entrances from the outside, space 2 and 5, and another possible entrance in space 6. House B1 25

surface area (m2)

20

Surface

15

Hearth(s) Bench(es) Basin(s) oven

10

platform jar

5

0 16

7

1

3&4

5

15

14

10

13

2

11

12

17

18

8

9

6

room number

Fig. 3.5. Internal space sizes and features of House B1

Fig. 3.5 shows the surface areas of each space and the fixed features present. Room 16 was the largest with a surface area of over twenty metres2. This space had the characteristics of a multipurpose room where the people could eat, sleep, cook, work, etc. Because it could accommodate many people, it was positioned at the opposite side of the main entrance and was originally equipped with three low benches alongside the bases of the western, southern, and eastern walls (Lebeau 1993, 91). Spaces 7 and 1 were larger than fifteen metres2. Two domestic ovens and a trapezoidal platform were found in space 1, which was a large space positioned near what seems to be the main entrance. A wellplastered floor has been found in room 3/4, equipped with a semi-circular basin in the northwestern corner and a rectangular basin on the northern limit separated from the rest of the room by a sort of bench (Lebeau 1993, 91). Space 7 was located at the centre of the dwelling and in an ideal place for an internal courtyard; however, Lebeau, in his last analysis, excluded the presence of courtyards in Melebiya houses, although he considered open courtyards in his previous reports. If space 7 was covered, it remains

North Region Case Study: Tell Melebiya, Titris Höyük and Tell Taya

77

unclear how it would take in light. It can be hypothesised that a higher roof covered spaces 7, 8, and 11 to allow for the existence of windows in the higher section of the walls in order to let light in. The presence of a basin with two conic depressions at floor level in space 7 suggests domestic or craft activities and also the need for light. The other spaces in House B1 were smaller than five metres2. In space 18 a basin and a low bench were found, and the room led to another small space (17) where no features were found (Lebeau 1993, 91). On the other side of room 16 a similar arrangement was found: space 15 was connected to space 16 and a low bench was found on its eastern wall. The room also had a niche along the southern wall, near the passage connecting with space 16, and a fireplace (Lebeau 1993, 91). Another space (14) was arranged in a position symmetrical to the arrangement of spaces 17 and 18 on the other side of room 16 (Lebeau 1993, 91–2). Spaces 3/4, 7, 10, and 18 had basins, and their presence can be explained by craft activity carried out in the building. In fact, there seems to be too many for only household necessities, and this can also explain the need for more than one entrance, useful in correspondence with workshops or commercial activities. Spaces 3/4, 5, 15, 16, 18, and 9 had benches. In space 9 a low bench was found, as well as a jar in the floor, and therefore the space may have been used as a storeroom because of the presence of these features and its relatively private position. From the flowchart it appears that spaces 14 and 17 were less accessible, and it was necessary to pass through other five spaces to reach them. The analysis of the circulation also highlighted that the southern sector had a more private character, being accessible only by room 11 (Lebeau 1993, 92). Space 11 does not seem to be connected with any other room. Ethnographic models can be useful for formulating hypotheses and they attest to the existence of storerooms that were periodically opened or closed according to the season. Perhaps we are looking at a similar situation. Lebeau noted that the majority of the floors were covered in white plaster with the exclusion of the floor in space 1 (Lebeau 1993, 91). Specifically, plaster was well preserved in rooms 3/4, 6, 10, 17, and 18; only partially preserved in rooms 11 and 13; and almost absent in all the other spaces (Lebeau 1993, 91).

Chapter Three T

78

3.1.2 Hou use B2

(b)

(a) House B2a B 25

surface area (m2)

20

Surface 15

altar basin bench cooking pot drain jar

10

oven(s)

5

0 2

1

4

6

8

7

3

5

space number

(c) Fig. 3.6. (a)) Plan of Housee B2 (Lebeau 19 993); (b) Flowcchart; (c) Intern nal space sizzes and featuress of House B2a

North Region Case Study: Tell Melebiya, Titris Höyük and Tell Taya

79

House B2 was a building composed of eight spaces, with only one entrance to the outside in space 1. Fig. 3.6(c) shows that space 2 was the biggest with fifteen metres2 of surface area. Three low benches (thirty to forty centimetres long and twenty centimetres high) were found along the northern, southern, and eastern walls. The southeastern corner was occupied by a basin with a curved edge covered with plaster that was coating the basin and the base of the wall up to seventy-five centimetres. A peculiar structure has been found in this space against the western wall: an arrangement of bricks covered in plaster decorated with niches that Lebeau describes as being similar to an altar. The arrangement of the altar benches and basin suggests it could have been a cultic space used to gather and conduct religious activities like a domestic chapel. As for House B1, the majority of the spaces were small rooms of less than five metres2, organised to the side of the building. Space 4 was equipped with a bench at the base of its western wall and also had a low transversal wall that limited the southern side of a very wellplastered floor. A jar was embedded in the floor near the southeastern corner of the room (Lebeau 1993, 92). In space 6 was a plastered basin occupying all the available space, which suggests that this room was used for craft activities. In room 3, a domestic oven and a cooking pot have been found (Lebeau 1993, 92). Room 5 was equipped with a big domestic oven (Lebeau 1993, 92). Room 7 was the smallest, being accessed from space 2 and partially blocked by a reservoir. A transversal bench oriented west-east was dividing the room, and at the base of the bench a well-plastered drain was found (Lebeau 1993, 92). Lebeau suggested that all the floors were originally plastered although few traces remain, found only on the floor in spaces 4, 6, and 7 (Lebeau 1993, 92). In a later phase, space 1 was divided into three spaces because of the erection of a wall.

80

Chapter Three T

Room 8 couuld have beenn a storage space becausee of its positio on in the layout and itts small dimennsions.

3.1.3 Hou use B4

(a)

North Region Case Study: Tell Melebiya, Titris Höyük and Tell Taya

81

(b) House B4 25

surface area (m2)

20

Surface basin bench drain hearth(s) oven(s)

15

10

5

0 8

7

9

10

3

12

11

2

1

6

4

5

space number

(c) Fig. 3.7. (a) Plan of House B4 (Lebeau 1993); (b) Flowchart; (c) Internal space sizes and features of House B4

82

Chapter Three

House B4 was a building with a surface area of 136,3 metres2 composed of twelve spaces with only one entrance that connected space 1, a narrow corridor, with the outside. The corridor was provided by a canalisation built of large stones sloping towards the entrance and the street (rue Koldewey). The doorways leading to spaces 2, 3, 4, and 5 were not found because of erosion, but Lebeau supposes that their presence was necessary to be able to move around the building (Lebeau 1993, 93). The bar chart in Fig. 3.7 shows that spaces 8 and 7 were the only ones with a surface area larger than ten metres2. The majority of the other spaces were between five and ten metres2. Spaces 6, 4, and 5 had a surface area less than five metres2. Space 2 was provided with a plastered basin at the base of the northern wall and another at the base of the southern wall, and a low bench was positioned on the northern wall on the right side of the doorway leading to space 3 (Lebeau 1993, 93). There was a hearth at floor level. Space 4 was a small room with a domestic oven. Space 3 was equipped with two basins at floor level, one in the northeastern corner and one more central, both plastered. Two benches were positioned along the walls, one on the northern wall and one on the eastern wall (Lebeau 1993, 93). Space 8 was a central area of distribution of the circulation, and Lebeau considers the space to have been a roofed room. On the floor was part of the canalisation that was continued towards corridor 1. Along the eastern wall a bench has been found (Lebeau 1993, 93). Space 7 was a large room with a floor coated with several layers of plaster. The floor was disturbed by a large pit. A first basin at the floor level had a curved edge and was well plastered, and also had a layer of bitumen. Adjacent to the first basin was a square arrangement of bricks, and a second basin was placed on the outside of this arrangement in the southeastern corner of the room. Four benches were present along the walls (Lebeau 1993, 94). From room 7 it was possible to reach room 10, which was equipped with a bench at the base of the northern wall, continuing southwards and dividing the space into two parts. On the western part was a small wall oriented

North Region Case Study: S Tell Meleebiya, Titris Hööyük and Tell Taaya

83

west-east. A semi-circullar basin wass found towaards the eastern wall, which was pplastered and coated c in bitum men (Lebeau 1993, 94). As is clear ffrom the flowcchart, the norttheastern sectiion of the building had a more secluuded characteer. It was com mposed of roooms 9, 12, and d 11, and was accessibble only by rooom 6. Benches havve been founnd in spaces 8, 8 7, 9, 3, annd 2, while basins b are present in sppaces 7, 9, 3, 12, and 2. Th he presence off several basin ns can be explained byy the possibiliity of craft acttivities that occcurred in the building, as stated earrlier for House B1. Drains were w also preesent in spacess 8 and 1. Rooms 5 annd 11 do not have any feattures and migght have been n used for storage baseed on their sm mall dimension ns and positionn in the housee. Space 8 is positionedd at the centree of the buildiing surroundeed by other ro ooms and, if covered, would have had h the same lighting probblems as discussed for rooms 7, 8, aand 11 in Houuse B1.

3.1.4 Hou use B7

(a)

Chapter Three

84

(b) Fig. 3.8. (a) Plan of House B7 (Lebeau 1993); (b) Flowchart House B7 25

surface area (m2)

20

surface (m2) altar basin bench drain oven(s)

15

10

5

0 15

12

1

17

11

3

2

8

10

5

13

14

6

9

4

7

space number

Fig. 3.9. Internal space sizes and features of House B7

16

North Region Case Study: Tell Melebiya, Titris Höyük and Tell Taya

85

House B7 was a very large building of 184.5 metres2 composed of seventeen spaces with two entrances on the same side of the building. The bar chart in Fig. 4.8 shows that spaces 15 and 12 are the biggest with a surface area between 17 and 19 metres2. The remaining spaces are of less than ten metres, with the majority less than five metres2. Space 1 was peculiar, being in the shape of a three-sided corridor that connects the two entrances, the small side rooms around it, and the central spaces 2 and 3. Two stone drains were found connecting two corridors with the outside. As has been explained, Lebeau reported an absence of internal courtyards in Melebiya houses and space 1 should be considered a roofed central space. Rooms 5 and 7 were equipped with small basins (Lebeau 1993, 95–6). In room 9, a domestic oven was found. Room 1 was surrounded by three sides to a central area composed by rooms 2 and 3. Room 2 had a doorway on the eastern side, and a bench was found in the room along the western wall. Room 3 was in the northern side of room 2, and an L-shaped bench was found along the eastern and southern walls (Lebeau 1993, 96). The central space 11 was badly disrupted by pit n. 2004, leading to the northern part of the building. Room 15 was equipped with two benches along the eastern and western walls. Lebeau considers them to be a protection from dampness (Lebeau 1993, 96). Room 15 communicated with a small trapezoidal space that has the characteristics of a storage room and that Lebeau considers to have been obtained from room 15 in a later phase (Lebeau 1993, 96). Room 10 had a bench along the southern wall, while room 13 was also provided with a bench along the southern and eastern walls and was connected to room 14 (Lebeau 1993, 96). Room 14 could have been a storage space because of its position in the layout and small dimensions. One opening in room 1 would also probably be the connection with room 12, which as we saw earlier was one of the largest rooms in the building. A bench was found in the room along the western wall and continued in a L-shape along the southern wall. A structure, interpreted by Lebeau as an altar with niches, was found in the northeastern corner. The structure was

86

Chapter Three T

identical to the one founnd in buildin ng B1 and waas plastered with w care Lebeau (Lebbeau 1993, 966). Space 12 waas connected to space 17, which w had thee interesting feature f of an orthostat made of greyy slabs. Lebeau did not find the distribution d off objects of anny help toward ds further understandinng the room functions fu (Leb beau 1993, 96)). As previoussly discussed in i House B1, the need for m more than onee entrance might be expplained by thee presence of workshops orr commercial activities inside the hoousehold.

3.1.5 Hou use G1 House G1 ooccupied 127.4 metres2 an nd was compoosed of eleveen spaces with an entrrance to the outside o in spacce 2, a small plastered spaace, and a possible second entrance in space 9. The remainss of Islamic tombs disturbeed the floors of spaces 8, 9, 9 and 10 (Lebeau 19993, 99).

(a)

North Region Case Study: Tell Melebiya, Titris Höyük and Tell Taya

87

(b) House G1 25

surface area (m2)

20

15

Surface oven(s) bench basin jar

10

5

0 7&4

12

9

8

10

11

6

5

1

3

2

space number

(c) Fig. 3.10. (a) Plan of House G1 (Lebeau 1993); (b) Flowchart; (c) Internal space sizes and features of House G1

Only three spaces have a surface area larger than ten metres2 (spaces 7/4, 12, and 9). The space could have been a multipurpose room, although its position in the centre of the dwelling poses a question about access to

88

Chapter Three

daylight. It seems in the most likely position to be an internal courtyard, but Lebeau excludes the possibility of an internal courtyard at Tell Melebiya. Lebeau specifies that the floor of space 12 was not plastered and he supposes that all the other floors were plastered, with spaces 5 and 6 in better condition, with several layers of white plaster (Lebeau 1993, 99). On the floor of space 12, a tannour (domestic oven) was found. A bench was found in space 7, running along the base of the western wall, and a small domestic oven was near the bench. Near space 7 was what appeared to be an enclosed room without doors that might have been a storage room with a door that was periodically closed (Lebeau 1993, 99). Space 6 was a well-plastered room with a small plastered basin in the northeastern corner. Space 6 was connected with space 8, a room where no features have been found. Space 9 was accessible from spaces 12 and 8, where again no features were found (Lebeau 1993, 99). Rooms 8 and 9, because of their relatively larger dimensions and position at the back of the house, could have been used as living quarters for the household. The other spaces were smaller rooms of approximately five metres2 or less. Space 10 was accessed by space 9 and a large basin was found in it. As seen in the previous houses, the presence of spaces for craft activities in relatively small rooms at the side of the dwelling seems to be a common pattern. In room 3 a small wall of brick was found as well as a jar embedded in the floor (Lebeau 1993, 99). The remaining rooms were found empty, like room 1, a square space opening to the right of the entrance room. Some of them could have been used for storage based on their dimensions and positions in the building. From the ethnographic record, we saw that storage was one of the essential functions of houses. Lebeau reports that very little has been found in House G1 in terms of objects, as only a fragment of a clay chariot from room 11 is mentioned (Lebeau 1993, 99).

North Region Case Study: S Tell Meleebiya, Titris Hööyük and Tell Taaya

89

3.1.6 Hou use B6

(b)

(a) House B6 25

surface area (m2)

20

15 Surface oven bench(es) 10

5

0 1

4

3

2

space numb ber

(c) Fig. 3.11. (aa) Plan of House B6 (Lebeau 1993); 1 (b) Flow wchart; (c) Intern nal space sizzes and featuress of House B6

90

Chapter Three T

House B6 w was a small building with four rooms withh a surface areea of 37.2 metres2. Thee entrance was from space 1, a room thatt had a floor covered in tiles of muddbrick. Two parallel benches were fouund at the base of the northern andd southern walls w (Lebeau 1993, 94–5). Originally, the t house was compossed of three spaces, s and in n a later phasee space 1 waas divided and two room ms were creatted out of one (Lebeau 19933, 95). The small rroom 2 was occupied o by a domestic ovven and a hig gh bench along the eaastern wall (Leebeau 1993, 95). Space 3 waas a small rooom connected d to the entraance through a sort of corridor, whhich was part of space 1. Two short bencches were fou und at the northeasternn and southeasstern corners of o the room (L Lebeau 1993, 95). Lebeau hypoothesises that through room m 3 it was pos sible to access room 4, a space provvided with a low l bench alo ong the westerrn wall. Spacee 4 was a relatively biig room and was w probably the deepest rroom in the house, if a door was preesent betweenn rooms 3 and 4.

3.1.7 Hou use C2

(a)

(b)

North Region Case Study: Tell Melebiya, Titris Höyük and Tell Taya

91

House C2 25

surface area (m2)

20

15 Surface 10

5

0 1

2 space number

(c) Fig. 3.12. (a) Plan of House C2 (Lebeau 1993); (b) Flowchart; (c) Internal space sizes and features of House C2

House C2 was a small building of 29.8 metres2 composed of two or three rooms; however, the number of spaces and the position of the doors seem uncertain. No features were found inside the building apart from the circular arrangement of flat stones in the western corner. Lebeau interprets the structure as the remains of a silo for cereals or of a workspace for grinding cereals. In support of his interpretation might be the presence of a domestic oven outside the building near the southwestern side (Lebeau 1993, 95). In such a small house with a simple layout, spaces do not have a very specialised use. Probably the first and biggest room was a multipurpose space used for cooking, eating, and sleeping. Rooms 2 and 3, whether separated or not, could have been storage spaces.

Chapter Three T

92

3.1.8 Hou use G2

(a)

(b)

House G2 G 25

surface area (m2)

20

15

Surface basin grave jar oven

10

5

0 1

2

3

er space numbe

(c) Fig. 3.13. (aa) Plan of House G2 (Lebeau 1993); 1 (b) Flow wchart; (c) Intern nal space sizzes and featuress of House G2

North Region Case Study: Tell Melebiya, Titris Höyük and Tell Taya

93

House G2 was a small building of 25.4 metres2 with three rooms. The outside entrance was situated in the first room, which was the biggest with a surface area of nine metres2. A domestic oven has been found at the northeastern corner of the room and a jar was embedded in the floor. There was a pillar against the western wall and a child’s tomb was found under the floor near the southeastern corner (Lebeau 1993, 100). The other two spaces were smaller rooms that could only be accessed from the first room. Room 2 was equipped with a basin at floor level (Lebeau 1993, 100). No features were reported for the other room. As in House C2, it is possible that the biggest room (n. 1 in House C2 and n. 1 in House G2) was a multipurpose room used for sleeping, eating, food preparation, and other family activities, while two smaller rooms at the back were used for storage. As we can see from the flow charts (see Figs. 3.4[b], 3.6[b], 3.7[b], 3.8[b], 3.10[b], 3.11[b], 3.12[b], 3.13[b]), to understand the movement of people in a house, the ideal condition would be a well-preserved plan of each floor with a clear indication of the position of the doors. With a complete ground-floor layout, it is possible to establish whether a room is closer to the entrance, and if it is more “private” or “public” relative to other rooms on the ground floor. In the case of the existence of an upper storey within a given house, the missing preservation of the upper floor represents a serious gap in the understanding of a house. In fact, we lack important information about the possible activities carried out in these levels and the relevant circulation. Nevertheless, a spatial analysis of the ground floor alone has value. In archaeology we rarely deal with perfectly preserved evidence, but we can still extrapolate valuable information from the evidence available to us. The first thing that appears from the eight diagrams is that Melebiya’s houses present a low variability on the permeability chart. In fact, two main groups are evident: houses B6, C2, and G2 are the smallest among the eight houses with an average of only three rooms and a more linear circulation system. When there are only three rooms in a house, the biggest room may be a multipurpose space in which the majority of activities take place, as we have seen through comparisons with the ethnographic records. The biggest room is the one closest to the entrance; therefore, it is the closest to the major source of light, the door, and is suitable for daily activities such as food production and consumption, family gatherings, the reception of guests, and potential craft activity. The

94

Chapter Three

smallest rooms are often used for storage because they are darker, as attested by ethnographic parallels (Hansen 1961, 38; Aurenche et al. 1997). Access analysis is not meaningful in these situations because no room is very far from the entrance. From an analysis of the bar charts (see Figs 3.5, 3.6[c], 3.7[c], 3.9, 3.10[c], 3.11[c], 3.12[c], 3.13[c]) it appears that the houses can be divided into two groups. The first group comprises buildings with a larger surface area (between eighty and two hundred metres2): B1, B2, B4, B7, and G1 belong to this group. These are the houses with the most rooms (between ten and nineteen), but they also have the most extended rooms in terms of surface. In fact, rooms are present in those houses with a surface area of ten metres2 up to twenty-three metres2. In the second group of houses, the pattern seems to be the presence of two or three large spaces with a surface area between ten and twenty-three metres2. All the others are medium or small-sized rooms with a surface area of less than ten metres2 and often around five metres2 or less. It appears that the large rooms and some medium-sized rooms were likely to have been multipurpose rooms in which the family ate and slept and conducted the majority of the household activities. Benches and hearths are sometimes present. Those spaces all have relatively high values of depth, and are in fact placed towards the back of dwellings. Bar charts and flowcharts indicate the same two groups/types of houses: small houses and large houses. Lebeau hypothesises that for all the houses in Melebiya there was an absence of open spaces (open space refers to a roofless space inside the house walls) in his last excavation report, after expressing the opposing opinion in previous excavation reports (Lebeau 1993, 104). Each house had one or two facilities for heating or cooking such as hearths and tannours or domestic ovens. In House B4, the ovens were in small rooms not far from the entrance, perhaps in order to easily evacuate the cooking smoke. On the other hand, in houses B1 and B2 a hearth is placed in rooms with the greatest depth value, which means that they are further away from the entrance. From the ethnographic record we can see that cooking facilities can be present in living rooms, in small, separate spaces, generally close to the outside, or in open spaces. All the large houses have one or more basins, which suggests that they were used for craft activities. A recognisable pattern is present: often,

North Region Case Study: Tell Melebiya, Titris Höyük and Tell Taya

95

basins were in small-sized rooms on the sides of the buildings, but in House B1 a basin was present in room 7, which is large and central. Sometimes, one or more basins are associated with one or more benches. Lebeau identifies these rooms as workshops, adding that the floors are often plastered (Lebeau 1993, 106). In a final report, Lebeau recognises a more complex function for those rooms, perhaps related to their “private” position in the layout (Lebeau 1993, 104). He seems to suggest that those rooms were not utilised as workshops only. In light of the ethnographic records exposed in chapter two, it seems probable that Lebeau’s “workshops” may have had multiple uses, especially in the case of the largest rooms. Lebeau identifies “gathering rooms” as the largest spaces with benches. Perhaps the concept of “benches” should be reconsidered in light of the ethnographic records. Hansen describes activities that take place almost exclusively on the floor, which is the place for work, sleeping, and eating. She describes a specific use for “benches along walls,” which is to store bedding and other household objects. Aurenche confirms that benches were used for storage purposes. A similar use of “benches” in Melebiya houses appears to be possible—probably a more appropriate name for these “benches” would be “shelves.” In every large house Lebeau identifies more than one “gathering” room. He distinguishes between these rooms and the “reception” rooms. Because of their position, he seems to recognise their function as places of private reunion for the household, i.e. guests were not admitted in the “gathering” rooms while the purpose of a “reception” room would be to receive guests. One could argue that there is not enough evidence to make this distinction. On the basis of the archaeological evidence, we can only suppose that the largest rooms were the ones in which people spent the majority of their time indoors, where they gathered to eat, sleep, work, etc. Domestic altars have been identified in room 2 of house B2 and room 12 of house B7, both of which could have been living rooms. A preliminary interpretation might indicate that the rooms were multifunctional family rooms, and not devoted entirely to cult activities. There is an absence of storage areas in Lebeau’s identification of room functions. He suggests that big jars could have been placed on the terraced roofs (Lebeau 1993, 105). Ground-floor analysis and space analysis seem to contradict this interpretation, and from parallels with ethnographic

96

Chapter Three

observation it appears that storage areas were very important and necessitated a lot of space. It seems that many small rooms present in Melebiya houses could have been used for storage purposes, especially the rooms with the highest depth value, i.e. those farthest from the entrance. Storage rooms seem to have also been the most private spaces in Melebiya houses. They were the farthest from the doors in the large houses, but even in the small three-room dwellings they were relatively private. The multipurpose rooms had a relatively high depth value in the large buildings. In the small dwellings they were the first rooms from the door. The complexity and sophistication of houses B1 and B7 make Lebeau speak about the presence of specialised people in the construction of private houses (Lebeau 1993, 96). In trying to estimate household size, it seems probable that there is a relationship between the size of the house and the number of occupants: the small houses could accommodate a nuclear family while the larger houses could have been inhabited by more than one nuclear family, probably more than one generation of the same family. In large properties, spaces may have been rented out to non-family members, or some rooms near the entrance or isolated from the rest of the house could have been places for commercial or business activities. Brusasco, to estimate the number of people per household, utilises the hypothesis that each living room corresponds to a nuclear family. By living room he means relatively large rooms, often with benches and plastered floors, that other authors call reception rooms or gathering rooms. It is not easy to identify those kinds of rooms in a third-millennium house, especially if the majority of the spaces in a house could been used as a “living room,” and this method does not appear to be accurate as it does not seem possible to assign an exact numerical correspondence between one living room and one nuclear family.

3.2 Titris Höyük Titris Höyük was a small, mid-late third-millennium BC urban centre of approximately thirty-five hectares located on the lower Euphrates in southeastern Anatolia (Algaze et al. 1992, 33; Algaze and Misir 1994: 153; Algaze et al. 2001; Nishimura 2008; 2012). The site was the capital

North Region Case Study: S Tell Meleebiya, Titris Hööyük and Tell Taaya

97

ped along thhe Euphrates between of a small city-state syystem develop a southwestern Anatoliaa. The settlem ment was northern Mesopotamia and a areaa on a high mound that has been composed oof a central acropolis occupied froom the EBA to the Seljuk periods, and a lower city in which the archaeollogists recognnised two Low wer Town are as and an Ou uter Town area (see Fiig. 3.15). The Tavuk Cay flowed f by thee High Mound d and the Lower Townn areas (Matnney, Algaze, an nd Rosen 199 9, 185).

Fig. 3.14. Third milleennium northern n Mesopotamiaa (Nishimura 20 012)

98

Chapter Three T

Fig. 3.15. Plaan of Titris (Alg gaze and Misir 11994, 154)

There were also suburbs external to th he town wall, identified on the basis of surface scatters of ceraamics, contem mporary with the mid-late EBA and extramural ccemetery areas (Algaze et al. a 1995, 16). The earliestt occupation of the site dates to the EBA, whilee the site achieved urbban size in the mid-EBA an nd collapsed aat the end of the EBA. After the ccollapse, occuupation persissted on the High Mound d but the settlement w was reduced at a less than ten n percent of itts former sizee (Matney and Algaze 1995, 46–7). In total, 2,,500 metres2 of non-elitee building pllans were un ncovered, comprising nine complete or partial houses recovvered in two separate neighbourhooods (Matney,, Algaze, and Rosen 1999, 186). In the Outer Town, fourr building uniits at either siide of a centrral paved street runninng SW-NE haave been iden ntified (Matneey, Algaze, an nd Rosen 1999, 187). The occupatiion was a coh herent buildingg phase. In th he eastern

North Region Case Study: Tell Melebiya, Titris Höyük and Tell Taya

99

sector the architecture is terraced into the slope of the mound (Algaze et al. 1996, 130–2).

Fig. 3.16. Titris Höyük Outer Town Architecture, with permission from the Titris Höyük Archaeological Expedition, Timothy Matney and Guillermo Algaze.

3.2.1 Building Unit I Unit I is a large building that has not been completely excavated (see Fig. 3.16). It is most likely composed by fifteen spaces; fourteen have been completely excavated, while space 15 is only partially excavated. The building is delimited on two sides by streets: street A on the southeastern side and Street B on the southwestern side. It is probable that Street B continued into the unexcavated area and this was the delimitation of the building, as in fact the “basic structure is highly regularized” with a “recurrent uniform division of space into city lots” (Algaze et al. 2001, 68–9). Having, in Unit I, an almost complete layout of the ground floor, it is possible to conduct an analysis of it.

Chapter Three

100

Fig. 3.17. Flowchart of Building Unit I Building Unit I 45

40

35

surface(m2)

30

25 surface (m2) 20

15

10

5

0 5

3

12

11

4

15

7

10

2

14

1

6

13

8

9

space

Fig. 3.18. Internal space sizes and features of Building Unit I

The only visible entrance is along Street A from room 1, which does not have a direct connection with other spaces in the building. Either the main entrance of the house is from Street B, through room 15, or the connection of room 1 to the rest of the building has not been found by the excavators. The published reports do not give information about the findings and it is

North Region Case Study: Tell Melebiya, Titris Höyük and Tell Taya

101

only possible to make hypotheses on the basis of the layout of the building. Space 5 has the aspect of a courtyard because of its large dimensions and central position in the building. All the relatively large spaces such as n. 2, 3, 7, 10, 11, and 12 could have been living rooms with multiple functions such as sleeping, consumption of food, and probably spaces to work. Smaller rooms at the side of them such as n. 6, 8, and 9 could have been storage spaces. It is difficult to give a more accurate interpretation of the use of space because of the lack of more-detailed information in the excavation reports.

3.2.2 Building Unit II Unit II is a large building that, from the plan in Fig. 3.16, appears to be composed of fifteen spaces. The only entrance is along Street A from room 9. The building seems to be divided into two distinct sectors. The first sector seems to be composed of eight rooms disposed around space 5, which has the aspect of a central courtyard. The second sector is composed of five large rectangular rooms and a small room (space 14) that is in the most secluded area of the house and might have been a storage room (Matney et al. 1999, 187–8). Also, in this case the published reports do not give information about the findings, and it is only possible to make hypotheses based on the layout of the building. The first sector seems to have the classical arrangement of a central courtyard with rooms opening around it. Spaces 3 and 6 may have been living rooms with multiple functions such as sleeping and consumption of food. Space 7 has the characteristics of a storage room. Without other information, it is difficult to make hypotheses on the use of the other spaces that might have been rooms for cooking, storage, bathrooms etc. In its final phase, room 13 was accessible from Street C and cut off from the rest of the house. A self-standing circular basin of white plaster that was probably used for domestic functions and was reused in a mortuary arrangement of the long bones of seventeen individuals, with the skulls arranged around the basin. Fifteen of the seventeen individuals were young males, and many had traces of head trauma. The deposition was secondary and no funerary gifts were found (Matney et al. 1999, 189–90). The mortuary arrangement is unique in Titris, where other domestic tombs were cists.

102

Chapter Three

(a)

(b) Fig. 3.19. Building Unit 2, flowcharts

North Region Case Study: Tell Melebiya, Titris Höyük and Tell Taya

103

Building Unit II 25

surface area (m2)

20

15 surface (m2) basin grave(s) 10

5

0 5

12

13

3

10

6

11

16

9

8

1

2

4

7

14

15

space number

Fig. 3.20. Internal space sizes and features of Building Unit II

3.2.3 Building Unit IV The building block is delimited by two streets on two sides (northwest and southeast). The southern part of the building is delimited by an open space that archaeologists believe was used as a communal space for production activities and rubbish depositing. The building belongs to the EBA and has three phases. There are few remains of phase A, while the remains of phase B and C are shown in Fig. 3.21. The general aspect of the house did not change much in the different phases; room partitions were removed or added and doorways opened or closed.

104

Chapter Three T

Fig. 3.21. Building B Unit 4VI (Algaze et a l. 2001)

North Region Case Study: Tell Melebiya, Titris Höyük and Tell Taya

Fig. 3.22. Flowchart of Building Unit IV c

Fig. 3.23. Flowchart of Building Unit IV b

105

Chapter Three

106

Building Unit IVc 45

40

surface area (m2)

35

30 surface (m2) basin bin grave(s) heart(s) jar

25

20

15

10

5

0 8

6

20

1

15

9

16

11

12

22

5

3

24

17

19

2

14

4

18

23

25

21

7

13

space number

Fig. 3.24. Internal space sizes and features of Building Unit IV c Building Unit IVb 30

25

surface(m2)

20 surface (m2) heart(s) oven(s) basin jar

15

10

5

0 8

6

20

1

15

9

11 12 22 14

5

3

24 17 19

2

4

16 18 23 25 21 10 27

7

26

space

Fig. 3.25. Internal space sizes and features of Building Unit IV b

North Region Case Study: Tell Melebiya, Titris Höyük and Tell Taya

107

The excavators have interpreted building IV as one large house of 434 metres2 with twenty-two rooms and three internal courtyards. The building has four different sectors (A, B, C, and D) (Fig. 3.21), each with its own entrance from the outside, and sectors A and B have a connecting passage only in sub-phase C. Room 7 has a staircase, leading archaeologists to hypothesise the presence of a second storey. Sector A is composed of three spaces, with room 1 directly connected to the street (the largest room with a carefully pebbled floor) (Algaze et al. 2001, 27). Sector B is the largest, with a surface of 171 metres2. Space 8 is a courtyard with a packed-earth floor and is surrounded by rooms. In subphase C, the courtyard had a garbage pit and a cist tomb with the remains of children, while in sub-phase B the pit and the tomb were sealed by a new floor and rooms 10 and 26 were added. Room 11 has a rectangular mudbrick hearth, an in situ vessel, and ash with a high density of burned animal bones (Algaze et al. 2001, 27–8). The remains in room 11 have the characteristics of areas devoted to food preparation. The results of microdebris analysis in this room showed that food was prepared and cooked first and then, along with drinks, was placed onto a serving vessel and moved elsewhere for consumption (Algaze et al. 2001, 31). Also, the smaller rooms 13 and 14 have traces of ash. Room 9 is one of the most private rooms (see the flowchart), and the archaeologist also found a door socket attesting to the presence of a door to seclude the space. The floor was carefully plastered and cups and a bowl were retrieved (Algaze et al. 2001, 28). The space seems to have been used as a living room for eating and probably sleeping. Also, the micro-debris analysis suggests a multipurpose function for this room (Algaze et al. 2001, 32). Room 19 has the deepest permeability value. It is a narrow space accessible only by room 9, and has the characteristic of a secluded storage room to keep precious items. A rectangular mudbrick bin has been retrieved that archaeologists interpreted as being used for grain storage (Algaze et al. 2001, 28). Room 4 is at the opposite side of room 9, and because of its position, small dimensions, and absence of particular features seems to have been another storage room.

108

Chapter Three

Sector C has an area of 102 metres2. Space 15 has been identified as a courtyard and was pebbled. In sub-phase C, an entrance opened from the street directly to the courtyard (Algaze et al. 2001, 28). In sub-phase B, room 16 was divided into what seems to be a vestibule with pebbled flooring and a smaller room with an oven (Algaze et al. 2001, 28). Room 12 had a well-plastered floor and walls, an oven, and a platform blocking the passageway that originally led to sector B (Algaze et al. 2001, 28). Room 12 in sub-phase B has the characteristic of the living room, where the main activities of cooking, eating, and sleeping used to take place. Room 18 has the deepest permeability value and has the characteristics of a storage room. Room 17 might have been another living room or a storage room. Sector D has an area of 107 metres2. The excavators hypothesise an entrance from room 21 which has not been completely excavated. Space 21 is a courtyard with a pebbled floor and a basin of limestone. A large amount of pottery was present in rooms 23, 24, and 25, including several complete pots in situ (Algaze et al. 2001, 29). There was also a circular burned low platform and an oven associated with a small mudbrick platform in room 23. In room 24, several grinding stones and burned stones were found. Room 22 was only partially excavated (Algaze et al. 2001, 29). The excavators are inclined to exclude the hypothesis of more than one house unit, saying that only sector B has a “typical” house plan for Titris: “two large rectangular rooms perpendicularly arranged in the corner opposite the entrance” (Algaze et al. 2001, 29). If the interpretation of the doorways by the excavators is correct, Building Unit IV might have been composed by three houses in sub-phase C and four houses in sub-phase B. Alternatively, sector A (composed of rooms 1–3) might have been a shop or a workshop. This has also been shown in ethnographic parallel with the presence of commercial activity with separate entrances and also the presence of more than one house unit in one building block, where the secondary smaller house unit is composed of one or only a few rooms. In this case, dwellings often do not have a “typical” house plan.

North Region Case Study: S Tell Meleebiya, Titris Hööyük and Tell Taaya

109

Höyük, a fortiification walll has also beeen found. Along A the In Titris H r ro ooms were buuilt and arrang ged in the internal sidee a series of rectangular niches of thee wall (see Figg. 3.26, area D). D The entrannce is a single doorway situated on the street, and the typiical characterristics of a domestic environmentt, such as plaastered floors, evidence off burning, heaarths, and storage pits, have been found in the roo oms (Matney eet al. 1999, 18 88).

Fig. 3.266. Outer Town (Matney ( et al. 11999)

f In Also in the Lower Townn a coherent neighbourhoood area was found. osed (Matney et al. 1997, 62). 6 Stone 1996, one thhousand metrees2 were expo foundations up to one metre high were preserved, suupporting a no ow eroded mudbrick suuperstructure. Some floorss were plasteered and the floors in rooms near the entrancess were generally carefully ppaved with flaat stones, while courtyyards were roughly paved. A terrace sysstem was emp ployed on the slope sim milar to the Ouuter Town (M Matney et al. 19997, 62–3).

110

Chapter Three T

3.2.4 Building Unit 1 Only one coompleted plann of a house was uncovereed in the Low wer Town (Matney et aal. 1997, 64). Building Unnit 1 is a larrge building of o approximaately 285 mettres2 with seventeen roooms. The hoouse appears to t be dividedd into two uniits. In the first unit (sppaces 1–9) thhe rooms are arranged a arouund a courtyaard (space 5), and the eexcavator repoorts that room ms 3 and 4 weere built at a laater time, and thereforre the courtyard was originally larger (M Matney et al. 1997, 1 64). The second unit is an annnex (spaces 13–17) with no internal courtyard c connected bby a single dooorway to the main house uunit. In room m 13 there was a doorw way to the outtside, and it could have beeen a shop or workshop w which mightt have been coonnected to th he rest of the bbuilding. Some floorrs were plasstered and were w probablly used as domestic preparation surfaces (Mattney et al. 199 97, 63). The sttreet system appears a to be deliberattely laid out (Matney et al. a 1997, 63)). The prolifeeration of cooking areaas within the houses h may also a reflect thee separation off cooking by socioeconnomic or rituaal status (Matn ney et al. 19977, 65).

(a)

North Region Case Study: Tell Melebiya, Titris Höyük and Tell Taya

111

(b) Fig. 3.27. Building Unit 1 Lower Town (Matney et al. 1997) Building Unit 1, Lower Town 35

30

surface (m2)

25

20 surface (m2) 15

10

5

0 5

7

15

8

6

14

12

10

13

2

9

4

16

3

11

1

17

room

Fig. 3.28. Internal space sizes and features of Building Unit 1, Lower Town

112

Chapter Three

In the suburb area, three hundred metres east of the Outer Town, a workshop for the production of Cananaean blades has been found in association with domestic structures dating to the mid-EBA. Rosen. who excavated. the area judged the production to be very efficient and of high quality (Algaze et al. 2001, 37–40) The archaeologists presume that most buildings in Titris had a second storey because all the excavated Late EBA houses were built with adjoining double walls, and in two cases possible staircases were identified (Algaze et al. 2001, 53). Generally, rooms near the entrances were carefully paved with large flat flagstones while other surfaces, especially courtyards, were irregularly paved (Matney et al. 1997, 63). Two principal characteristics immediately appear which are not present in other sites analysed so far. The first is the presence of distinct sectors into which houses are often divided. The outline of each sector is evident from the layout of each house (for example, sectors C and D in Building Unit IV, and the sector composed by rooms 10 to 16 in Building Unit II). The second peculiar characteristic is the presence of “rings” in the circulation of the houses. There is a ring in a flowchart when one space is accessible in more than one way (see chapter three section 2). Hillier and Hanson associate “rings” in circulation with a more-accessible house and a less-hierarchical household. It is probably not possible to demonstrate if this is true for Titris houses, but it is important to observe the presence of at least one ring in the four houses analysed at Titris when often none are present at other sites. Other characteristics of Titris houses seem to be common to other sites, such as the presence of a central courtyard. On the other hand, the excavators identified in the “two large rectangular rooms perpendicularly arranged in the corner opposite the entrance” (Algaze et al. 2001, 29) a peculiar characteristic at Titris, although this arrangement is present only in Building Unit II and Building Unit IV. Building Unit I has a slightly different spatial organisation and its main entrance has not been retrieved. Algaze considers Titris buildings to be non-elite domestic houses, “since a comparable plan was recovered in several of the houses exposed in the contemporary neighbourhood” (Matney et al. 1997, 64). He has two

North Region Case Study: Tell Melebiya, Titris Höyük and Tell Taya

113

hypotheses for the presence of the division of the houses in two sectors (a main house with a central courtyard surrounded by rooms and an annex structure connected by a single door). The first hypothesis is the separation of private space from public space (the main house being private, the annex being public). The second hypothesis is the gender separation of the space. The first hypothesis seems more probable. Archaeologists made several attempts to demonstrate the existence of gender-specific areas in Mesopotamian houses, but it seems very difficult to find archaeological evidence for this. In the best cases, there is evidence of craft or commercial activities that are not demonstrated to be gender related and that could had taken place in specific separated areas. On the other hand, the scarce evidence for the condition of women in Mesopotamian society seems to suggest the absence of segregated gender-specific areas (Brusasco 1999–2000). Algaze points out that the plans of houses in Titris, Tell Asmar, Khafajah, and Abu Salabikh are similar and typically Sumerian, but that at Titris there are also tri and bi-partite room arrangements at either side of a central courtyard which is an “archaic” plan similar to Uruk Period houses. We seem to be in the presence of a coexistence of different architectural traditions (Algaze et al. 2001, 29–30). Algaze points out signs of planning in Titris neighbourhoods such as the shared use of terrace walls and wall alignments that cut across individual houses and streets. The space of each dwelling has uniform and regular measures, but the individual houses have individual characteristics, especially in the building of their annexes (Algaze et al. 2001, 68–9). These characteristics, together with the terracing walls shared by several houses, are indicators of supra-households and labour mobilization, perhaps due to a centralised administration (Matney et al. 1999, 193). In the Outer Town, Street C runs parallel to the city wall, Street A is almost perpendicular to Street C, and the small portion of Street B that has been exposed seems to be perpendicular to Street A. The Outer town does not seem to be arranged around a main street and cul-de-sac as Stone suggested for Mesopotamian cities, although the area excavated is too small to allow any conclusions. There is evidence of food preparation in Titris houses such as hearths, blackened plastered floor patches, cooking pots, and grinding tools

114

Chapter Three

(Matney et al. 1997, 65). The archaeologists think that Titris houses accommodated extended families and that the number of cooking areas may give an indication of nuclear families within the household, although they also report Henry Wright’s observation that this “may also reflect the separation of cooking by socio-economic or ritual status ” (Matney et al. 1997, 65). Several basins are an indication of production activities such as the washing and processing of wool or fleece, as well as the processing of grapes for wine production because of the presence of tartaric acid in the plaster (Matney et al. 1997, 65). Other presences of craft activities inside several houses are stone loom weights for textile production and even the workshop for the production of high-quality Canaanite blades associated with domestic structures. These are indicators of the strong economic role of the households in Titris (Matney et al. 1997, 65). There seem to be indications of intra-site spatial differentiation due to differences in status. In the Lower Town, houses are larger compared with Outer Town houses, intramural burials were richer, and a jewellery mould was found in a Lower Town house (Matney et al. 1997, 70–1). Although there is a presence of extramural cemetery areas (Algaze et al. 1995, 16), many intramural cist tombs have been found in houses. They were communal burials mostly with the disarticulated remains of multiple individuals, probably family members occupying the house. One of the functions of many Titris houses was to be the place of mortuary rituals (Matney et al. 1999, 189). Unfortunately, there is an absence of textual documentation in Titris (Algaze et al. 2001, 67).

3.3 Tell Taya Tell Taya lies in the northeast of the Sinjar-Tell ’Afar plain (Reade 1968, 234). The site is situated on the hills near the plain at the confluence of two non-perennial rivers (Reade 1968, 239). The site was first mentioned by Lloyd in his account of the survey of the Jebel Sinjar (Lloyd 1938, 123–42). The site was excavated by Reade and his team between 1967 and 1973 (Reade 1968; 1971; 1973). Excavation in the citadel at the top of the mound exposed nine levels of occupation that have been dated, on the

North Region Case Study: Tell Melebiya, Titris Höyük and Tell Taya

115

basis of the pottery remains, from the late Early Dynastic III to the Parthian Period (Reade 1968, 240–60). No complete plans of houses have been excavated on the site. Abundant surface remains in the form of limestone foundations enabled a drawing of a large area of the site plan (Reade 1973, 179). In the plan, the outline of the neighbourhoods is clearly visible, with streets running alongside the residential areas.

Fig. 3.29 (a) Tell Taya (Reade 1973; Dibo 2016)

116

Chapter Three T

Fig. 3.29 (b) Teell Taya, areas Tcc and Ucc R eade (1973)

North Region Case Study: S Tell Meleebiya, Titris Hööyük and Tell Taaya

Fig. 3.29 (c) Telll Taya, areas Py y, Qaa and Qbb (Reade 1973)

117

118

Chapter Three T

Fig. 3.29 (d) Tell Taya, area a Vx (Readee 1973)

North Region Case Study: S Tell Meleebiya, Titris Hööyük and Tell Taaya

Fig. 3.29 (e) Tell T Taya, areas AAr and Yr (R Reade 1973)

119

120

Chapter Three T

The overvieew of the toown plan is impressive, with neighb bourhoods composed oof clusters of o houses cleearly visible, in some caases with complete grround-floor laayouts. Unforttunately, thosee that remain have not been excavaated but are coonsidered to be b dated at thhe third millen nnium (R. Matthews ppersonal comm munication). An analysis of the layou ut of the houses whicch appear to be complete or almost co mplete has been done based on suurface evidennce. Nine hou uses have beeen selected with this criterion. Nuumbers have been b assigned to each spacee in the houses in order to build flowcharts of the internal circulation, and an app proximate calculation of the surfaace of the ground floor has been made. The calculation could not bee very precisse having beeen obtained from the published pplan of Tell Taya, but it is useful foor giving an order of magnitude ffor comparisoon among ho ouses at Tell Taya and with w other houses at diifferent sites. The surfacess of each spaace in houses have not been calculaated because the error wo ould be too l arge to obtaiin even a rough measuurement. Therrefore, it was not possible to produce a bar chart for the surfaaces of spaces as for the oth her sites.

3.3.1 Areea AAr

(a)

(b) Ar (Reade 1973)); (b) Flowcharrt Fig.. 3.30. (a) Plan of house in AA

North Region Case Study: Tell Melebiya, Titris Höyük and Tell Taya

121

The house in area AAr seems to have been composed of six spaces, which occupied about fifty-seven metres2 (see Fig. 3.30). The building seems to be complete at the ground-floor level because it appears to be adjacent to at least two houses: one on the northeastern side and one on the southwestern side. The southeastern side of the building seems to be flanked by an alley. The only entrance seems to be placed at the eastern corner of an open space. In the open space, traces of kiln debris have been found (marked with stars in the plan). All these observations are at the level of hypothesis because it is not possible to know if the absence of structures is due to the open space or poor preservation. The house is located at the eastern end of the settlement in an area rich with kiln debris, and fifty metres further east there are the remains of at least one kiln. With this observation one can make the hypothesis of a production of ceramic in the area, perhaps even in the open communal space in front of the house in AAr. Based on the plan of the house, space 1 appears to be the largest. It might have been an open courtyard or, if it was a covered space, a multipurpose room. There is not sufficient information to make a more accurate hypothesis on the use of this space or the others.

3.3.2 Area Py The house in area Py is located at the western end of the settlement. It appears to be isolated from other buildings, although it is very likely that the buildings on the western side of the wall that delimits the southwestern end of the house have disappeared because of preservation problems (see Fig. 3.31). The building appears to be composed of six spaces but it is incomplete. It is not clear if the entrance was located in space 3. Space 4 might have been a central courtyard, rooms 1–3 multipurpose rooms for sleeping, working, and eating, and rooms 5–6 storage spaces. The surface of the ground floor was approximately 120–130 metres2.

122

Chapter Three T

(a)

(b) Figg. 3.31: (a) Plann of house in Py y (Reade 1973);; (b) Flowchart

3.3.3 Areea Qaa

(a)

(b) Fig. 3.32. (a) Plan of house in Qaaa (Reade 1973)); (b) Flowchartt

North Region Case Study: S Tell Meleebiya, Titris Hööyük and Tell Taaya

123

The house inn area Qaa allso appears to have been coomposed of siix spaces, delimited onn the eastern side s by a street and on the northern and d southern sides by houuses (see Fig. 3.32). Together, thhe three housees seem to hav ve formed a soort of insula, delimited on two sidess at least by sttreets. The surface of the groundd floor, which h seems to be complete, is about a 190 metres2. Spaace 4 looks likke a large internal courtyardd, but it is posssible that at least one room was preesent in the no orthwestern coorner of spacee 4. These interpretatioons are, as for the other Tay ya houses, at tthe level of hypothesis from the obbservation of the t plan. It iss possible thatt the recorded d remains belonged to different buillding phases. Hypotheticaal in the samee way is the interpretation i of the function of the rooms. Room m 5 is the larrgest and mostt distant from m the entrance; it might have been a multipurposee room for thee household too sleep, eat, and a work, while the oother rooms might have been b used foor storage orr cooking purposes.

3.3.4 Areea Qbb

(a)

(b)

(c)

Reade 1973); (b)) Flowchart Ho ouse I; Fig. 3.333. (a) Plan of hoouses in Qbb (R (c) Flowchart House II

124

Chapter Three

In area Qbb there are two houses that seem to have an almost complete layout of the ground floor (see Fig. 3.33). The first house is delimited by a street on the southern side. It appears to be composed of six rooms with a total surface of around 127 metres2. The entrance might have been located at space 1, which now seems to be a large room but might have been divided into smaller spaces. A second hypothesis is that the house was larger with more rooms on the eastern side that have not been preserved. Space 3 might have been an internal courtyard, space 5 a multipurpose room, and space 6 a storage room. The northern side the building is delimited by another house. The layout of the ground floor seems almost complete and the house is not surrounded by other buildings on the eastern, northern or western sides. The entrance could have been in room 1, but the walls at the northeastern corner are poorly preserved. The functional interpretation of the other rooms is difficult because the plan does not follow a regular pattern. No internal courtyard seems to be present, but this observation is based on the dimensions and position of the spaces in the house only.

North Region Case Study: S Tell Meleebiya, Titris Hööyük and Tell Taaya

3.3.5 Areea Tcc In area Tcc a large house is present (seee Fig. 3.34).

(a)

(a)

Figg. 3.34. (a) Plan of house in Tcc (Reade 1973)); (b) Flowchartt

125

Chapter Three T

126

3.3.6 Areea Ucc In area Ucc the layout off the almost co ompleted grouund floor of a house is present (see Fig. 3.35).

(a)

(a)

Fig. 3.35: (a) Plan of house in Ucc (Reade 1973)); (b) Flowchartt

The house iss delimited onn the southwesstern side by a street. The rest r of the house is isollated with no other building gs around it, ssave for a hou use on the southeasternn side. The tw wo houses are not part of a bblock, and thee walls of the two houuses have a different alignm ment. The buuilding is com mposed of eleven spacces and has a surface off approximateely 130 metrres2. The ground-floorr plan is squaare with a square central coourtyard surrounded by rooms. Thee entrance seeems to be from f a corriddor that lead ds to the courtyard, bbut because the layout is not complete it is not clear iff there are additional oopenings on thhe southern siide along the street. On th he eastern side, the couurtyard is flannked by a row of relatively small rooms for f which it is difficultt to give a funnctional interp pretation. Theey could have been, for example, stoorage or cooking facilities. On the westeern side there are a larger rooms. Spacce 9 might havve been a mu ultipurpose rooom. Spaces 7 and 8 do not seem to be not connected to the maain house andd are open to the t street; they might hhave been rennted to tenantss or constituteed a shop or workshop. w Near the maain entrance are a the remain ns of a curvedd wall, which does not seem to havve been part of the house and possibbly part of a different constructionn phase.

North Region Case Study: S Tell Meleebiya, Titris Hööyük and Tell Taaya

127

3.3.7 Area Vx ood (see Fig. 3.36). A streeet runs on Area Vx is a densely buillt neighbourho the southeasstern corner of o the area an nd there is ann almost perp pendicular alley flankedd on both sidees by building gs. The first buuilding on thee southern side seems to be an inncomplete hou use plan with th a central courtyard c surrounded by rooms. Thhe following building b on thhe same side is i another house with a seemingly coompete plan of o the ground floor.

(a)

128

Chapter Three

(b) Fig. 3.36. (a) Plan of house in Vx (Reade 1973); (b) Flowchart

The building is composed of eleven spaces with a surface of about three hundred metres2. The entrance is at the northeastern corner of the house in room 1. Space number 6 has the characteristics of a central internal courtyard surrounded by rooms. Space 12 has the characteristics of a multipurpose living room for sleeping, eating, and working because of its large dimensions and position far away from the entrance. Space 13 could have been a storage place. Also, rooms 4 and 2 could have been multipurpose spaces because of their large dimensions. Room 7 could have been a storage space, and room 3 could have had some other facilities, for example for cooking. Space 11 is an isolated room opening directly to the outside that might have been a shop or workshop. Rooms 8 and 9 are connected to each other but do not seem to be connected with the rest of the house. A door could

North Region Case Study: S Tell Meleebiya, Titris Hööyük and Tell Taaya

129

w what look ks like an alleey flanking th he eastern have conneccted room 8 with side of the bbuilding.

3.3.8 Areea Yr/s In area Yr/ss there seems to be a well-p preserved com mplex of twen nty spaces covering aboout 326 metrees2 (see Fig. 3..37).

(a)

130

Chapter Three

(b) Fig. 3.37. (a) Plan of house in Yr (Reade 1973); (b) Flowchart

Space 1 seems to be a courtyard surrounded by doors leading to different sectors of the complex. The first sector (spaces 2–8) could have been a house. The rectangular shape appears to be similar to the houses in area Qbb (I) and Qaa (see Figs. 3.33 and 3.32). The other sectors could have been used for production or commercial activities. The layout is different than other buildings at Tell Taya, and without more information it is difficult to understand the function of the rooms. In summary, among the nine houses analysed at Tell Taya, five were organised with a central courtyard surrounded by rooms (Py, QbbI, Tcc, Ucc, and Vx). Qaa and QbbII have a long rectangular shape and might have also had an internal courtyard. Yr also might have been a house with a long rectangular shape, as well as annexes with workshops connected by an internal courtyard. Besides the two types of Tell Taya houses (the type

North Region Case Study: Tell Melebiya, Titris Höyük and Tell Taya

131

with a central courtyard and the rectangular type) there is AAr, which apparently does not belong to either type. It is the smallest house, less than sixty metres2, while the other houses are all greater than 120 metres2. The impression is that AAr belongs to the category of “small houses” that do not have a standard layout but seems to be a vernacular architecture that is mainly determined by the spaces available such as the “small houses” of Tell Melebiya or the house of the schoolteachers in Sulaimani (see chapter two). The other two types of houses seem to have been planned following a project. Additionally, the neighbourhoods in Tell Taya bear evidence of planning, such as wall alignments that cut across individual houses and streets that are considered indicators of supra-household organisation due to a centralised administration (Matney et al. 1999, 193). The presence of a kiln and abundant kiln debris in the southeastern area of the town might be a sign of a household-based pottery production district.

3.4 Summary At the end of this chapter it is useful to summarise the characteristics of the houses analysed from the three sites. The presence of dwellings with a central space around which rooms are organised is a common aspect of the three sites. Each site has peculiar characteristics: at Melebiya, the absence of an internal courtyard has been reported, although this interpretation seems questionable. The four cases of Titris have the model of the house with a central courtyard always with an adjoining annex. At Taya, five out of nine houses seem to be organised with rooms surrounding a courtyard. At Titris, self-standing “small houses” have not been excavated. The four dwellings analysed have between fifteen and twenty-two spaces and are all more than two hundred metres2. Building Unit VI is composed by sectors, some of which could have been separate dwellings. In addition are rooms built in the niches of the fortification wall that have a domestic function, but in general the neighbourhoods appear quite homogeneous in their composition, with large courtyard houses with annexes organised in regular slots that suggest an homogeneous social composition. Moreover, there seem to be indications of differentiation in status between the two areas excavated. When compared to Outer Town houses, Lower Town

132

Chapter Three

houses were larger and had richer intramural burials (Matney et al. 1997, 70–1). On the other hand, at Melebiya houses of two to four rooms are adjacent to large courtyard houses of eleven to sixteen rooms that suggest a mixed social composition of the neighbourhoods. In the case of Taya, the interpretation is more difficult because of the incomplete surface remains. From the observation of the detailed plan, however, it is possible to note long streets and symmetry in wall alignments that extend across the partition walls of individual houses and streets. The neighbourhoods appear to be mixed, composed of different types of houses with various dimensions and layouts. There is evidence of food preparation in Melebiya and Titris houses, such as hearths, tannours, blackened plastered floors patches, cooking pots, and grinding tools. A similar observation cannot be made at Taya due to the lack of excavation. Several basins found inside rooms at Melebiya and Titris are an indication of production activities such as the washing and processing of wool or fleece, and the processing of grapes for wine production because of the presence of tartaric acid in the plaster at Titris (see 3.2 above and Matney et al. 1997, 65). The presence of craft activities inside several houses is demonstrated by stone loom weights for textile production and even workshops for the production of high-quality Cananaean blades associated with domestic structures in Titris. Although there is not such detailed information for Taya, the presence of a kiln and abundant kiln debris in the southeastern domestic area of the town might be signs of household-based pottery production in the area. It is clear from the excavated evidence of Melebiya and Titris that houses were a centre of production, and in some cases the facilities seem to be too many for only household necessities and might have been the base of commercial activities.

CHAPTER FOUR CENTRAL REGION CASE STUDY: KHAFAJAH AND TELL ASMAR

The methodology discussed in chapter two is applied here to two late third-millennium sites where houses have been excavated: Khafajah and Tell Asmar. These two cities have been chosen because they are among the best remains of domestic architecture in the region, and in particular an unusually large amount of houses from the same period have been excavated in the two sites. Around forty houses analysed in this thesis from the two sites present a complete plan of the ground floor. The two archaeological sites were first excavated at the same time by a team of the University of Chicago in the 1930s (Frankfort, Jacobsen, and Preusser 1932; Frankfort 1933; 1934; 1935; 1936), and specific analyses of Khafajah and Tell Asmar houses were carried out by Delougaz, Hill, and Lloyd (1967) and Henrickson (1981). Some fixer-features, graves, and objects were found in the building but no modern techniques such microartefact analysis are available to help in the functional analysis.

4.1 Khafajah The site of Khafajah, or Khafaje, is located eleven kilometres east of Baghdad in the Diyala region (Delougaz, Hill, and Lloyd 1967, 1), an area named after the Diyala River, which is a tributary of the Tigris. The texts found in the archive of the Sin temple identify Khafajah as the ancient town of Tutub (Harris 1955, 32). The archaeological site is composed of three apparently unconnected mounds (Frankfort, Jacobsen, and Preusser 1932, 61).

134

Chapter Four

Fig. 4.1.. Diyala Regionn (Frankfort, Jaccobsen, and Preeusser 1932, Fig g. 1).

Central Regioon Case Study: Khafajah and T Tell Asmar

135

Fig. 4.2. P Plan of Khafajahh (Frankfort, Jaacobsen, and Prreusser 1932, Fiig. 19).

136

Chapter Four

Fig. 4.3. Aeriaal view of Khafa fajah (Frankfortt 1933, 154)

Fig. 4.4. Aerial view v of Khafajaah (Frankfort 19934, Fig. 54).

Central Region Case Study: Khafajah and Tell Asmar

137

Every find in the site belongs to the Early Dynastic Period (Frankfort 1935, 79), specifically the Early Dynastic II and III and the Protoimperial (Henrickson 1981, 44).

4.1.1 Khafajah Mound A Mound A was the largest among the three mounds and had signs of the earliest occupation, which began in the Uruk Period and continued until the Akkadian Period (Harris 1955, 33). Three temples were excavated in the mound: the Oval temple, the Sin temple, and the Nintu Temple (Harris 1955, 33). Private houses were also excavated in Mound A, ranging from about the middle of the Protoliterate Period to Protoimperial times (Delougaz, Hill, and Lloyd 1967, 1). Delougaz, Hill, and Lloyd (1967) reported the presence of twelve levels of occupation in the excavated domestic area (named houses 12 to 1, abbreviated as Hss 12–1). Henrickson studied the settlement patterning of late ED Khafajah, and did not take into consideration houses 12–7 from previous levels because no complete house plan has been found at those levels, only sparse architectural remains (Henrickson 1981, 44–6). Henrickson determined three domestic occupational phases: Hss. 5–6, Hss. 3–4, and Hss. 1–2, regrouping Delougaz, Hill, and Lloyd’s (1967) domestic levels. There is a difference between Hss 6 and the previous levels. The domestic area was planned anew, and the rooms of houses were larger and more regular in shape. The walls were thicker, up to 1.40 metres, with different heights at various points. Hss 6 to 1 present various typologies of plans with a very different number of rooms at the ground-floor level (from 2 to 12 rooms for each house).

138

Chapter Four

Fig. 4.5. Khafajah conntour map of Mound M A showinng excavated arreas (Delouggaz, Hill, and Lloyd L 1967, platte 1)

Central Regioon Case Study: Khafajah and T Tell Asmar

4.1.1.1 Hou use XXXII/X XXXV/XXXV VIII/XLI

(a)

(b) Fig. 4.6. (a) Plan of Houuse XXXII in grrey and the adjaacent temple in white (Delougaz, Hill, and Lloyd d 1967); (b) Floowchart.

139

Chapter Four

140

House XXXII 60

Surface area (m2)

50

40

surface (m2) grave(s) jar oven(s)

30

20

10

0 27

28

23

9

25

24

12

4

20

22

16

17

21

Space number

Fig. 4.7. Internal space sizes and features of House XXXII and the adjacent temple.

Building XXXII in Hss 6 is a house organised around a central court, space 27, with nine rooms around it (see Fig. 4.6). Henrickson calculates that the building has a surface of 310 metres2 (Henrickson 1981, 52). She suggests that a building of this type could have been the dwelling of several nuclear families, part of an extended family, or a lineage unit (Henrickson 1981, 54). The building appears to be complete, although the southwestern side has not been excavated. The eastern side was delimitated by an alley, on the western side there was the temple of Sin, and on the northern side the building was bound by what could have been a street, but the area has not been further excavated. Not all the doorways were found in the excavation. The report specified that the position of some doorways has been suggested taking into consideration the spatial organisation of the rooms (Delougaz et al. 1967, 10). Two entrances to the block have been hypothesised: one from room 9 for the courtyard house and one from room 16 for the Sin temple. Space 12 had a large oval kiln in it, and must therefore have been used as a cooking facility, while in the courtyard there were semi-circular structures

Central Region Case Study: Khafajah and Tell Asmar

141

built of unbaked bricks of unspecified use (Delougaz et al. 1967, 10). Two vaulted graves had been built below the floor of the courtyard. A pit burial has been found in space 20. The room might have been a storage room based on its dimensions and high depth value. Three connected vaulted tombs were discovered below a secondary floor in space 23, which was the largest room, and together with room 24 and room 28 could have been a multipurpose space where the family ate, slept, and worked. It is possible to see in rooms 9 and 28 what looks like a circular installation, but there is no description of it in the excavation report. Room 22 was a very small space that Delougaz, Hill, and Lloyd (1967, 10) suggested could have been a storage area or a staircase. It is difficult to propose a function for room 25—it could have been another living room or a storage room. The fact that a jar containing many objects was found in this room could point towards its use as storage. A flint blade, six shell animal amulets with clay, faience, shell, and stone beads, two copper fishhooks, a pinhead, and other small objects were found in the jar (Delougaz, Hill, and Lloyd 1967, 28). A cylinder seal has been found in both room 12 and room 20, indicating that the inhabitants of the house may have been part of the high stratum of society, and that someone might have been part of the temple’s administrative staff, given the proximity of the house to the Sin temple. A stone cosmetic jar with traces of green pigment and a stone macehead were found in room 24. Adjacent to the house is a smaller building composed of rooms 16, 17, and 21. No objects were found in the building, but space 21 has an altar along the southeastern wall and has been identified as a small temple. The location of the small temple is at almost the same since level as Hss 11 and continues through the levels until Hss 3. The continuity through time is a characteristic of sacred areas, which usually do not change their location even when the layout of the buildings around them does. In level Hss 6, the shrine (space 21) became part of a bigger building. Without more information, it is difficult to understand the functions of the other spaces: room 16 might have been an open courtyard and room 17 could have hosted facilities for the temple. A kiln was present in the plan (Fig. 4.6) in what had the appearance of an open space between buildings that had not been completely excavated. In his excavation report, Delougaz distinguished a new phase of the building that he called Hss 5 (Delougaz, Hill, and Lloyd 1967, 11), but Henrickson in her analysis considered Hss 6 and Hss 5 to be part of the same phase because very little changes were reported from one to the other

142

Chapter Four

(a)

(b) Fig. 4.8. (a) Plan of House XXXV X (Delougaz, Hill, and Llloyd 1967); (b) Flowchart

Central Region Case Study: Khafajah and Tell Asmar

143

House XXXV 50 45 40

Surface area (m2)

35 30 surface (m2) bench

25 20 15 10 5 0 27

28

23

9

25

24

12

4

20

22

Space number

Fig. 4.9. Internal space sizes and features of House XXXV

(Henrickson 1981, 46). House XXXII was called House XXXV in level Hss 5 of the Delougaz report (see Fig. 4.8), and appears to remain almost unchanged from Hss 6 to Hss 5 levels. The few variations consist of the absence of the graves that were present at level 6, a vaulted tomb that was present in space 9, in room 12 the oven was no longer there, and a large quadrangular structure appeared. Delougaz interpreted this as a possible storage bin (Delougaz, Hill, and Lloyd 1967, 11). The so-called semicircular “domestic” structure that has not been better identified is present in court 27 at both levels. It is possible to see in room 2 an installation, but there is no description of it in the excavation report. Room 24 in Hss 5 had a bench and a stone stamp seal was found during the excavation of the space. A stone lion-head stamp seal amulet was also retrieved in space 13 between the houses. The spatial organisation of House XXXV is almost identical to that of House XXII, and as a consequence the use of rooms seems to be the same. The small temple is composed of spaces 16, 17, and 21 is almost identical between level Hss 6 and level Hss 5.

144

Chapter Four

(a)

(b) Fig. 4.10: (a) Plan of House XXXVIII X (Delou ugaz, Hill, and L Lloyd 1967); (b) Flowchart

Central Region Case Study: Khafajah and Tell Asmar

145

House XXXVIII 60

Surface area (m2)

50

40 surface (m2) altar(s) bench drain grave(s)

30

20

10

0 20

27

28

23

24

9

4

12

26

25

22

11

Space number

Fig. 4.11. Internal space sizes and features of House XXXVIII

At the next level, Hss 4, the building is called House XXXVIII and has a central court surrounded by almost the same rooms as at the previous levels (see Fig. 4.10). In Hss 4, the excavators were able to identify all the doorways. The modifications to the structure were: a reduction in the dimension of room 12 where in Hss 4 also appeared to be isolated from the rest of the house and opened directly to the outside, suggesting a commercial function or a different use that could involve people outside the household; room 9 was no longer connected with room 12, but had a connection with a small space that looked like a storeroom that Delougaz called a closet (room 11); in room 9 a seal impression was found. It is possible to see in the room what looks like a circular installation, but there is no description of it in the excavation report. Space 25 was divided into two different rooms that could have been used for storage proposes. Room 24 stayed the same size, but two new structures were uncovered that the excavator interpreted as altars; if this interpretation is correct, space 24 had a cultic area. Six stone maceheads, a stone needle, stone bowls, a stone figurine fragment and a copper fishhook, a copper spearhead, a copper needle, and pins were also found in room 24. Room 28 had a drain, a bench, and a vaulted tomb. The space presents the characteristics of a living room (Delougaz, Hill, and

146

Chapter Four

Lloyd 1967, 12). In the court c (space 27) 2 a stone boowl was foun nd, and in space 23 a pplaque fragment, a copper fishhook, annd a cylinder seal s were found (Deloougaz, Hill, annd Lloyd 1967 7, 32). In the smalll temple on the t southwestern side of thhe house, the partition between spaace 16 and 17 was not present anymore, and at this leevel space 16/17 appeaared like a largge internal cou urtyard. The sshrine remaineed almost unchanged. A kiln was built in the rectangular open o space onn the southern n side of House XLI aand a grave was w found in th he eastern partt of the space.. At level Hsss 3, the buildiing changed itts name againn and was called House XLI. This leevel, being near the surface, was affectedd by illicit exccavations. The house w was reconstruccted using thee preserved w walls and the plan p from the previouss level, and diid not presentt architecturall variations. No N burials were found at this level (D Delougaz, Hilll, and Lloyd 1967, 13). A shell face amulet was found in room m 26.

(a)

Central Region Case Study: Khafajah and Tell Asmar

147

(b) Fig. 4.12. (a) Plan of House XLI (Delougaz, Hill, and Lloyd 1967); (b) Flowchart

The small temple has a new doorway connecting the shrine directly with the alley (15), and there is a new circular structure in the southwestern corner of space 16. There is a kiln in the open area adjacent to the house and the small temple, and in the southwestern part of the area a fragment of a stone cosmetic jar was found near the facility while a stone face amulet was found in the courtyard (Delougaz, Hill, and Lloyd 1967, 36). After level Hss 3, the area was probably abandoned because level Hss 3 was covered by an extensive layer of ash (Delougaz, Hill, and Lloyd 1967, 16). 4.1.1.2 House XXXIII and House XXXIV At level Hss 6, across alley 15 that flanked the small temple, there is a block composed of House XXXIII and House XXXIV. The block with the two houses has the aspect of an “insula” surrounded by streets, although the area has not been completely excavated (Fig. 4.13).

Chapter Four

148

(a)

(b)

(c)

Fig. 4.13. (a) Plan of Houuse XXXIII and d XXXIV (Deloougaz, Hill, and d Lloyd XXIV 1967); ((b) Flowchart of House XXXIIII; (c) Flowcharrt of House XX

Central Region Case Study: Khafajah and Tell Asmar

149

House XXXIII 30

Surface area (m2)

25

20

15

surface (m2)

10

5

0 21

23

23(35)

Space number

(a) House XXXIV 45

40

Surface area (m2)

35

30

25 surface (m2) 20

15

10

5

0 6 12

20

22

27

19

11

Space number

(b) Fig. 4.14. (a) Internal space sizes and features of House XXXIII; (b) Internal space sizes and features of House XXXIV

150

Chapter Four

Building XXXIII is a dwelling of 112 metres2 (Henrickson 1981, 52) composed of three rooms (Fig. 4.13[a] and 4.15[b]). The division of space is not typical and there are no fixed features to help in the interpretation of the room usage. Two objects were found in room 23: a miniature glazed fruit bowl and a stone lid (Delougaz, Hill, and Lloyd 1967, 28). The entrance to the house is on the northwestern side, from room 23. The adjacent building XXXIV is large at 184 metres2 (Henrickson 1981, 52) and is composed of six rooms (Fig. 4.13[a] and 4.15[c]). The excavators found three sides of the building, the southeastern side was not preserved and has been reconstructed (Delougaz, Hill, and Lloyd 1967, 10). The archaeologists hypothesised that the entrance was from room 6/12. A grave was found under a wall in room 6/12 that seems to have existed before the construction of the house. There was another grave just outside room 6/12 (Delougaz, Hill, and Lloyd 1967, 10). House XXXIV also does not have fixed features to help interpret the use of space. By analysing only the surface of the rooms and their accessibility, it seems probable that space 11 was a storage room, as spaces 22 and 19 could also have been. The others could have been multipurpose rooms for processing food or other types of work and for sleeping. A shell bead was found in room 20, and a baked-clay cone for a wall mosaic was found in room 22 (Delougaz, Hill, and Lloyd 1967, 28). A wide street separates the group composed by houses XXXII and XXXIV from the wall of the Temple Oval (Delougaz, Hill, and Lloyd 1967, 10). 4.1.1.3 House XXXVI and House XXXVII At level Hss 5 the two houses changed names. The three-room dwelling on the eastern side is called House XXXVI and occupies the same area as House XXXIII at level Hss 6 (see Fig. 4.15). A large oval kiln is present in room 35, a wall was shifted between room 21 and 23, and graves are present in both spaces.

Central Regioon Case Study: Khafajah and T Tell Asmar

151

(a)

(b)

(c)

Fig. 4.15. ((a) Plan of Houuse XXXVI and d XXXVII (Del ougaz, Hill, and d Lloyd 1967); (bb) Flowchart off House XXXV VI; (c) Flowcharrt of House XX XXVII

Chapter Four

152

House XXXVI 40

35

Surface area (m2)

30

25 surface (m2) 20

kiln grave(s)

15

10

5

0 23

21

35

Space number

Fig. 4.16. Internal space sizes and features of House XXXVI

The organisation of House XXXVII is almost identical to the one at level Hss 6, the only difference from that at level Hss 5 being the presence of several graves. Each space, with the exception of room 19, has at least a tomb. In room 22 a child burial in a basket was found (Delougaz, Hill, and Lloyd 1967, 11). 4.1.1.4 House XXXIX and House XL At level Hss 4 the area on the northeastern side of the Temple Oval was reorganised. House XXXIX is a small dwelling composed of two rooms (spaces 14 and 3) accessed from the alley (15), with a lane separating the block of dwellings from the small temple (see Fig. 4.17 and 4.17[c]). From the plan it is possible to see a grave in each room and what looks like a circular installation, but there is no description of it in the excavation report.

Central Regioon Case Study: Khafajah and T Tell Asmar

153

(a)

(b) Fig. 4.17. (a) Plan of House XXXIX and XL X (Delougaz, H Hill, and Lloyd 1967); (b) F Flowchart of House H XXXIX; (c) ( Flowchart oof House XL

Chapter Four

154

House XXXIX 20 18 16

Surface area (m2)

14 12 surface (m2) grave(s)

10 8 6 4 2 0 14

3 Space number

(a) House XL 35

30

Surface area (m2)

25

20 surface (m2) grave(s) 15

10

5

0 N6

15

12

O6

19

14

2

5

18

7

Space number

(b) Fig. 4.18. (a) Internal space sizes and features of House XXXIX; (b) Internal space sizes and features of House XL

Central Region Case Study: Khafajah and Tell Asmar

155

House XL occupies the rest of the area (Fig. 4.17[a] and 4-17[b]). It is a large dwelling composed of ten spaces distributed in an area of around 280 metres2. The plan is quite original. Room 18 appears to have been an entrance that divided the house into two distinctive parts. The reason for this differentiation could have been the separation of two nuclear families belonging to the same household or a commercial activity in the smallest part of the house. Space 15 had the aspect of an internal courtyard and a grave was found in the floor. The court is similar in size to court 27 in House XXXVIII, but it is not located in the centre of the house (Delougaz, Hill, and Lloyd 1967, 12). In room O6 (O43), and perhaps also in room 2, an oven was excavated. Room 14 was particular because it was the only one not sharing a wall with the outside nor adjacent to a court. This room seems to have no light source, a situation suggesting a storage function. Spaces N6 and 12 could have been multipurpose living rooms, and spaces 7 and 19, because of their smaller dimensions and positions (Fig. 4.17[c]), could have been storage rooms. Six graves have been found in House XL: a simple grave in the courtyard, two built tombs in room 12, a vaulted tomb with a simple grave beside it in room 5, and a vaulted tomb with thick walls in room N6 (Delougaz, Hill, and Lloyd 1967, 12). A cylinder seal was found in room O6, a stone bead and a jar in room 12, in the courtyard a copper dagger blade, possibly from grave 132, was found together with a pottery box fragment, a stone stamp seal, a cylinder seal, and a shell cap for a cylinder seal. In room 5 a clay seal with string impressions and a shell bird amulet were found. In room N6 another cylinder seal and a bitumen plaque were found. In room 14 a stone cosmetic jar and a bowl were found (Delougaz, Hill, and Lloyd 1967, 31). 4.1.1.5 House XLII and House XLIII At level Hss 3 the two dwellings have different names: House XXXIX is called House XLII, but appears to be almost identical to the dwelling at level Hss 3. It is in fact composed of the same two rooms (room 3 and 14). Two graves were present, one in each room, that were vaulted tombs parallel to the long walls of the rooms (Fig. 4.19[a]). In room 3, a copper chisel, a stone bead, and a cosmetic container were found (Delougaz, Hill, and Lloyd 1967, 36).

Chapter Four

156

(a)

(b)

(c)

Fig. 4.19. (a)) Plan of House XLII and XLIIII (Delougaz, H Hill, and Lloyd 1967); (b) F Flow chart of House H XLII; (c) Flow chart of H House XLIII

Central Region Case Study: Khafajah and Tell Asmar

157

House XLII 20 18 16

Surface area (m2)

14 12 surface (m2) grave(s)

10 8 6 4 2 0 14

3 Space number

(a) House XLIII 35

30

Surface area (m2)

25

20

surface (m2) grave(s) jar oven(s)

15

10

5

0 15

19

12

2

O6

N6

14

5

7

Space number

(b) Fig. 4.20. (a) Internal space sizes and features of House XLII; (b) Internal space sizes and features of House XLIII

158

Chapter Four

House XL is called House XLIII at level Hss 3 (Fig. 4.19[a] and 4.19[c]). The area occupied by the dwelling is the same as at the older level, and the plan has only minor alterations. The entrance room was enlarged, one room was eliminated, and spaces 6 and 14 remain the same. The rest of the layout is the same except for some alterations to the dimensions of the spaces. The southwestern side of the house was not found (Delougaz, Hill, and Lloyd 1967, 13), and is only suggested on the plan (see Fig. 4.19). The excavator suggested an entrance from space 6 and no connection between the group of rooms 6, 5, 7, and the rest of the house. If this was the case, the three separated rooms could have comprised a different dwelling or a commercial activity. Three vaulted tombs were found in the house, grave 137 was cut into an oven in room O6, there was a grave in room 7, and a third grave was cut into the wall between court 15 and room 19. The grave cut in the wall was from a later period than grave 137, but covered by the same layer of ash that covered the ruins of level Hss 3 before the rebuilding of the later level (Delougaz, Hill, and Lloyd 1967, 13). In room 2, a stone bowl was found, in room 12 a copper jar and a stone stamp seal were found, in courtyard 15 a shell cosmetic container was found, in room 5 a stone lion-head amulet was found, and in room N6 a stone bowl was found (Delougaz, Hill, and Lloyd 1967, 36). An extensive area of houses was excavated on the northwestern side of House XLIII, but no one complete dwelling plan was exposed; the remains are all fragmentary and of limited use for the analysis (Delougaz, Hill, and Lloyd 1967, 14). All the remains at level Hss 3 were covered by a layer of ash that the archaeologist identified as a sign of discontinuity in occupation (Delougaz, Hill, and Lloyd 1967, 15). 4.1.1.6 House XLV House XLV is a large dwelling of 356 metres2 (Henrickson 1981, 52), organised around a trapezoidal court (space 14) and surrounded by streets and alleys (Fig. 4.21[a]). The house is composed by nine spaces and it is almost complete, the archaeologist reconstructing the southwestern side (Delougaz, Hill, and Lloyd 1967, 15). The entrance from room 6 is the only certain entrance, the other two being conjectural. The house has a

Central Regioon Case Study: Khafajah and T Tell Asmar

159

a peculiar spaatial organisation because itt appears thatt court 14 is accessible only by room m 13 (Fig. 4.221[b]). The ho ouse seems to have been div vided into two parts, one composeed of room 11, which iis a very lon ng space occupying m more than fortty metres2, and d room 2, whiich is an adjaccent room with a roundded corner. The second parrt includes thee rest of the house h and the court. A Apparently, a central courrtyard surrounnded by room ms was a common hoouse plan, buut the spatiaal analysis hiighlights the peculiar organisationn of the circullation. It is ex xceptional to hhave a court connected c with only one room, usuually when theere is a centraal space surrounded by rooms, eachh room openinng towards thee central court rt. Moreover, there t was probably a bbench along thhe northwesterrn wall of thee court, which h does not seem to occuur often. It is possible to seee in the courttyard what loo oks like a circular instaallation, but thhere is no desccription of it inn the excavatio on report. Because of tthis unusual arrrangement, the archaeologisst suggests thaat building XLV was noot an ordinary house h (Deloug gaz, Hill, and L Lloyd 1967, 15 5).

(a)

Chapter Four

160

(b) House XLV 100 90 80

Surface area (m2)

70 60 surface (m2) bench

50 40 30 20 10 0 14

11

4

13

6

2

5

12

1

Space number

(c) Fig. 4.21. (a) Plan of House XLV (Delougaz, Hill, and Lloyd 1967); (b) Flowchart; (c) Internal space sizes and features of House XLV

Because of this peculiarity, it is difficult to make hypotheses on the use of space. In a standard house, rooms 1 and 12 could have been storage spaces because of their relatively small dimensions and position. Room 5 could have been a storage space for the same reasons or a facility for cooking because it was near the entrance, as many cooking facilities were. Rooms

Central Region Case Study: Khafajah and Tell Asmar

161

4 and 13 could have been living rooms but even their dimensions are conjectural. In the courtyard, a bone cap for a cylinder seal and a carnelian bead were found; in room 1 a copper axe and a copper chisel were found together with a cylinder seal impression; in room 2 a copper needle, a cylinder seal, a shell, an amulet, a shell pendant, and a baked-clay spindle whorl were found; and in the long room, number 11, a copper disk, a cylinder seal, and a lapis lazuli bead were found. 4.1.1.7 House XLVI House XLVI is a very large building occupying 528 metres2 and is composed of sixteen spaces (see Fig. 4.22[a]). The building is trapezoidal in shape with a rectangular internal court (space 15), around which five or six rooms opened. The excavators supposed that space 2 could have been a courtyard because of its dimensions. No doorway has been found but the building was not complete. The archaeologists reconstructed a doorway in space 2, but one could have also been in room 30 if the long room was organised like the long room 11 in House XLV. Indeed, room 30 in House XLVI was similar to room 11 in House XLV, as they are both long rectangular rooms around three metres wide. The archaeologists supposed that there was a separation between room 30 and room 22 which was only partially preserved, but even if there was no doorway on the northwestern side of the room there was an entrance from court 2 nearby. Moreover, the presence of a separating wall between space 2 and space 29 is conjectural, and the space could have been a very large courtyard. In either case, room 30 was separated from the main part of the house (courtyard 15 and surrounding rooms). This arrangement suggests that, in the separate section of the house, activities involving people unrelated to the household were carried out. Because of the unusual layout, it is difficult to understand the function of rooms like number 31 and 1, and they might have been storage or facilities rooms related to the activities going on in room 30.

Chapter Four

162

(a)

(b) House XLVI X

60

Surface area (m2)

50

40

surface (m2) grave(s)

30

20

10

0 15

2

3 30

3

29

4

1

22

44:6

18

31

43:6

12

7

16

23

Space number

(c) H XLVI (Deelougaz, Hill, aand Lloyd 1967); (b) Fig. 4.222. (a) Plan of House Flowchart; (c) Inteernal space sizees and features oof House XLVII

Central Region Case Study: Khafajah and Tell Asmar

163

The rest of the house has a deep permeability value (see Fig. 4.22[b]). To reach court 15 it was necessary to pass through four other spaces unless a doorway on the eastern side was not preserved or there was an opening to the alley outside the house in room 3. Rooms 6 and 4 could have been multipurpose spaces used by the household to sleep, eat, and work. Rooms 18 and 7 could have been storage spaces because the doorways appear to be blocked. It is difficult to understand the use of rooms 23 and 16 in the absence of other features. The archaeologists explained the unusual layout of the house and the location of courtyard 15, far away from the entrance, by the presence of earlier developments that influenced the construction of the house (Delougaz et al. 1967, 16). A grave was found in space 2 of House XLVI along with other objects: a copper needle, a stone spindle whorl and a carnelian bead in room 1; a clay spindle whorl, a shell, a stone bead, and a whetstone in space 2; two cylinder seals, a shell pendant, and a stone bowl in room 30; a stone weight and a lapis lazuli bead in room 3; a stone inlay and a carnelian bead in room 4; a cylinder seal impression, a copper pin and chisel, a bone toggle and punch, a stone weight, and a clay bead in room 7; a cylinder seal, an animal horn, copper wire and pin, a stone, and two baked-clay spindle whorls in room 6; a lapis lazuli, a shell and a stone bead, a copper pin, and a stone weight in room 12; an agate bead, a cylinder seal, a shell ring, and a copper nail and pin in court 15; a shell container and ring, and a stone macehead in room 16; and a clay model chariot wheel and a shell stamp-seal foot amulet in room 18 (Delougaz, Hill, and Lloyd 1967, 40– 1). It is possible to see in the courtyard what looks like a circular installation, but there is no description of it in the excavation report. Street number 3 separated House XLVI from House XLV; street 6, perpendicular to street 3, separated House XLVI from the Temple Oval. Street 6 was blocked at the end by two rooms, 24 and 17, which have no connection with House XLVI. The excavators believe that these rooms were related to the temple because they were plastered with bitumen and had drains connected to the outer enclosure wall of the temple (Delougaz, Hill, and Lloyd 1967, 16). There were rooms on the southeast of House XLVI, but it is difficult to determine to which house they belonged because they were partially covered by the remains of the later level Hss 1.

Chapter Four

164

4.1.1.8 Level Hss 1 .

Level Hss 1 is characterised by a complete re-planning at the summit of Mound A (4.5). A 2.60 metre wide enclosure wall was found that limited on three sides (southeast, northeast, and northwest) an area of domestic buildings. The southwestern side of the wall was not present and the archaeologists hypothesise that the area was limited by the wall of the Oval Temple on that side (Delougaz, Hill, and Lloyd 1967, 17). The enclosed private house area occupied 1,700 metres2 and showed signs of planning, as did the at least 35-metre long straight street crossed by perpendicular alleys. This grid defined the orientation of the wall of the houses in the neighbourhood (Delougaz, Hill, and Lloyd 1967, 17). No doorways have been found in the preserved section of the enclosure wall. The “walled quarter” appeared densely built. It was delimited on the east side by the Sin temple and on the western side by the Oval temple, perhaps because of space limitations. The area of the houses was notably smaller compared to those at the earlier levels. 4.1.1.9 House XLVII House XLVII occupied 89 metres2 and is organised in six spaces (see Fig. 4.23[a] and 4.23[b]). The entrance of the house was from a vestibule (2) and it lacked a central court. However, vestibule 2 could have been an accessible area, as was the case for many other entrance rooms in the “walled quarter” (see, for example, House XLVIII, House XLIX, House LI, and House LIII). Room 3, which appeared isolated from the rest of the house, could have been a reception room for guests, a space for a different nuclear family, or an area for business. Room 5 is the biggest in the house, and was probably a multipurpose room for the family; room 4 could have been a storage room because it is small and far away from the entrance; room 7 could have been a distribution room and room 6 another distribution room or a space with facilities like a cooking area. In space 2, a clay model chariot wheel was found, and in space 3 a copper needle and a stone weight were found. In room 4, two cylinder seals were found, as was a silver earring, a pin and a blade in copper, and a stone weight. In room 5 a clay model chariot and bed were found, along with two carnelian beads and a cylinder seal. Room 7 had a copper needle, a miniature stone bowl, a cylinder seal, and a lapis lazuli bead (Delougaz, Hill, and Lloyd 1967, 49).

Central Regioon Case Study: Khafajah and T Tell Asmar

165

(a)

(b) House XL LVII

20 18 16

Surface area (m2) ( )

14 12 10

surface (m2)

8 6 4 2 0 5

3

6

2

7

4

Space numberr

(c) Delougaz, Hill, aand Lloyd 1967 7); (b) Fig. 4.233. (a) Plan of House XLVII (D Flow wchart; (c) Inteernal space sizess and features oof House XLVII

166

Chapter Four

4.1.1.10 Hou use XLVIII Adjacent to House XLVII on the sou uthwestern sidde is House XLVIII, X a 2 building occcupying 108 metres m (Henrrickson 1981, 52) distributeed among seven spaces (see Fig. 4.224[a]). Court 8 served as a vestibule and d could be entered direectly from the southern end d of the main street (Delougaz, Hill, and Lloyd 11967, 18). Thhe central cou urt is surroundded by smaller spaces that could hhave been stoorage spaces (especially roooms 6 and 13 which have higherr permeabilityy values) (seee Fig. 4.24[b]]), or could have h been cooking or w working faciliities or living rooms (especcially room 12 2 which is relatively biigger). Room 1 could havee been the maain living room m for the family becaause it was the biggest. In room 12, a sstone pestle, two t stone weights, andd a lapis lazuuli bead were found. In rooom 4 a copperr pin was found, in rooom 6 a stone weight w and a stone s shell-shaaped bowl weere found, and in room m 13 a clay moodel chariot wh heel was founnd (Delougaz, Hill, and Lloyd 1967, 48). In courrt 8, a stone bead b and a stoone cylinder seal s were found. In rooom 1, three copper pins, a copper braccelet, a shell and shell ring, two sttone faces, a clay c model ax xe, a lapis lazzuli, and a sttone bead were found (Delougaz, Hill, and Lloyd d 1967, 48).

(a)

(b)

Central Region Case Study: Khafajah and Tell Asmar

167

House XLVIII 30

Surface area (m2)

25

20

15

surface (m2)

10

5

0 8

1

12

13

4

5

6

Space number

(c) Fig. 4.24. (a) Plan of House XLVIII (Delougaz, Hill, and Lloyd 1967; (b) Flowchart; (c) Internal space sizes and features of House XLVIII

4.1.1.11 House XLIX House XLVIII had its southwestern wall in common with House XLIX, a smaller dwelling extending over 86 metres2 (Henrickson 1981, 52) and organised in four spaces (see Fig. 4.25[a] and [b]). The entrance is on a long lane with the doorway 14 metres from the main street. The doorway directly accesses what looked like a courtyard of around 16 metres2 (space 9). Two long, narrow rectangular rooms, almost symmetrical, flank the courtyard and both have direct access to it. At the back of the house is another room and a drain plastered with bitumen leads from the southern corner of the room to the outside of the house. The peculiar organisation of the space makes the interpretation of the use of the rooms difficult. All three rooms are sufficiently large to have been multipurpose family rooms where household activities could be carried out. The simple tripartite arrangement is reminiscent of the Mesopotamian prehistoric houses (Delougaz, Hill, and Lloyd 1967, 18). Two cylinder seals were found in room 12 (Delougaz, Hill, and Lloyd 1967, 48). In room 7, a bead, a weight, and a stone bowl were found (Delougaz, Hill, and Lloyd 1967, 48).

Chapter Four

168

(a)

(b) House XLIX X

18

16

Surface area (m2)

14

12

10 surface (m2) 8

6

4

2

0 9

12

7

13

er Space numbe

(c) Fig. 4.255. (a) Plan of House H XLIX (Deelougaz, Hill, aand Lloyd 1967); (b) Flowchart; (c) Inteernal space sizees and features oof House XLIX X

Central Region Case Study: Khafajah and Tell Asmar

169

On the northeastern side of the house is a long rectangular room that the excavators considered independent from House XLIX because it is separated from it by a very thick wall (Delougaz, Hill, and Lloyd 1967, 18). The isolated room is reached by a long lane, but in a later phase the access was blocked (Delougaz, Hill, and Lloyd 1967, 18). Delougaz, Hill, and Lloyd’s interpretation is that the thicker wall sustained a second floor accessed from a stairway ending where the alley appears to be blocked, but the long room 10 was accessed nonetheless (Delougaz, Hill, and Lloyd 1967, 18). 4.1.1.12 House L At 169 metres2 (Henrickson 1981, 52), House L is the largest dwelling preserved in the “walled quarter.”. The irregular plan was probably composed of eight spaces (see Fig. 4.26[a] and 4.26[b]). The house is accessed from an alley perpendicular to the main street. Room 11 was defined as a “vestibule” by the excavators, but it may have had other functions; for example, many Mesopotamian houses had cooking facilities near the entrance. The southern wall of the room was not preserved, like the doorway, and it probably had a passage for the internal courtyard. The open space was around 27 metres2 and was surrounded by rooms. Room 24 was a long and narrow space at the back of the house, probably used to receive guests and for household activity. Room 21 could have been another multipurpose room for household activities. Room 8 and 14, being quite small and secluded (see Fig. 4.26[b]), could have had a storage function. In room 10, many pottery vessels were found in situ (Delougaz, Hill, and Lloyd 1967, 18). In Room 6 a pestle, a weight, a pendant, and a bead all made from stone were found, along with a flint blade and a shell containing red pigment. A clay model of a boat fragment was found in room 24 (Delougaz, Hill, and Lloyd 1967, 48). In room 9 it is possible to see what looks like a circular installation, but there is no description of it in the excavation report. House L was flanked on the northeastern and southeastern sides by a series of rooms, mostly single spaces opening directly onto the streets. Because of the street access, excavators disregarded a domestic function for these rooms (Delougaz, Hill, and Lloyd 1967, 18–19)—a commercial function seems to be a more likely hypothesis.

Chapter Four

170

(a)

(b) House L

30

Surface area (m2) ( )

25

20

surface (m2) jars

15

10

5

0 22

24

11

21

10 0

9

8

14

Space numberr

(c) Fig. 4.26. (a) Plan of House L (Delougaz, Hill, H and Lloyd 1967); (b) Flow wchart; (c) Internal space s sizes and d features of Hoouse L

Central Regioon Case Study: Khafajah and T Tell Asmar

171

4.1.1.13 House LI de of the alleey, lies Housee LI. The In front of House L, on the other sid house is 1224 metres2 (H Henrickson 19 981, 52) with four spaces (see Fig. 4.27[a] and 4.27[b]). Deloougaz defined d the entrancee, space 2, as a “central room,” but in my opinioon it could haave been an oopen space (D Delougaz, Hill, and Llloyd 1967, 199). To the easst of space 2 is a long an nd narrow room (43:8)), and on thee other side of o space 2 weere two almo ost square interconnectted rooms. This T simple organisation makes it diffficult to formulate hyypotheses on the use of spaace, as any sppace in House LI could have been uused for any household acctivities. Therre were no feeatures to help the inteerpretation, as only a copperr pin was founnd in space 2. House LI shhares a block with w a non-ressidential buildding composed d of three rooms with a possible com mmercial funcction (see Fig.. 4.27).

(a)

(b)

Chapter Four

172

House LI 25

Surface area (m2)

20

15 surface (m2) 10

5

0 2

1

44:8

43:8

Space number

(c) Fig. 4.27. (a) Plan of House LI (Delougaz, Hill, and Lloyd 1967); (b) Flowchart; (c) Internal space sizes and features of House LI

4.1.1.14 House LIII House LIII opens onto the main street, a dwelling of around fifty-five metres2 (Henrickson 1981, 52) with five spaces (see Fig. 4.28[a] and 4.28[b]). It is symmetrically organised with rooms around the central court (space 3). A hearth is present in room 12, indicating that it was used as a kitchen. The other rooms could have been multipurpose living rooms or, in the case of space 23, storage rooms. Two lapis lazuli and a carnelian bead, a whetstone, a stone blade and a copper blade, and copper bowls where found in space 23, while a copper pin and a clay spindle whorl were found in space 3 (Delougaz, Hill, and Lloyd 1967, 48).

Central Regioon Case Study: Khafajah and T Tell Asmar

(a)

173

(b) House LIII L

14

12

Surface area (m2)

10

8 surface (m2) heart(s) 6

4

2

0 3

11

12

5

23

Space numberr

(c) F Fig. 4.28. (a)) Plan of Housee LIII (Delougazz, Hill, and Llooyd 1967); (b) Flowchart; (c) Internal space sizes and d features of Hoouse LIII

174

Chapter Four

The “walled quarter” is generally well preserved, and six houses have complete ground plans. This abundance of information allows for the consideration of general/common characteristics. As previously stated, the neighbourhood was well planned with a straight main street and perpendicular alleys. The houses were smaller than the dwellings at the previous levels, the majority of them smaller than 130 metres2, and half of them less than ninety metres2 (Henrickson 1981, 52). Three houses had a tripartite organisation (XLIX, LI, and LIII), and four houses had at least one long and relatively narrow rectangular room (room 5 in House XLVII, rooms 7 and 12 in House XLIX, room 24 in House L, room 43:8, and 44:8). Delougaz hypothesised that the necessity of a walled quarter inside an already walled city could have been protection for a “local garrison established by recent conqueror” (Delougaz, Hill, and Lloyd 1967, 20) from the inhabitants of the city who were from a different ethnic group to the new conquerors. In support of this hypothesis, the “walled quarter” was built on top of the previous Nintu Temple and possibly also the small temple with no respect for the previous traditions (Delougaz, Hill, and Lloyd 1967, 20).

4.1.2 Khafajah Mound B and C A fortress called Dur-Samsuiluna, which was built by Samsuiluna, the son of Hammurabi, was discovered on this site (Harris 1955, 33). In mound C at least one house has been identified, but the plan and the section of this sector were either never made or lost. One Old Babylonian cylinder seal, four pottery vessels dating to the Kassite Period, and a bronze sickle from the Hammurabi Period were found in this area.

4.1.3 Khafajah Mound D Mound D is occupied by a fortified citadel. The site has not been excavated extensively, but numerous isolated soundings have been carried out. At the centre of the citadel is a temple of Sin and the surrounding area seems to have been almost entirely occupied by private houses. Tablets with cuneiform texts were found in the temple archive (Harris 1955, 33). The texts give accounts of public activities officially carried on by the temple, but do not give information about the private lives of the citizens of Tutub (Harris 1955, 37). The main street identified probably led from the gateway to the temple. Several doorways opening off the northern side

Central Region Case Study: Khafajah and Tell Asmar

175

of this long street were of a size that indicated that the buildings into which they led were private houses. The excavators inferred the presence of narrow alleys and passageways between the houses or groups of houses, as was common for other settlements of the same period. No passageways seemed to have existed along the inner face of the thick fortified walls of the citadel. The only complete house excavated was near the southwestern corner of the citadel. The floor plan appears to be the standard house plan with a central court that leads to a series of rooms of different sizes. The entrance was from the southern side through a vestibule, and because of its size and position room 4 seems to have been a living room.

4.2 Tell Asmar Like Khafajah, Tell Asmar is located in the Diyala region. The two archaeological sites were first excavated at the same time by a team of the University of Chicago in the 1930s (Frankfort, Jacobsen, and Preusser 1932; Frankfort 1933; Frankfort 1934; Frankfort 1935; Frankfort 1936). Tell Asmar is located fifty miles northeast of Baghdad in the area that takes its name from the Diyala River, a tributary of the Tigris (Frankfort, Jacobsen, and Preusser 1932, 1). At the site the excavators found a palace of the Isin-Larsa Period; a temple dedicated to Su-Sin, king of Ur, by his vassal an ensi of Eshnunna; a palace from the Ipiq-Adad Period and an “Audience Hall” built by NaramSin, another ruler of Eshnunna. The earlier remains were a temple of Abu, a palace and the private houses from the Protoliterate to the Akkadian periods. In the Akkadian and Ur II periods, the city spread out in order to make room for large public buildings and further houses (Martin 1988, 116). The latest remains at Tell Asmar were from the Old Babylonian Period (Harris 1955, 32). The site of Tell Asmar has been identified with the ancient city of Eshnunna (Frankfort, Jacobsen, and Preusser 1932, 3–4).

176

Chapter Four

Fig. 4.299. Diyala Regioon (Frankfort, Jaacobsen, and Prreusser 1932, Fig. F 1)

Central Regioon Case Study: Khafajah and T Tell Asmar

177

Fig. 4.30. Coontour map of Tell T Asmar (Deelougaz, Hill, annd Lloyd 1967, plate 23)

178

Chapter Four

In the archives, around six hundred clay tablets have been found dating between 2300–1900 BC (Frankfort, Jacobsen, and Preusser 1932, 12–13, 24). The texts identified Tell Asmar as the capital of a small state ruled by an independent Akkadian dynasty testified by Akkadian names and king lists (Frankfort, Jacobsen, and Preusser 1932, 26, 40–1). Hill explains that it was difficult to divide the domestic building’s area into occupational levels because there was an unbroken continuity in the occupation without general episodes of destruction and, “each house had his own history, not necessarily paralleled by that of its neighbours” (Delougaz, Hill, and Lloyd 1967, 143). As for Khafajah, Henrickson studied the settlement pattern of the late ED in Tell Asmar and found that strata Vb and Vc had comparable house plans that suggest a continuous occupation, while stratum Va appeared to be a new phase of rebuilding (Henrickson 1981, 50).

4.2.1 Stratum Vc The archaeologists dated Stratum Vc to the Early Dynastic Third (Delougaz Hill, and Lloyd 1967, 164). Hill attributed four houses to Stratum Vc; three of them have complete ground-floor plans, and for that reason they have been analysed below. The fourth house has not been analysed because it is only partially preserved.

Central Regioon Case Study: Khafajah and T Tell Asmar

179

Figuree 4-

Fig. 4.31. S Stratum Vc of Tell T Asmar (Dellougaz, Hill, andd Lloyd 1967, plate p 24)

180

Chapter Four

Fig. 4.32. S Stratum Vb of Tell T Asmar (Dellougaz, Hill, annd Lloyd 1967, plate p 25)

4.2.1.1 House H I House I is composed off nine spaces (see Fig. 4.3 ) on a surfacce of 156 metres2 (Hennrickson 19811, 52).

Central Region Case Study: Khafajah and Tell Asmar

181

(a) House I Stratum Vc 25

Surface area (m2)

20

15

surface (m2) 10

5

0 J40

38

41

49

47

H40

48

42

43

Space number

(b) Fig. 4.33. (a) Flowchart; (b) Internal space sizes and features of House I Stratum Vc

The ground floor level is complete apart from the external wall on the southwestern side, but the layout is clear. The house is flanked by Middle

182

Chapter Four

Road, where the entrance is. On the northeastern side it shares a wall with House II. Based on the plan (see Fig. 4.31), the other two sides appear to have been occupied by houses that have not been excavated. The only entrance is in room 42. The house is organised into eight rooms surrounding space J40, which had the position and dimensions of an internal courtyard measuring twenty-three metres2. Because of its dimensions (14.5 metres2), room 38 could have been used as a living room for the inhabitants of the house, a place where food consumption, sleep, and work took place. Space 43 could have been a cooking facility or a storage room based on its small dimensions and high depth value. Room 48 is small and far away from the entrance, and has the aspect of a storage space. It is difficult to perceive the uses of rooms 41, 40, and 47; they seem too large to be storage or facilities spaces, and therefore could have been other living rooms. In room 47 a shell containing red pigment and cylinder seal were found; in room 48 a lapis lazuli pendant was found; in space J40 two cylinder seals were found; and a stone weight was found in room 41. 4.2.1.2 House III House III was composed of four spaces and occupied 66 metres2 (Henrickson 1981, 52). It is a complete unit adjacent on the southwestern side to House III and on the northern side to House IV. House IV was only partially excavated and the area on the northeastern side was not excavated at all (see Fig. 4.35). The house is reached from Middle Road through an entrance in room 53, a space that contained a bread oven in the eastern corner (Delougaz, Hill, and Lloyd 1967, 165). The rest of the building did not follow a standard plan, the author observing that House III is “a poorer building fitted inbetween larger dwellings.” The archaeologists consider room 66 to have been used as the “main room,” although it is not larger than the others spaces (Delougaz, Hill, and Lloyd 1967, 165). Room 66 has a niche built on the northwestern wall, but no other features to suggest the use of the room as a domestic chapel. The only other feature was a pot on the floor in the northern corner belonging to a second Vc occupation level 0.40 metres above the first one (Delougaz, Hill, and Lloyd 1967, 165).

Central Region Case Study: Khafajah and Tell Asmar

183

(a) House III Stratum Vc 9

8

Surface area (m2)

7

6

5

surface (m2) jar oven(s)

4

3

2

1

0 50

53

76

66

Space number

(b) Fig. 4.34. (a) Flowchart; (b) Internal space sizes and features of House III Stratum Vc

It is very difficult to assign a function to the remaining rooms with their unusual layout without any features.

Chapter Four

184

4.2.1.3 House II House II is a large building of 196 metres2 composed of twelve spaces (see Fig. 4.35) (Henrickson 1981, 52).

(a) House II Stratum Vc 25

Surface area (m2)

20

15

surface (m2) oven(s) basin bench jar

10

5

0 27

79

52

37

16

59

64

36

41

13

77

68

Space number

(b) Fig. 4.35. (a) Flowchart; (b) Internal space sizes and features of House II Stratum Vc

House II is adjacent to House I and flanked by Middle Road. On the northeastern side it is flanked by House III, and on the northwestern side by House IV and another building that has not been excavated (see Fig. 4.31).

Central Region Case Study: Khafajah and Tell Asmar

185

House II is the largest found in Stratum Vc. The ground-floor layout is complete and of an irregular shape. The only entrance to the building is from Middle Road in space 52, and it may be supposed that room 52 was previously part of House III and subsequently added to House II (Delougaz, Hill, and Lloyd 1967, 155). The northern half of room 52 was coated with gypsum and a pot was found on the floor near the entrance, Delougaz, Hill, and Lloyd suggest a use of water in the room and food preparation (Delougaz, Hill, and Lloyd 1967, 155). Through a small room (space 68), the entrance is connected with rooms 41, 27, and 64. In room 41 were cooking facilities, comprising a bread oven of 1.75 metres long, 1.40 metres wide, and about 0.5 metres tall that occupied the majority of the space, and another, smaller bread oven. The house seems to have been organised around an almost square large space (27) surrounded by smaller rooms. Space 27 is large and central with the characteristics of an internal courtyard. An area plastered in gypsum (a rectangle of 0.8 metres wide and 3 metres long) was found in the eastern corner of space 27. A basin-like circular depression (40 centimetres in diameter and 18 centimetres deep) was at the northeastern end of the gypsum rectangle, and the gypsum plaster extended 12 centimetres along the foot of the wall (Delougaz, Hill, and Lloyd 1967, 155). This feature suggests the use of water. A bench of unbaked bricks 0.6 metres high was found in the southern corner (Delougaz, Hill, and Lloyd 1967, 155). A pot was found in the southern corner of room 64. A small bench was found in the eastern corner of room 79, which was a relatively large space of around ten metres2 that could have been used as a family room for eating, sleeping, and working. Room 77 could have been a storage room, its alignment suggesting that it originally belonged to House VI (Delougaz, Hill, and Lloyd 1967, 156). It was possible to follow the evolution of House II at the more recent levels. At level Vb a preparation for rebuilding was carried out: the walls were removed to a height of 0.5–0.6 metres above the Vc floors (Delougaz, Hill, and Lloyd 1967, 156). The rebuilding took place in the second Early Dynastic III. House II Vb (see Fig. 4.32) incorporated almost all of House III Vc, except for rooms 76 and 77.

186

Chapter Four

Fig. 4.36. Reeconstruction of o House II Stratum Vb of Telll Asmar (Delougaz, Hill, and Lloyd 196 67, plate 35)

The only enntrance to the building is frrom Middle R Road through space 52, as at the pprevious levell. A dotted line on the pplan indicates that the doorway waas widened duuring a second dary occupatioon (Delougaz, Hill, and Lloyd 1967,, 156). There was w a bin in th he northern coorner of room m 52 and a pot on the flloor along thee northeastern wall, and the archaeologistts suggest that this room m was used foor food preparration.

Central Region Case Study: Khafajah and Tell Asmar

187

(a) House II Stratum Vb 25

Surface area (m2)

20

15

surface (m2) bin(s) oven(s) bench jar

10

5

0 27

10

13

52

53

16

59

50

64

66

41

68

Space number

(b) Fig. 4.37. (a) Flowchart; (b) Internal space sizes and features of House II Stratum Vb

188

Chapter Four

In the adjoining room 68 a pot was partially buried in the floor, and the archaeologists suggest that it was also used for food preparation (Delougaz, Hill, and Lloyd 1967, 156–7). An arched doorway connects room 68 with room 52. There is a window in the wall between room 68 and 64 built around 1.70 metres above the floor, which is hypothesised to be for ventilating room 64 and to inspect visitors before they entered the more private part of the house (Delougaz, Hill, and Lloyd 1967, 157). There was also a hole, perhaps for a pivot stone, by the jamb of the doorway between space 68 and space 27 (Delougaz, Hill, and Lloyd 1967, 157). Room 68 has kitchen facilities built on top of the facilities of the previous level and similar to them (Delougaz, Hill, and Lloyd 1967, 157). The doorways between space 27 and spaces 59, 64, and 10 are arched. The other doorways in space 27 had been destroyed during later rebuilding but the archaeologists supposed that they were also arched, as shown in the hypothetical reconstruction (see Fig. 4.3). The doorways were high at approximately 1.50 metres, and the archaeologists note that, “it was necessary to stoop in order to pass through them, as is the case in many houses in modern villages in the neighborhood” (Delougaz, Hill, and Lloyd 1967, 158). A pier, added to the main room later, was interpreted as an indication of roofing, supporting Delougaz’s theory that Tell Asmar houses did not have internal courtyards (Delougaz, Hill, and Lloyd 1967, 158). As at the previous level, there was a bench in the southern corner (Delougaz, Hill, and Lloyd 1967, 159). In space 27, a cylinder seal was found (Delougaz, Hill, and Lloyd 1967, 212). Room 59 remained as in Stratum Vc, with the addition of a bin in the eastern corner (Delougaz, Hill, and Lloyd 1967, 159). Room 10 is a newer large space that replaced two earlier rooms, with two large bread ovens suggesting its use as a cooking facility (Delougaz, Hill, and Lloyd 1967, 159). The archaeologists suggest that the preparation or storage of food also took place in room 64 because they found large jars, thick-walled ribbed pots, goat horns, and a quern with a rubbing-stone (Delougaz, Hill, and Lloyd 1967, 159). Room 64 was the only connection between the main house and rooms 66, 50, and 53. Those rooms had a deep permeability value; to reach room 53, for example, it was necessary to pass through six spaces (see flowchart). There were no indications of the use of these rooms, but because of their position the archaeologists suggest that they could be private living quarters (Delougaz, Hill, and Lloyd 1967, 159).

Central Region Case Study: Khafajah and Tell Asmar

189

It is difficult to understand what room 16 was used for, as no door has been found to access it. A large storage jar was set in the floor (Delougaz, Hill, and Lloyd 1967, 159) and a clay bead was found (Delougaz, Hill, and Lloyd 1967, 212). It is possible that the room could have been used for storage and closed seasonally. House II was rebuilt in stratum Va (see Fig. 4.42), its floor being a metre higher than that of Vb, but the majority of the walls are in the same position as the previous stratum (Delougaz, Hill, and Lloyd 1967, 159). The entrance is between room 41 and the wall from this space and room 68 has been removed (Delougaz, Hill, and Lloyd 1967, 160). There is access to room 52 from room 68. In room 41, the cooking facilities from stratum Vb have been covered by the new floor, but a bread oven has also been added (Delougaz, Hill, and Lloyd 1967, 160).

(a) Fig. 4.38. (a) Flowchart of House II Stratum Va

Chapter Four

190

House II stratum Va 25

Surface area (m2)

20

15 surface (m2) oven(s) 10

5

0 27

10

13

53

16

52

59

41

66

50

64

Space number

(b) Fig. 4.38. (b) Internal space sizes and features of House II Stratum Va

The archaeologists supposed that room 10 was not in use during Va because, although the door was not blocked, no floor has been found for stratum Va (Delougaz, Hill, and Lloyd 1967, 160). No doorway was found between room 64 and 66. It has been reconstructed on the basis of stratum Vb as well as the doorways on the southeastern side of rooms 50 and 66 (Delougaz, Hill, and Lloyd 1967, 160). In some cases, traces of an occupation level between stratum Vb and stratum Va have been found, but the remains could not be traced throughout the structure (Delougaz, Hill, and Lloyd 1967, 160). In the following stratum, IVa, there was a rebuilding of House II dating to the Akkadian Period. The house was rebuilt one metre above the previous stratum Va and enlarged (Delougaz, Hill, and Lloyd 1967, 161–3). In this thesis, the Akkadian Period of Tell Asmar is not going to be analysed because all the other sites taken into consideration belong to the Early Dynastic Period. This thesis is going to be limited to the ED Period in order to present consistent and comparable evidence. There are also

Central Region Case Study: Khafajah and Tell Asmar

191

remains of later levels in House II in the later strata III, II, and I (Delougaz, Hill, and Lloyd 1967, 163–4).

4.2.2 Stratum Vb Stratum Vb was also dated to the Early Dynastic III (Delougaz, Hill, and Lloyd 1967, 165). The excavations revealed the complete ground plan of four houses situated at the northern side of Middle Road (see Fig. 4.32) and two others that were only partially excavated. 4.2.2.1 House I House I in stratum Vb was rebuilt on top of House I of the previous stratum. Only a few changes were made to the ground plan (Delougaz, Hill, and Lloyd 1967, 166). The surface of the ground plan was 140 metres2 and the entrance was shifted to the middle front room, and in this way access to the “main space,” 33, was more direct (see Fig. 4.). In the entrance, room 32, a cylinder seal was found (Delougaz, Hill, and Lloyd 1967, 212).

(a) Fig. 4.39(a). Flowchart of House I Stratum Vb

Chapter Four

192

House I Stratum Vb 20 18 16

Surface area (m2)

14 12 surface (m2) jar altar

10 8 6 4 2 0 J33

J31

30

32

29

H31

38

34

32

H33

Space number

(b) Fig. 4.39(b). Internal space sizes and features of House I Stratum Vb

Space 33 gives access to rooms 30 and 31 with a modification of the doorway positions from the previous stratum and a variation of the room surfaces. A fragment of cylinder seal was also found in space 33 (Delougaz, Hill, and Lloyd 1967, 212). Room 29 also opens onto the main space (33) and contained a large storage jar and what Hill defined as an “altar”—a structure of unbaked bricks, quadrantal and situated in the western corner of the room. “Altars” found at Tell Asmar were usually decorated features that suggested the hypothesis of a non-utilitarian purpose (Delougaz, Hill, and Lloyd 1967, 151, 166). In room 29 a lapis lazuli beetle amulet and a stone weight were found (Delougaz, Hill, and Lloyd 1967, 212). The small room 32 was an addition in stratum Vb and looks like a storage space because of its dimensions. Room 34 could also have been for storage in stratum Vb since there was no entrance, as in the previous stratum. A flint blade set in bitumen was found in this room (Delougaz, Hill, and Lloyd 1967, 212).

Central Region Case Study: Khafajah and Tell Asmar

193

Adjacent to House I on the northeastern side was House II, the “Arched House” described earlier, while House VI and House V were on the southwestern side.

Fig. 4.40. Reconstruction by Viviana Russo: Stratum Vb, remains of House I, reconstruction of elevation and features of House II and reconstruction of House IV completed with roof.

4.2.2.2 House V Hill thinks that House V and House VI occupied the site of an earlier building because their combined area was equal to the surface of House II (Delougaz, Hill, and Lloyd 1967, 165). An alternative explanation might be that houses V and VI each occupied half a lot while House I occupied one lot. The uniformity of measures of land within a neighbourhood together with symmetry in wall alignments that extend across the partition

Chapter Four

194

(a) Fig. 4.41(a). Flowchart of House V Stratum Vb House V Stratum Vb 12

Surface area (m2)

10

8

surface (m2) bin(s)

6

4

2

0 30

41

34

30

40

35

Space number

(b) Fig. 4.41(b). Internal space sizes and features of House V Stratum Vb

Central Region Case Study: Khafajah and Tell Asmar

195

walls of individual houses is usually a sign of urban planning (see Matney, Algaze, and Pittman 1997, 70). The “coincidence” of the sum of the areas of houses V and VI equalling the area of House I can be seen as a consequence of urban planning. House V was composed of six spaces, and none of them seems to have been a central room around which the house was organised (see Fig. 4.40). The entrance was from Middle Road to room 30, which is also the largest space of the house. Room 42 had a function of distribution because it was connected with other three spaces. Room 35 had the deepest permeability value of the house, and to reach room 35 it was necessary to pass through three spaces. This room was also the smallest in the building and had the characteristics of a storage space, an interpretation that seems to be confirmed by the presence of a square bin of unbaked bricks built against the northeastern wall (Delougaz, Hill, and Lloyd 1967, 165). The function of the other spaces is difficult to interpret without the presence of installations and with the non-standard layout of the house. Rooms 34 and 41 could have been family rooms for sleeping, eating, and working because they are relatively larger, while room 40 could have been a storage space or another family room. 4.2.2.3 House VI House VI is on the northeastern side of House V. It is composed of six spaces without a central space around which to organise the others rooms (Delougaz, Hill, and Lloyd 1967, 165). The entrance is from West Street to room 45, which is the largest room and it was equipped with a bread oven and a rectangular bin of unbaked bricks (Delougaz, Hill, and Lloyd 1967, 165). A copper ring was found in this room (Delougaz, Hill, and Lloyd 1967, 212). Room 38 was also used for cooking because a second bread oven was found here (Delougaz, Hill, and Lloyd 1967, 165). The archaeologists also found a cylinder seal in this space (Delougaz, Hill, and Lloyd 1967, 212). No features were found in the other four rooms—these spaces were smaller than the two rooms in front (Delougaz, Hill, and Lloyd 1967, 166).

Chapter Four

196

Room 43 had the deepest permeability value in the house, and to reach room 43 it was necessary to pass through four spaces.

(a) Fig. 4.42(a). Flowchart of House VI Stratum Vb House VI Stratum Vb 14

12

Surface (m2)

10

8

surface (m2) bin(s) oven(s)

6

4

2

0 45

38

43

44

37

36

Space

(b) Fig. 4.42(b). Internal space sizes and features of House VI Stratum Vb

Central Regioon Case Study: Khafajah and T Tell Asmar

197

4.2.3 Strattum Va m Va to the Prrotoimperial P Period (Delougaz, Hill, Jacobsen assigns Stratum nrickson neveertheless consiiders it to and Lloyd 1967, 145 and 166), but Hen D Perriod becausee it is Pre--Sargonid belong to the Early Dynastic (Henricksonn 1981, 50). There is som me continuity between strattum Vb and sstratum Va. Th he streets around whicch the area iss organised are a still in plaace: Middle Road, R the principal strreet, and Wesst Street, a peerpendicular ssmaller street (see Fig. 4.42). The excavaated area in stratum Va is i much largeer, comprisin ng fifteen houses that were almostt completed and a some unccompleted plans were uncovered bby the excavattion on the sou utheastern sidde of Middle Road. R The limit of the excavation was w East Road d, an almost straight street running north-south,, and South Rooad, a curved prosecution oof East Road.

Fig. 4.43. S Stratum Va of Tell T Asmar (Dellougaz, Hill, andd Lloyd 1967, plate p 26)

198

Chapter Four

The archaeologists interpret stratum Va as a period of general decline, a sign of which is the presence of smaller houses and the repairs to House II (Delougaz, Hill, and Lloyd 1967, 144, 166). The presence of numerous smaller houses in the southeastern area of Middle Road could also have occurred in stratum Vb, although in some cases the excavators distinguished the outlines of the unexcavated houses below that were larger than the houses in stratum Va (Delougaz, Hill, and Lloyd 1967, 167). Moreover, there are also large houses in stratum Va organised along East Road and South Road. The area on the northwestern side of Middle Road, in particular houses I, II, V, VI, and VIII, remains the only one with signs of urban planning. It shows evidence of perpendicular streets, uniform measures of land, and symmetry in wall alignments that extended across the partition walls of single houses. The other areas, starting from the incomplete excavation area on the northwestern side of houses VI and I, lack signs of planning (Delougaz, Hill, and Lloyd 1967, 167). 4.2.3.1 House I There were only few changes between House I in stratum Vb and stratum Va. A bread oven was found at the entrance which, in a later phase, was blocked, a new entrance being opened in room 3, and a bread oven was added to that room at the same time (Delougaz, Hill, and Lloyd 1967, 167). These details seem to confirm that the preferred position for cooking facilities was near the entrance.

Central Region Case Study: Khafajah and Tell Asmar

199

(a) Fig. 4.44(a). Flowchart of House I Stratum Va House I Stratum Va 20 18 16

Surface area (m2)

14 12 surface (m2) oven(s)

10 8 6 4 2 0 2

3

1

14

13

35

17

10

15

9

Space number

(b) Fig. 4.44. (b) Internal space sizes and features of House I Stratum Va

200

Chapter Four

The use of space in the remaining rooms was not very different from the previous levels. Space 2 had the characteristics of a central courtyard, rooms 13 and 14 were the largest and could have been used for working, sleeping, and family gatherings. Rooms 9, 10, and 35 could have been storage spaces. A cylinder seal was found in room 10 (Delougaz, Hill, and Lloyd 1967, 214). A shell bead was found in room 35 (Delougaz, Hill, and Lloyd 1967, 213). It is difficult to understand the function of the long and relatively narrow room 17. Several objects were found in it (e.g. a cylinder seal, a stone bead, a stone weight, and clay model horns) (Delougaz, Hill, and Lloyd 1967, 214). In an earlier phase room 15 was connected with House V, which was part of House I, and in a later phase that door was closed and House V became independent (Delougaz, Hill, and Lloyd 1967, 166). 4.2.3.2 House V House V in stratum Va is a small building of 48 metres2, and together with houses VII and I it occupies the place of the former House V of stratum Vb, confirming the tendency for smaller houses in stratum Va (Delougaz, Hill, and Lloyd 1967, 166). House V was in an earlier phase part of House I, and the connecting doorway between the two houses was later blocked and a doorway opening onto Middle Road was created. In the later configuration, House V was a small dwelling with three rooms, and the use of space in such a simple layout was usually organised with a multifunctional room, which was the largest, in the front with small storage spaces behind. In room 6 two bread ovens were found (Delougaz, Hill, and Lloyd 1967, 166). Several objects were found in room 6: a cylinder seal, a copper blade, a stone pendant, and a clay plaque fragment. A bead was found in Room 6 (Delougaz, Hill, and Lloyd 1967, 214).

Central Region Case Study: Khafajah and Tell Asmar

201

(a) House V stratum Va 20 18 16

Surface area (m2)

14 12 surface (m2) oven(s)

10 8 6 4 2 0 6

5

4

Space number

(b) Fig. 4.45. (a) Flowchart; (b) Internal space sizes and features of House V Stratum Va

4.2.3.3 House VI House VI in stratum Va occupies eighty-two metres2—almost the same surface area as House VI in stratum Vb, but with a partially different layout. The entrance to the building is from West Street to room 24. The room was a kitchen with an oven in the western corner and a pot set into

Chapter Four

202

the floor in the northern corner surrounded by a low screen wall of unbaked bricks (Delougaz, Hill, and Lloyd 1967, 166–7).

(a) House VI stratum Va 25

Surface area (m2)

20

15 surface (m2) oven(s) jar 10

5

0 3

24

15

22

8

Space number

(b) Fig. 4.46. (a) Flowchart; (b) Internal space sizes and features of House VI Stratum Va

Central Region Case Study: Khafajah and Tell Asmar

203

Space 3 is the largest, having a central position, and all the other rooms open towards it. There was a pot of about sixty centimetres in diameter near the centre of the space and a bin in the western corner with another installation that the archaeologists defined as a “semi-circular table.” They hypothesise that the bin was for grain storage and the pot was for water (Delougaz, Hill, and Lloyd 1967, 167). Many objects were found in room 3: a stone bowl, a copper tool, a lapis lazuli, paste beads, a shell ring for a belt, two stone weights, and a stone pendant (Delougaz, Hill, and Lloyd 1967, 213). No features were present in the other rooms so it is difficult to make hypotheses on their use. They could have been storage spaces if the household activities were limited to the main space (3). Many objects were found in room 8: a faience hut symbol, a faience bead, a cylinder seal, and a clay model horn (Delougaz, Hill, and Lloyd 1967, 214). In room 15 a copper pin with a lapis lazuli head was found along with a pierced shell, a cylinder seal, and a hematite weight (Delougaz, Hill, and Lloyd 1967, 214). In room 22 there was a group of shells and pin fragments (Delougaz, Hill, and Lloyd 1967, 214). 4.2.3.4 House VIII House VIII is a small house of forty-four metres2 composed of three rooms (Henrickson 1981, 52). The entrance is from Middle Road into room 26, which is the largest room, equipped with a bread oven in the southern corner (Delougaz, Hill, and Lloyd 1967, 165). Room 26 was probably a multifunction room where the majority of household activities took place; room 16 was the smallest room and the farthest from the entrance. It was probably a storage room. A clay model horn, a copper vanity set and case, and a cylinder seal were found in it (Delougaz, Hill, and Lloyd 1967, 214). It is difficult to understand the function of room 25. It seems too large to be a storage room, and was probably another room used for sleeping, eating, and/or working.

Chapter Four

204

(a) Fig. 4.47(a). Flowchart of House VIII Stratum Va House VIII Stratum Va 14

12

Surface area (m2)

10

8 surface (m2) oven(s) 6

4

2

0 26

25

16

Space number

(b) Fig. 4.47(b). Internal space sizes and features of House VIII Stratum Va

Central Region Case Study: Khafajah and Tell Asmar

205

4.2.3.5 House X The excavator explains that houses X, XI, and XII once formed a single house (Delougaz, Hill, and Lloyd 1967, 167). In stratum Va, House X is a small dwelling composed of four spaces occupying 54 m2 (Henrickson 1981, 52). The entrance was from Middle Road into room 28. Space 13 is the largest space and presumably a multipurpose room where the household activities took place. Rooms 23 and 27 at the back were the most private. Rooms 23 and 27 could have been storage spaces because of their small dimensions and position. In room 23, a baked-clay female figurine and stone duck weight were found (Delougaz, Hill, and Lloyd 1967, 216).

(a)

(b)

(c)

(d)

(e)

(f)

Chapter Four

206

(g)

(h)

Fig. 4.48. (a) Flowchart of House X; (b) Flowchart of House XI; (c) Flowchart of House XII; (d) Flowchart of House XIII; (e) Flowchart of House XIV; (f) Flowchart of House XV; (g) Flowchart of House XVI; (h) Flowchart of House XVII House X Stratum Va 14

12

Surface area (m2)

10

8 surface (m2) 6

4

2

0 13

28

23 Space number

(a)

27

Central Region Case Study: Khafajah and Tell Asmar

207

House XI Stratum Va 14

12

Surface area (m2)

10

8 surface (m2) 6

4

2

0 29

25

26

8

Space number

(b) House XII Stratum Va 16

14

Surface area (m2)

12

10

8

surface (m2)

6

4

2

0 24

23 Space number

(c)

Chapter Four

208

House XIII Stratum Va 12

Surface area (m2)

10

8

6

surface (m2)

4

2

0 35

26

19

Space number

(d) Fig. 4.49. (a) Internal space sizes and features of House X; (b) House XI; (c) House XII; (d) House XIII

4.2.3.6 House XI House XI is slightly larger than House X of sixty metres2 (Henrickson 1981, 52). It is composed of four spaces: a small entrance, room 8, opening onto Middle Road; and two large spaces, 29 and 25, probably multipurpose rooms. In room 25, a cylinder seal was found (Delougaz, Hill, and Lloyd 1967, 216). Room 26 is very small. It had a doorway in stratum IVb, but in stratum Va this was blocked by unbaked-clay sills (Delougaz, Hill, and Lloyd 1967, 167). The excavators could not find a purpose for room 26 (Delougaz, Hill, and Lloyd 1967, 167). In my opinion, it could have been a storage space because of its small dimensions—it has been attested in the ethnographic records that storage spaces were sometimes seasonally or periodically closed (see chapter two section 4).

Central Region Case Study: Khafajah and Tell Asmar

209

4.2.3.7 House XII House XII is composed of only two spaces for a total of fifty-two metres2 (Henrickson 1981, 52). Because the remaining walls are only fifty centimetres high, less than the walls of the other houses, the excavators thought that the dwelling was unroofed (Delougaz, Hill, and Lloyd 1967, 167). The entrance is from an alley perpendicular to Middle Road into space 24, a large area that could have been a multipurpose area. No doorway was found to access space 23. A doorway was reconstructed by the archaeologist. 4.2.3.8 House XIII On the southeastern side of houses XII and XI is House XIII, another small dwelling of fifty metres2 (Henrickson 1981, 52). It is composed of three spaces. The entrance is from the narrow alley flanking House XII into space 35, the largest room in the house. It is the only possibility for an entrance because the house appeared to have been fitted into the space left by the surrounding buildings (Delougaz, Hill, and Lloyd 1967, 168). The dwelling has an irregular plan. No other features were found in the excavation and the function of the rooms can be only suggested by the dimensions and position. Room 19, for example, could have been a storage space because it is narrow and small. The other two rooms could have been multipurpose spaces where the household activities took place. 4.2.3.9 House XIV House XIV is adjacent to House XIII on the northeastern side. It is another small dwelling of three rooms occupying forty-eight metres2 (Henrickson 1981, 52). It appears to have be fitted into the space left by the surrounding buildings (Delougaz, Hill, and Lloyd 1967, 168). The entrance to the building is at the end of the alley perpendicular to Middle Road through space 44, the largest of the house. This was probably the multipurpose space where the household activities took place. The other two rooms were smaller and at the back of the building. Their function was probably of storage or some private household activity. No other features were found in the excavations.

Chapter Four

210

House XIV Stratum Va 12

Surface area (m2)

10

8

6

surface (m2)

4

2

0 44

45

43

Space number

(a) House XV Stratum Va 7

6

Surface area (m2)

5

4 surface (m2) 3

2

1

0 27

28

30 Space number

(b)

29

Central Region Case Study: Khafajah and Tell Asmar

211

House XVI Stratum Va 16

14

Surface area (m2)

12

10 surface (m2) bin(s) oven(s)

8

6

4

2

0 22

27

21

Space number

(c) Fig. 4.50. (a) Internal space sizes and features of House XIV; (b) House XV; (c) House XVI

4.2.3.10 House XV The excavator hypothesises that House XV and House XVI were originally part of the same building—a bigger house existing at the previous stratum (Delougaz, Hill, and Lloyd 1967, 168). In stratum Va, House XV appears to be a building of four rooms occupying fifty-one metres2 (Henrickson 1981, 52). The entrance is from the alley to room 27, a relatively small and long room. At a different time there was a doorway in the wall of room 28 to access the house, and a pot was placed on the floor in room 28 (Delougaz, Hill, and Lloyd 1967, 168). There is no room substantially bigger than another, and rooms 27 and 28 both took the role of entrance, probably with cooking facilities. Rooms 29 and 30 are smaller and at the back of the house, and could have been storage spaces or rooms for more private household activities. 4.2.3.11 House XVI House XVI occupies forty-eight metres2 (Henrickson 1981, 52) and is divided into three rooms. Room 27 is the entrance to the building from

212

Chapter Four

Middle Road, and is equipped with a bread oven (Delougaz, Hill, and Lloyd 1967, 168). Room 21 is connected to room 27 and contains a bin of unbaked bricks. It may have been a storage space. Room 22 is the largest room, and probably a multipurpose space for household activities. In a later phase a doorway was opened in the wall of room 22 giving access to the alley flanking the house (Delougaz, Hill, and Lloyd 1967, 168). Hill suggested that the new arrangement was implemented because the previous situation—two small rooms in front and a large room at the back—was not satisfactory (Delougaz, Hill, and Lloyd 1967, 168). It was changed to a more typical layout with a large room at the entrance and small rooms at the back (Delougaz, Hill, and Lloyd 1967, 168). 4.2.3.12 House XVII The excavators suggest that perhaps houses XVII–XIX were built on the walls of a single earlier house (Delougaz, Hill, and Lloyd 1967, 168). House XVII was a small dwelling of sixty-four metres2 (Henrickson 1981, 52) organised into three rooms. Rooms J20 and K20 both opened along Middle Road and were connected with the largest room, 31. A V-shaped fireplace is present in room 31. It is formed of baked bricks set on edge on the floor, and contained traces of ash and burning (Delougaz, Hill, and Lloyd 1967, 154, 169). The northeastern section of the room is divided by a thin wall thirty metres high that the excavators interpreted as a low screen wall. There was a pot on the floor of room K20. Because the house has two doorways opening onto the main street and a peculiar layout, the archaeologists hypothesised that the building was used as a shop with a workroom or storeroom at the back. One of the front rooms could have been used as a private room by the owner or tenant of the commercial activity.

Central Region Case Study: Khafajah and Tell Asmar

213

House XVII Stratum Va 18

16

Surface area (m2)

14

12

10

surface (m2) jar heart(s)

8

6

4

2

0 31

J20

K20

Space number

(a) House XVIII Stratum Va 8

7

Surface (m2)

6

5

surface (m2) oven(s)

4

3

2

1

0 33

18

21 Space

(b)

36

214

Chapter Four

(c) Fig. 4.51. (aa) Internal spacee sizes and feattures of House X XVII; (b) Housse XVIII; (c) Housee XIX

4.2.3.13 Hou use XVIII House XVIIII and Housee XIX were partially desttroyed by a round r pit excavated inn stratum IVbb, but the rem mnant of the w walls was suffficient for the archaeologists to recconstruct a possible layouut of the grou und floor (Delougaz, H Hill, and Lloyyd 1967, 169). House XVIIII was probaably organised d into four sspaces extend ding over forty-three m metres2 (Henrrickson 1981, 52). The enntrance was lo ocated in room 21. T The doorway opened onto Middle Roaad. A bread oven o was found in thee corner of thhat room nearest to the enttrance (Delougaz, Hill, and Lloyd 1967, 169). Space S 33 was probably thhe main place for the household’ss activities. Thhe two rooms at the back o f the house co ould have been used ass storage spacces.

Central Region Case Study: Khafajah and Tell Asmar

(a)

(b)

(c)

(d)

215

Fig. 4.52. (a) Flowchart of House XVIII; (b) Flowchart of House XIX; (c) Flowchart of House XX; (d) Flowchart of House XXI

4.2.3.14 House XIX House XIX has been reconstructed as a dwelling composed of four rooms extending over seventy-five metres2 (Henrickson 1981, 52). A doorway from Middle Road was opened to space 22. The walls in space 23 were poorly preserved, but reconstruction was possible because of the presence of a solid wall in the stratum below (Delougaz, Hill, and Lloyd 1967, 169). The connecting doorway between room 26 and room 22 has also been

216

Chapter Four

reconstructed because no doorway was found in the badly preserved wall (Delougaz, Hill, and Lloyd 1967, 169). No features were preserved to help in the functional interpretation of the spaces. Moreover, the layout of the house is irregular and the spaces have similar dimensions. It is possible to suppose that room 22, being the entrance, had cooking facilities like the majority of other Tell Asmar dwellings; room 19, being far away from the entrance, could have been a storage space, and room 25 could have been the room for the main household activities. 4.2.3.15 House XX House XX is a large dwelling of 136 metres2 (Henrickson 1981, 52) positioned at the corner of Middle Road and East Road. The house is organised into nine spaces. The entrance is from East Road through room 35. The house is organised around a large space (27) surrounded by smaller rooms opening onto it. Space 27 had the characteristics of an internal courtyard, and a quadrantal structure was found in the southwestern corner that Hill interpreted as an altar (Delougaz, Hill, and Lloyd 1967, 169). Room 3 is the only room that did not open towards the main space (27), and was connected only with the entrance (35), perhaps being a storage area. Room 34 was equipped with a platform of unbaked bricks covered with a layer of gypsum plaster. A wide-mouthed pot was put on top of this. This structure was interpreted by the excavators as a washing place (Delougaz, Hill, and Lloyd 1967, 169). A pot containing silver ornaments was found buried near the doorway of room 2, which could have been used as a storage space (Delougaz, Hill, and Lloyd 1967, 169). Room 24, at the back of the house, was an unusual size and could have been the room were people gathered, ate, slept, and worked. There were no other features to help identify the use of space.

Central Region Case Study: Khafajah and Tell Asmar

217

House XX Stratum Va 25

Surface (m2)

20

15 surface (m2) jar altar 10

5

0 27

24

34

23

35

1

3

2

Space

(a) House XXI Stratum Va 18

16

Surface area (m2)

14

12

10 surface (m2) 8

6

4

2

0 40

28

27

39

41

Space number

(b) Fig. 4.53. (a) Internal space sizes and features of House XX; (b) House XXI

218

Chapter Four

4.2.3.16 House XXI House XXI was reconstructed by Hill because the walls at the southwestern corner were no longer in place (Delougaz, Hill, and Lloyd 1967, 169). It probably occupied eighty-four metres2 (Henrickson 1981, 52) distributed over five rooms. The entrance was found in an open passage just off East Road (Delougaz, Hill, and Lloyd 1967, 169). There were no other indications to interpret the use of space in the dwelling. Space 40 was probably a large room for the household activities, as was room 28 because of its dimensions, while room 39 was probably a secluded storage space. 4.2.3.17 House XXII House XXII is a large dwelling of 132 metres2 (Henrickson 1981, 52) organised with a large central space (37) probably surrounded by nine rooms. The eastern corner of the house was not preserved and was reconstructed based on the alignment of the walls and the remains of the previous stratum (Delougaz, Hill, and Lloyd 1967, 170).

(a)

Central Region Case Study: Khafajah and Tell Asmar

219

House XXII Stratum Va 16

14

Surface area (m2)

12

10 surface (m2) bin(s) jar

8

6

4

2

0 37

33

34

32

14

15

13

10

38

35

Space number

(b) Fig. 4.54. (a) Flowchart; (b) Internal space sizes and features of House XXII

The entrance to the house is through an open passage from East Road. Rooms 13, 14, and 15 are interconnected and almost the same size. In room 14 a low bin was found in the western corner. These rooms could have been equipped with cooking facilities, but none were found in the excavation. Room 33 was large and could have been a place for the household’s people to gather and for activities such as eating, working, and sleeping. Room 34 was relatively large and had a low screen wall in the southern area (Delougaz, Hill, and Lloyd 1967, 170). Room 34 could have also been used as a multipurpose space for household activities. Room 32 was smaller and connected with room 34. A pot was found on the floor of this space that could have been used as a storeroom. Rooms 38 and 35 were small and connected, and they could have been storage spaces too. The doorway between spaces 38 and 37 was reconstructed by Hill on the basis of the remains from the previous stratum (Delougaz, Hill, and Lloyd 1967, 170).

220

Chapter Four

4.2.3.18 House XXV houses XXIII and XXIV were not completely excavated; only the outer walls were traced, and the internal walls were reconstructed by the archaeologists (Delougaz, Hill, and Lloyd 1967, 170). House XXV, on the contrary, was fully excavated and appears to be one of the largest dwellings on the site (Delougaz, Hill, and Lloyd 1967, 170), occupying around 248 metres2 (Henrickson 1981, 52). Thirteen spaces composed this dwelling, organised with a central, large, rectangular space (1), surrounded by rooms that opened into it. An additional row of rooms opened towards South Road with two doorways connecting the street with room 9 and room 3. Several bins were found in room 9. Because of this organisation and the presence of another entrance to the house, room 9 has been identified as a shop by the archaeologist. The space was connected to the private owner (Delougaz, Hill, and Lloyd 1967, 170). The bins in space 9 were organised in two rows of square compartments of around forty centimetres wide and long. They were formed by thin, low partitions of unbaked clay and the inner row was about fifteen centimetres above the other. There was no indication of the contents of the bins, but Hill saw a strong similarity with the stalls of the modern bazaar, “where the commodities are displayed around the walls of the booth and the proprietor sits in the centre.”

(a)

Central Region Case Study: Khafajah and Tell Asmar

221

House XXV Stratum Va 18

16

Surface area (m2)

14

12

10

surface (m2) bins(s) oven(s)

8

6

4

2

0 1

4

9

12

11

17

7

10

8

5

6

3

2

Space number

(b) Fig. 4.55. (a) Flowchart; (b) Internal space sizes and features of House XXV

Room 3 has been identified as a private entrance and was equipped with a bread oven, as was often the case for Tell Asmar houses (Delougaz, Hill, and Lloyd 1967, 170). The doorway from room 3 to the street is smaller than the doorway connecting space 9 with the street. This has been seen by the archaeologist as a sign distinguishing the private use of space 3 from the public use of space 9 (Delougaz, Hill, and Lloyd 1967, 170). Room 7 connects the living quarters with the commercial activity (Delougaz, Hill, and Lloyd 1967, 170). In the western corner a plastered rectangle of unbaked bricks about a metre high has been found, the purpose of which is difficult to interpret (Delougaz, Hill, and Lloyd 1967, 170). House XXV in stratum Va had evidence of a later occupation. A bread oven and a low screen wall in room 17 belonged to the later occupation as well as the blockage of the doorway accessing room 11 (Delougaz, Hill, and Lloyd 1967, 170). Hill hypothesises that the presence of many ovens in the second occupation can be interpreted as a subdivision of the large house into many smaller dwellings in a later period, in line with the trend

Chapter Four

222

of decline observed on the site overall (Delougaz, Hill, and Lloyd 1967, 170).

4.3 Summary As in the case study in chapter three, the houses in this chapter have been analysed with regards to three main aspects: • • •

Space size Installations and other features within each space Access and privacy

Installations like ovens, tannours, and other cooking facilities obviously indicate cooking activities in the spaces in which they were found. In Khafajah and Tell Asmar, cooking installations were often found in the entrance room or in a nearby space (see, for example, House XXXII at Khafajah, House I of Stratum Vc, and House XVIII at Tell Asmar). Sometimes, hearths were found in central large rooms, and Henrickson saw this as an indication that the room must have been an unroofed courtyard where there would have been adequate ventilation to evacuate the smoke (Henrickson 1981, 62). Large bins are possible indicators that storage took place in a room (see, for example, House XXXII at Khafajah). In the ethnographic records, storage rooms were always present in houses. From the examples in chapter two, it appears that the minimum requirement for a dwelling was one room for people and one for storage. In many houses, one or more storage rooms were present (see, for example, the three store rooms in Sheikh Taifur’s House at Topzawa [Hansen 1961, 33]), often positioned in an area relatively less accessible in comparison to other rooms (see all the dwellings in Hansen’s ethnographic records [Hansen 1961, 33–42]). In the ethnographic evidence from the Near East there is no information on room size limitation, for example on how smaller sized rooms could be used or how many people could work/live in them. It seems impossible to completely separate storage rooms from living/working/sleeping rooms. A certain degree of storage was going on in any room (for example, in Watson’s ethnographic records almost every living room had niches for storage, wooden pegs, and grain pits [Watson 1979, 124–6]). On the other

Central Region Case Study: Khafajah and Tell Asmar

223

hand, some rooms could have been used for storage purposes only, and it is possible to try to identify them. Henrickson analysed the distribution of storage vessels and found that large, long-term storage bins were found in small rooms that were relatively less accessible from the outside in later phases, while in the earlier strata they were also found in larger, more-accessible rooms. She recognised a trend towards smaller, less accessible rooms, but the reason for this has not been found (Henrickson 1981, 64–6). Small, more portable storage vessels were found distributed throughout houses that were probably used in storage rooms, cooking areas, and working areas (Henrickson 1981, 66). Working tools like grinding stones and needles were found throughout the houses and not in specific working areas, which confirms the hypothesised fluidity of the use of space (Henrickson 1981, 66). Water-related facilities were usually drains and basins, and from their distribution Henrickson could not deduce a functional interpretation. Taking into consideration the limited number of toilets it appears that water-related facilities belonged to large buildings more often than small houses, a fact that suggests their relation to wealth and status (Henrickson 1981, 64). An isolated room opening directly to the outside suggests a commercial function or a different use that could involve people outside the household (see, for example, House XXXVIII at Khafajah). These guidelines have been used to formulate hypotheses on the use of space in houses at Khafajah and Tell Asmar. For the remaining spaces, a multiple use of space has been hypothesised on the basis of the archaeological observations. As can be seen from chapter two, in traditional houses in Iran, Iraq, and Turkey the spaces had very flexible uses, and there were no modern Western concepts like bedrooms or dining rooms. The sleeping and eating patterns were constantly changing, there was no privacy as intended in contemporary Western dwellings, and there was no nuclear family division when different generations were present in a house.

224

Chapter Four

The lifestyle described above produces a material culture very similar to the one we see evidence from in archaeological excavations on sites in the same geographical area that date to the third millennium BC. As a consequence, in this thesis it has been hypothesised that spaces without a specific function were probably used for household activities such as sleeping, eating, craft production and all the other activities that did not require specific features like ovens. Henrickson analyses the distribution of serving vessels and notices that in Tell Asmar they are more often found in rooms closer to the outside and in Khafajah more often in less accessible rooms (Henrickson 1981, 62). In general, she found that serving vessels were used throughout the house. Her interpretation of that evidence is that serving vessels were easily moved and different groups could perhaps have eaten in different rooms at different times, possibly based on sex and age (Henrickson 1981, 63). If we combine this information with the evidence from the ethnographic records analysed in chapter two and mentioned above, in the ethnographic analysis we find that the eating patterns were constantly changing and people arranged in different groups and in different rooms for each meal (Hansen 1961, 42). The method for understanding the use of space in the houses is the following: analysis of specific semi-fixed or fixed structures that indicate specific function (e.g. ovens, hearths, bins, and basins) and analysis of objects and other materials (e.g. ceramic and ashes). The next step is the analysis of the size and position of spaces—a small room farther from the entrance is more likely to be a storage space, while a large room is more likely a multipurpose living space, as explained before. This interpretation of spaces is perhaps more realistic than the interpretations that identified bedrooms and assigned areas to different nuclear families within an extended family. The bar chart in Fig. 4.5 shows the surface of each house analysed. The different sizes of houses in Asmar and Khafajah exhibit different quantitative and qualitative features. For Asmar, we consider data from all the analysed strata (all belonging to the Early Dynastic).

Central Region Case Study: Khafajah and Tell Asmar

225

First of all, we can observe that houses found in Khafajah are in general larger than those in Asmar. In fact, we can split houses into two bands: those smaller or equal to one hundred metres2 and those larger than one hundred metres2. We find that in Khafajah only around twenty-nine percent of the total number of houses fall into the first band and around seventy-one percent fall into the other band. For Asmar, around sixty-four percent fall into the first band and around thirty-six percent fall into the second. 45%

40%

Percentage of total number of houses

35%

30%

25% Asmar Khafajah

20% Asmar + Khafajah

15%

10%

5%

0% 0-50

51-100

101-150

151-200

201-250

251-300

301-350

351-600

Surface area bands (m2)

Fig. 4.56. Surface area distribution of houses in the central region

Furthermore, the average size in Khafajah is 196 metres2 whereas it is 102 metres2 in Asmar. We can now take into consideration the relative distribution of sizes by assigning houses to size groups and counting the percentage of houses in each group. We can observe from Fig. 4.55 that the most populated group for both Khafajah and Asmar is the one relative to 51–100 metres2. As already noted, sixty-four percent of Asmar’s houses fall in the 0–100 metres2 band, compared with twenty-nine percent for Khafajah. If we take into consideration other elements, like the number of rooms and the

Chapter Four

226

presence of a courtyard or a central room, it is possible to see that there are two distinct types of dwellings: “large” houses with more than 130 metres2 and an average of ten or more spaces, usually with a central space, and relatively “small” houses of less than ninety metres2 and an average of five or fewer rooms, and generally no courtyard or central room. Henrickson also notes the patterns and suggests that, because there are no houses smaller than forty metres2, that area must have been the minimum space acceptable for a family dwelling (Henrickson 1981, 54). Henrickson also suggests two different interpretations for the two types of houses; the first is that houses from 40 to 100 metres2 may have been inhabited by nuclear families, and houses between 130 and 140 metres2 may have been occupied by extended families (Henrickson 1981, 54). The second interpretation is that the difference in size was dependent on wealth (Henrickson 1981, 54). Henrickson took various elements into consideration as indicators of wealth: • • • •

the dimensions of the dwelling (a large house could have belonged to a wealthier family) the presence of luxury objects in the rooms or/and the burials associated with the house the presence of built tombs (for example vaulted tombs) in contrast with simple pit graves more often associated with small houses the presence of water-related facilities (Henrickson 1981, 69–74)

A comparison between the two sites highlights a difference: graves were almost absent in Tell Asmar houses, while at Khafajah many tombs were found under house floors (Henrickson 1981, 77). On the basis of the previously stated criteria, it is possible to see that, among the almost forty houses analysed from the sites of Khafajah and Tell Asmar, two groups are present: wealthy households, generally occupying large houses and owning luxury objects, and less-wealthy households, owning small houses often composed of three to four rooms. The interesting fact is that both groups were coexisting in the same areas, and very wealthy houses were often sharing walls with non-wealthy ones. In Khafajah and Tell Asmar, there does not seem to be a differentiation of neighbourhoods on the basis of wealth as often happens in modern contemporary cities.

CHAPTER FIVE SOUTH REGION CASE STUDY: TELL ABU SALABIKH AND SHURUPPAK

The methodology described in chapter two is applied to two Early Dynastic sites where several houses have been excavated in the south of Mesopotamia. The two cities have been chosen because they have the remains of domestic architecture, in particular examples of complete house plans, from the same period (the second half of the second millennium BC) in order to have comparable evidence. As has been said in chapter two, there are a number of limitations in studying houses for this period in Mesopotamia given the amount of evidence suitable for analysis. Some issues are: the uncertainty of the existence of an upper story, or the need for a complete ground floor plan to carry out the spatial, analysis which is a rare occurrence because of difficulties in preservation and sometimes bad excavation or little interest in domestic quarters by excavators. These problems reduce the amount of evidence available for study, often to very few units per site. This study concentrates on sites richer in data and well published. The sites of Tell Abu Salabikh and Shuruppak in southern Mesopotamia have been chosen following those criteria.

5.1 Tell Abu Salabikh Tell Abu Salabikh was a mid-third millennium BC city located in the alluvial plain of southern Iraq (Matthews et al. 1994, 171). The site is composed of several low mounds (see Fig. 5.1) and occupies an area of one kilometre by five hundred metres (Matthews et al. 1994, 171). The Sumerian name of the city is not known for certain, but could have been Eresh (Postgate and Moon 1976, 161). The site was excavated by the Oriental Institute of the University of Chicago for eight weeks between 1963 and 1965 (Hansen 1974, 5) and by Cambridge University under the direction of J. N. Postgate between 1975 and 1989 (Postgate 1983, 1).

228

Chapter Five

Fig. 55.1. Third-milleennium southerrn Mesopotamiaa (Postgate 1994)

Textual eviddence was fouund in the exccavations of buuildings on th he eastern mound in A Area E in the highest h part off the Tell slighhtly north of the t centre (Hansen 19774, 5). The teexts have beeen dated to thhe Early Dyn nastic IIIa (Hansen 19774, 17). The analysed tabllets gave infoormation abou ut: “royal stables, largge flocks of shheep and goats and the alllocation of ag gricultural land to city officials” (Maatthews et al. 1994, 171). A Attempts weree made to match archaaeological annd written tesstimonies of life in Abu Salabikh (Matthews eet al. 1994, 1771). Domesticc housing wass exposed on the main mound (Figg. 5.2). For thhe purpose of this study, onnly six buildiings have been considdered, i.e. the ones that sho ow an almost complete ground-floor layout. Thesse are located in Area A and d Area E (see Figs. 5.3 and 5.4). Not all of these buildings havve been identiified as housees, but their laayouts do look very sim milar to thosee of houses.

Soouth Region Caase Study: Tell Abu A Salabikh an and Shuruppak

Fig. 5.2. Site plan of Teell Abu Salabik kh (scale 1:10,0000) (Postgate 1983) 1

229

230

Chapter Five

Fig. 5.3. S Southeastern Quuarter of Abu Saalabikh Main M Mound (Postgatee 1990)

Unfortunateely, ground floors f were not optimallly preserved in Abu Salabikh. Inn some cases, it is not clearr where the m main entrance is i located or where thee exact locatioons of doors are. a This lackk of informatio on affects the understaanding of thee internal cirrculation, an essential eleement for carrying outt the permeabbility analysis.. It was possibble to draw fllowcharts for four Abuu Salabikh hoouses (see Fig gs. 5.5[b], 5.7[[a], 5.8[b], and d 5.9[b]). It has not beeen possible too draw flowch harts for the tw wo remaining buildings (the southernn and northernn buildings in n Area A) in thhe database.

Soouth Region Caase Study: Tell Abu A Salabikh an and Shuruppak

231

Fig. 5.4. A Area A of Abu Salabikh S Main Mound M (Mattheews and Postgatte 1987)

At present, tthere is not suufficient inforrmation to forrmulate hypottheses for the missing elements of the t ground flo oors, but one oof the outcom mes of this work couldd be the form mulation of such s hypotheeses or expreessing an opinion aboout the naturee of the fourr Abu Salabiikh buildings (i.e. the Southern Unnit and the Buurned Buildin ng in Area E, and the N Unit U and S Unit in Areaa A). As previiously stated, it i is not clear if these buildiings were houses.

5.1.1 Buildiing 5G65 i a large As can be sseen in Fig. 5.5(a), Housee 5G65 in Abbu Salabikh is building com mposed of thiirteen spaces with one entrrance from th he outside from space 113. The bar chaart in Fig. 5.5((c) shows the surface valuees of each spacce. Space 6 is the larggest with a surrface area of almost a forty-ffive metres2. Matthews M identified thhis as an innternal courty yard because of the diffficulty of acquiring suufficient sturddy timber lon nger than fouur metres to cover c the

232

Chapter Five

space. Matthews’s analysis supports this interpretation (Matthews et al. 1994, 201). Room 3 is around thirty-four metres2 and was identified as a reception room by the excavator. Rooms 2, 7, 13, and 10 were between fourteen and nineteen metres2. They could have been living rooms for eating, working, and sleeping. The other spaces were smaller, being less than ten metres2. Space 11 is a long and narrow corridor; spaces 1, 5, and 9 could have been storage rooms because they are relatively small and have a high depth value. No cooking facilities were found. The lack of features in this house makes it more difficult to interpret the use of space. Matthews analysed the floors of the building and found them all well prepared and of good quality, while in other buildings there was a difference in quality between, for example, the courtyard and other rooms (Matthews et al. 1994, 188). The courtyard in building 5G65 was “wellmaintained with laid plaster floors and bitumen pathways” (Matthews et al. 1994, 194). In room 7 she found abundant remains associated with food preparation (Matthews et al. 1994, 197). On the basis of her analysis of the succession of type of floor deposits, she gave an interpretation of the use of space. For example, spaces 3 and 1 were very clean with well-prepared floors, characteristics she finds consistent with their interpretation of space 3 as a “reception” room with a relative annex or perhaps a shrine, but she found the same degree of “cleanness” in rooms 2 and 8, although with slightly different remains. In room 8 it is associated more with food consumption, while in room 2 it is perhaps caused by frequently shaking mats used for sitting and sleeping on (Matthews et al. 1994, 196–7). She also found evidence of use of water with the baked brick foundations and bitumen unearthed in room 12. Room 9, on the other hand, changed function, having been a well-plastered and well-maintained room in an earlier phase, becoming a room with a bitumen floor that suggests use of water (Matthews et al. 1994, 197). In room 5 no food remains were found, but evidence of a mat that was not often moved was found. Matthews suggests that space 5 was used as a storage room (Matthews et al. 1994: 198–9). In room 10, deposits associated with cooking were found (Matthews et al. 1994, 201).

Soouth Region Caase Study: Tell Abu A Salabikh an and Shuruppak

(a)

233

(b) Building 5G65

50 45 40

Surface area (m2)

35 30 25

surface (m2)

20 15 10 5 0 6

3

2

7

13 3

10

8

5

11

1

12

9

4

Space numbe er

(c) Fig. 5.5. (a) P Plan of House 5G65 5 (Matthews et al. 1994); (b (b) Flowchart; (c) ( Internal space sizes s and featurres of House 5G G65

234

Chapter Five

5.1.2 Building 6H82 House 6H82 was an even more extended building composed of around fourteen spaces in phase 1B and thirteen spaces in phase 1C. There was one entrance to the outside in room 66 (see Fig. 5.6). Two other openings seem to be present in spaces 60 and 72. In the bar chart for phase 1B we can see that the building had one large space (65) with an area of almost forty-five metres2 that Matthews identified as an open internal courtyard for the same reasons as expressed above for House 5G65. Also in this case, Matthews’s analysis supports this interpretation (Matthews et al. 1994, 201). For phase 1C, the bar chart in Fig. 5.7(c) shows that space 65 was larger than in the previous phase, and in fact had a surface area of sixty-four metres2. From the plan, it seems that there was no separation among rooms 69/70/71. If this was the case, the room would have been eighty metres2. In house 6H82, three spaces with cooking facilities have been identified (61, 64, and a kitchen unit composed of rooms 67 and 69). Matthews analyses the composition of the floor of rooms 67, 68, and 69, which was unusual and highly water resistant. She notes that the floor is appropriate for a cooking area (Matthews et al. 1994, 188). Moreover, she notes anthropogenic deposits that she interpreted as the consequence of food preparation and storage on the floor in rooms 67 and 69 (Matthews et al. 1994, 190) along with deposits associated with food cooking (Matthews et al. 1994, 201). Room 68, on the other hand, has been identified as a possible area below a wooden staircase on the basis of the architectural layout and also includes a large drain or latrine. Matthews’s analysis suggests the presence of a microclimate supporting the interpretation of an area under the stairs and attestations of the post-depositional percolation of water (Matthews et al. 1994, 199). House 6H82 also has remains of a tannour (in room 64), an oven (in room 69), and several hearths (rooms 70+71, 62, and 66). Rooms 70+71 and 62 could have been multipurpose rooms for eating, sleeping, and working because of their dimensions and the presence of hearths. Moreover, Matthews’s analysis shows that the floor in room 70 together with room 63 was of better quality compared to the other floors of the building (Matthews et al. 1994, 188); for this reason, she interprets the rooms as constituting a “reception” room with a relative annex (Matthews et al. 1994, 197).

Soouth Region Caase Study: Tell Abu A Salabikh an and Shuruppak

235

The identificcation of the use of the oth her spaces is ddifficult due to t its lack of features. IIt is possible to t hypothesisee that one or m more rooms co ould have had the funcction of archivves because fraagments of tabblets were fou und in the filling of thhe graves. Thhese fragments appear to bbe concerned with the managemennt of land annd workmen for the cityy’s public in nstitutions (Matthews aand Postgate 1987, 100–1). This can connfirm that thee function of private hhouses was as a places of bureaucratic activities reelated the government, as in the case of the city of o Fara.

Fig. 5.6. (a) Plan of House 6H H82 (Matthews et al. 1994)

Chapter Five

236

(a)

(b) House 6H82

90 80

60 surface (m2) heart(s) grave(s) oven(s)

50 40 30 20 10

61

64

72

60

66

56

55

59

62 /6 3

67

58

65

/7 0/ 7

1

0

69

Surface area (m2)

70

Space number

(c) Fig. 5.7. (b) Flowchart; (c) Internal space sizes and features of House 6H82

South Region Case Study: Tell Abu Salabikh and Shuruppak

237

5.1.3 Southern Unit The Southern Unit in Area E is composed of eleven spaces with only one entrance connecting space 38 with the outside. A clay tablet was found in the floor probably belonging to an earlier level of the building (Hansen 1974, 5). The bar chart in Fig. 5.9(c) shows that space 39 is the biggest at thirty-four metres2. Matthews identified space 39 as a “reception room,” probably following the popular model which tries to identify a “reception room” as differentiated from a “living room” in houses. Space 39 has a high depth value and is among the farthest spaces from the entrance; it is necessary to cross five other spaces to reach it and it is possible to say that the largest rooms were probably the best suited to being multipurpose spaces for eating, working, and social activities. A clay bowl was found set into the floor of room 39 (Hansen 1974, 9). There are tree spaces of between twenty-five and thirty metres2, one of which has been identified by Matthews as a courtyard (number 41) with the same methods used for the other buildings. Also, spaces 44 and 40 could have been multipurpose rooms because of their relatively large dimensions. In the northwestern corner of room 44, several fragments of tablets have been found (Hansen 1974, 6). Spaces 30 and 46 could have been storage rooms because of their small dimensions, positions, and high depth values. There is a long corridor composed of spaces 29 and 43, and spaces 48 and 45 also have peculiar narrow and long shapes. It is difficult to interpret the function of those spaces especially in the absence of other features. Originally, a door connected corridor 29 with space 40, but at the analysed level the doorway was filled by baked plano-convex bricks (Hansen 1974, 6). The ovens and firepits shown in Fig. 5.8 have been attributed to later deposits of ash, and were in use after the abandonment of the building (Hansen 1974, 6).

238

Chapter Five

Fig. 5.8. Plan of Area E (Hansen 1974, Fig. 1)

Soouth Region Caase Study: Tell Abu A Salabikh an and Shuruppak

(a)

239

(b) Area E Southern Unit

40

35

Surface area (m2)

30

25

20

surface (m m2)

15

10

5

0 39

41

38

45

44

29 9

40

43

48

46

30

Space number

(c) Fig. 5.9. (a) Plan of Area E Southern Unit (Matthews et aal. 1994); (b) Fllowchart; (c) Internal spacee sizes and featu ures of Area E S Southern Unit

240

Chapter Five

Figg. 5.10. Entrancce to the Southeern Unit (Hanseen 1974, Fig. 2)

Fig. 5.11. P Partially excavatted bowl set intto the floor of rroom 39 in the Southern S Unit (Hansen 1974, 1 Fig. 7)

South Region Case Study: Tell Abu Salabikh and Shuruppak

Fig. 5.12. Room 29 in the Southern Unit (Hansen 1974, Fig. 3)

241

242

Chapter Five

5.1.4 Burned Building The Burned Building is composed of fourteen spaces with one entrance in room 7, although Hansen suggested that the main entrance was in the northeastern corner of the building and had not been reserved (Hansen 1974, 13). Space 9 is the biggest room at 39.5 metres2. As in the Southern Unit, this space was identified as a “reception” room by the archaeologist. It was a room with a high depth value, as, like room 39 in the Southern Unit, it is necessary to cross five spaces to reach it. As explained for the Southern Unit, it is only possible to say that the largest rooms were probably the best suited to being multipurpose spaces for eating, working, and social activities. Space 4 is a big square space of thirty-six metres2 that has been identified as an open courtyard by the archaeologist with the same method used for the others buildings. Hansen found bitumen runners on its southern and western sides (Hansen 1974, 13). Spaces 15 and 3 could have been other living rooms because of their large dimensions and high depth values. The doorway connecting the courtyard and room 3 was coated with bitumen, and the same material coated a low platform in the room. Hansen judged it as being too low to be an altar. A bench is also mentioned in the report but is not visible in the plan of the building (Hansen 1974, 13). In room 6 a drain was found along with a structure that Hansen interpreted as being a bath, while ten centimetres above the floor a backed tablet was found. Hansen also considered space 22 to be a room in which water was used because of the partially bitumen coated floor, but no drainage system was found in the room (Hansen 1974, 13). He also considered space 17 as a washing area because of finding a bitumen-coated depression in the floor (Hansen 1974, 13). Spaces 10 and 14 could have been storage rooms because of their small dimensions and positions in the flowchart. There was the presence of a long and narrow corridor as in other buildings of this site, but it is difficult to understand the use of many rooms because they were not well preserved (Hansen 1974, 13). Cooking facilities made of mud with the occasional addition of some mudbricks were identified in space 7, and a low bench was fitted into the northeastern corner of the room (Hansen 1974, 13).

Soouth Region Caase Study: Tell Abu A Salabikh an and Shuruppak

(a)

243

(b) Burned Bu uilding

45

40

Surface area (m2)

35

30 surface (m2) platform hearth(s) kiln bench drain

25

20

15

10

5

0 9

4

24

5

15

18

3

8

17/19

6

14

22

10

7

er Space numbe

Fig. 5.13. (aa) Plan of the Buurned Building (Matthews et aal. 1994); (b) Fllowchart; (cc) Internal space sizes and feattures of the Burrned Building

244

Chapter Five

A kiln was ffound in space 24 as well as a in space 255, which does not seem to be part off the building (Hansen 1974 4, 13). Hansen repoorted the pressence of a staiircase betweenn corridor 8 and a space 19 but did nnot provide eviidence of it (H Hansen 1974, 13).

Fig. 5.14. Courtyard (Rooom 4) with bitu umen runners inn the Burned Bu uilding (Hansen 1974 4, Fig. 13)

In summaryy, in all of the four building gs a main opeen space was identified i which, from m the flow chaarts, seems to be the centree of distributio on for the circulation. R. Matthewss seems to bee certain thatt these squareed spaces lacked a rooof “because of the difficulty of acquiiring sufficien nt sturdy timber for spanning a width w greater than 4 m oor so” (Matth hews and Postgate 19887, 115). These spaces had d dimensions llarger than fou ur metres both in lenngth and widdth (for exam mple, the couurtyard of thee Burned Building is 6 metres by 6 metres; the courtyard c of H House 5G65 iss 7 metres by 6.4 metres etc.). All the t other “larg ge” rooms, onn the other haand, were rectangular iin shape, the smallest s side being b no long er than four metres. m The archaeoologists noted that the variation in courtyyard area dimeensions in the analysedd building of Abu Salabikh h had a very cclose relationship with the variatioon of the ovverall house area. They hhypothesised that the constructionn of a large courtyard c imp plied the buillding of a larrge house

Soouth Region Caase Study: Tell Abu A Salabikh an and Shuruppak

245

because the number and size s of the surrrounding room ms were deterrmined by the courtyarrd size (Mattheews and Postg gate 1987, 1177).

Fig. 5.15. F Fireplaces in rooom 7 in the Bu urned Building ((Hansen 1974, Fig. F 11)

Matthews iddentifies one reception r room in all the aanalysed build dings and two in Houuse 6H82 (roooms n. 70+71 1 and 62). Unndoubtedly, there t is a recognisablee pattern for thhose spaces th hat have high depth values,, and they were all in ffact as far as possible p from the t main buildding entrance and were moreover thhe biggest spaaces, generallly as wide ass possible for a roofed room (Matthhews and Posstgate 1987, 117). The areaas of those spaaces were generally coonstant, and not related to the housee area (Matth hews and Postgate 19887, 118). A rooom of thirty or forty metrees2 seems to have h been a conditio siine qua non foor the Abu Sallabikh buildinngs. Given the sttandardisationn of these room ms, it seems ppossible to hy ypothesise that they were fulfilliing some so ort of sociaal function. Another characteristiic is the preseence of a long g narrow corriidor as in the Southern Unit, whichh Hansen com mpared to the one in Housee D in Khafaajah or in House XIII at Fara. Beccause Deloug gaz has suggeested that Ho ouse D at Khafajah m might have been b the resiidence of a person of high h rank connected w with temple administratio on, Hansen suggested something similar for thhe building inn Area E (Han nsen 1974, 18)).

246

Chapter Five

Fig. 5.16. Drain in room 6 in the Burned Building (Hansen 1974, Fig. 12)

South Region Case Study: Tell Abu Salabikh and Shuruppak

247

One cannot exclude the possibility that these buildings had a different function given the extension of the constructions and the lack of cooking facilities in some of them, although this might be due to bad conservation. The four buildings in the flow chart (houses 5G65 and 6H82, Area E Southern Unit, and the Burned Building) have a high variability in depth, while House 6H82 has many uncertain elements in its flow chart, such as many rooms without access and two entrances to the outside that are not certain. This analysis is not very meaningful for Abu Salabikh because the number of houses is too small to build a statistical survey, but the findings could be useful in future in comparison with other buildings from different sites. Matthews’s analysis gives more information on the buildings: for example, she found more bones in courtyards and streets than in roofed spaces (Matthews et al. 1994, 176). Her analysis also showed that pigs may have run free in the city streets and that “Unroofed space, street and ash-tip samples were richer in animal and fish bones than room samples” (Matthews et al. 1994, 176). Matthews noted that the location of fire installation suggests cooking (Matthews et al. 1994, 181). On the basis of her analysis, she identified in buildings G and H a pattern organised as follows: “a room for cooking and food preparation, and an adjoining ancillary room with indication of food storage and food preparation” (Matthews et al. 1994, 201). She noted that plastered floors in courtyards were poorly prepared, while plastered floors in “reception” rooms appeared to be well prepared (Matthews et al. 1994, 188). Matthews found the lowest percentage of over-floor deposits in all living and “reception” rooms, which suggests that floors were covered by rugs and rooms were kept clean. On the other hand, the highest percentage was present in courtyards and open spaces (Matthews et al. 1994, 191). No storage of grains, such as mudbrick bins or pottery jars, have been found in Abu Salabikh, and similar a situation has prevailed in the majority of third-millennium Mesopotamian sites, with perhaps the only exception being some brick-lined pits excavated at Fara (Martin 1988; Matthews et al. 1994, 194). Matthews’s thin-section analysis does not indicate a concentration of animals in the four Abu Salabikh buildings taken into consideration. The

248

Chapter Five

only place in the city that held evidence of intermittent stabling was located in H58 and was a large square area, which was most probably unroofed (Matthews et al. 1994, 196). No evidence for cottage industries was found within the large wealthy households, the only industrial area encountered being on the Main Mound pottery production area at the northern end of 4I-4J (Matthews et al. 1994, 204).

5.2 Shuruppak Shuruppak, on the site of modern Tell Fara in Iraq, was probably one of the largest and most important Sumerian cities. It was built on the banks of the River Euphrates on the Mesopotamian alluvial plain (Martin 1988, 10– 12, 129). Numerous small branches of the Euphrates flowed through what is today an arid desert (Martin 1988, 125). The climate in the south of modern Iraq is characterised by insufficient rainfall for agriculture without irrigation (Martin 1988, 12). Andreae described the area around Shuruppak, in Sumerian times, as being marshy with sufficient water to allow cultivation (Andreae 1903). The other means of substance at the time, other than agriculture, were fish, sheep, goats, and cattle (Martin 1988, 125). Today, the mound occupies 120 hectares, considering the high mound and the lower rise of one to two metres surrounding it (Martin 1988, 113). The site was excavated in 1902–3 by the Deutsche Orient-Gesellschaft, which excavated long criss-crossing trenches and thoroughly tested the mound (Martin 1988, 113). In 1931, the site was excavated by a mission of the University Museum of the University of Pennsylvania (Martin 1988, 10, 16–17). In 1973, Martin conducted a three-day survey on the site (Martin 1988, 113). The finds were only completely published in 1988 (Martin 1988). Houses from Fara were analysed by Starzmann in her Master’s thesis with the use of spatial analysis (Starzmann 2007).

Soouth Region Caase Study: Tell Abu A Salabikh an and Shuruppak

Fig. 55.17. Third-milllennium southerrn Mesopotamiia (Postgate 199 94)

249

250

Chapter Five

Fig. 5.18. Plan P of Fara (Staarzmann 2007, Karte 1)

South Region Case Study: Tell Abu Salabikh and Shuruppak

251

The objects found in the site have been dated from the Jemdat Nasr to the Old Babylonian (Isin-Larsa) periods (Martin 1988, 10). Martin hypothesised that Shuruppak was a village at the top of the mound during the Jemdet Nasr Period (Martin 1988, 114). The finds indicate that the site was probably already seventy hectares large by the end of the Early Dynastic I (Martin 1988, 113). The Early Dynastic IIIa is well represented. The stratum was destroyed by a large fire which, as consequence, preserved many objects like tablets, seals, and pots that were left in the buildings (Martin 1988, 115). There is evidence that the ED III city was extending further east than the previous levels, although the area has been eroded, and the site probably reached the size of one hundred hectares in that period (Martin 1988, 115–16). Martin estimated that this size could accommodate between fifteen to thirty thousand inhabitants (Martin 1988, 127). There was little evidence of the Akkadian and Ur III periods, probably because of the erosion. For the Ur III Period there was evidence of a “city wall” that was possibly not present in previous periods (Martin 1988, 116). After 2000 BC the city declined and was no longer mentioned in the texts (Martin 1988, 14). Martin hypothesised that there was a gradual shift of the River Euphrates towards the west that caused the city to be abandoned (Martin 1988, 14). More than nine hundred tablets were found at Fara. The vast majority dated to the Early Dynastic IIIa, although there were examples from all the occupied periods from Jemdet Nasr to Ur III (Martin 1988, 82). The tablets generally came from burned houses near the surface (Martin 1988, 85). Martin tried to associate the architectural remains with the location of tablets, but the buildings were very poorly defined (Martin 1988, 86). From the texts it appears that Shuruppak had a king because the title lugal is mentioned several times, but GAR-ensis probably governed the city, organised in groups for the short term (Martin 1988, 118–19). Those interpretations are at the level of hypothesis because palace or temple archives have not been found, only household archives (Martin 1988, 127). Martin reported that: “Fara has the largest number of ED II seals and sealings found in Mesopotamia” (Martin 1988, 76). Information on thirteen buildings found in Fara has been published by Heinrich in 1931 and Martin in 1988. The data have been judged good for

252

Chapter Five

quantity andd quality by Starzmann in n order to perrform spatial analysis, although theere is no clearr information about a the topoography of thee site, the canalisation of the water,, and the road ds (Starzmannn 2007, 30). Generally, G the remains belong to the latest phase of o buildings w which dates to the Early Dynastic IIIa (Starzmann 2007, 33). Objects werre found in Fara’s houses which havee been catalo ogued by Starzmann, who built a database, but b they weere not usefu ul in the reconstructioon of room function f (Starzzmann 2007, 53). Heinrich h divided the site withh a grid of onee hundred by one o hundred m metres, subdiv vided into squares of teen metres idenntified by letteers. Eight builldings were ex xposed in areas IIIa/IIIIa-b, VIak-al, north VIIp-q, VIIs-u, andd XVac-ad, and a south IXab-ac andd XIIIf-i, XVaa-d (Heinrich 1931). 1

5.2.1 Housee Vlak-al 2 House Vlakk-al is a dwellling of 140 metres m (Starzm mann 2007, 155). 1 It is composed of eight spacess and at least one o more room m, space 9, in which no limits were found on the northwestern n side. Excavattors did not ev ven find a wall on the w western corneer of space 6 so s they hypothhesised the presence of a second enntrance or a continuation c of o the buildinng on the wesstern side (Starzmann 2007, 30).

(a)

South Region Case Study: Tell Abu Salabikh and Shuruppak

253

(b) House VIak-al 20 18 16

Surface area (m2)

14 12 10

surface (m2)

8 6 4 2 0 2

6

9

3

1

8

5

4

7

Space number

(c) Fig. 5.19. (a) Plan of House VIak-al (Heinrich 1931); (b) Flowchart; (c) Internal space sizes and features of House VIak-al

Starzmann, who published a study in 2005, hypothesised that a wall closed space 9 on the basis of a Minglus categorisation of plans on houses that see House Vlak-al as a house with a central court surrounded by rooms opening towards an internal court. On the same basis, she identified room 9 as a reception room and room 6 as the Houptwohraum of the Minglus classification (Starzmann 2007, 168).

254

Chapter Five

In the passage between spaces 6 and 7, a vaulted beam was found. In space 8 some small installations of a structure that are supposed to contain water have been found. In the central space, in front of room 8, a little wall was positioned. This arrangement made Heinrich hypothesise that space 8 was used as a toilet or a bathroom (Heinrich 1931, 15). The only definite entrance to the building was on the southeastern side through room 1. The remains of walls on the northern side of the dwelling made archaeologists think of the existence of an adjacent building (Starzmann 2007, 168). Starzmann applied the spatial analysis to Fara houses and built a flowchart for House Vlak-al. The flowchart highlighted that space 2 had connections with five other rooms and was the most permeable space with the lower RA. Room 7 has been reasonably interpreted as a place for the storage of important items because of its position at the end of the chain of permeability. The hypothesis of space 2 as an open courtyard also seems to be correct. On the contrary, the Minglus classification, as explained in chapter two, seems too rigid. It appears that the use of space in third-millennium Mesopotamia was very flexible, and it therefore seems impossible to identify a room with a defined use as “reception.” Moreover, the reports about the objects found in the house (a clay tablet, a lapis lazuli seal, an alabaster vase, and other metal objects) do not give a room location for them and do not help in the interpretation of the use of space (Starzmann 2007, 198). We can only say that spaces 9 and 6 were the largest rooms and the best suited to being multipurpose spaces for eating, working, and social activities.

5.2.2 House Vlls-u In EF area at the centre of the Tell, near one of the highest points, excavations exposed two buildings sharing a wall.

Soouth Region Caase Study: Tell Abu A Salabikh an and Shuruppak

Fig. 5.20. House H VIIs-u (H Heinrich 1931, Fig. 7a)

(a)

255

Chapter Five

256

(b) House VIIs-u 35

30

Surface area (m2)

25

20 surface (m2) heart(s) 15

10

5

0 1

6

5

8

3

11

4

10

7

9

2

Space number

(c) Fig. 5.21. (a) Plan of House VIIs-ur (Starzmann 2007); (b) Flowchart; (c) Internal space sizes and features of House VIIs-ur

South Region Case Study: Tell Abu Salabikh and Shuruppak

257

The remains have been identified as two houses, both burned and both with foundations of baked brick and with walls up to 1.16 metres (Starzmann 2007, 171–2). Only four rooms remained of building 2 and part of the court (twelve by eight metres). Building 1 was a house of eighteen by thirteen metres composed of rooms surrounding an irregular court of 4.87 x 6.78 metres. Heinrich hypothesises an entrance in the northern corner of the building that has not been preserved leading into room 3, connecting rooms 2 and 4 with the court (Heinrich 1931, 13). Martin, on the other hand, hypothesises the presence of an entrance at the southern side of the building through room 10 as the main and only entrance or alternatively as a secondary entrance (Martin 1988, 96). Room 5 is a large room opening onto the court in the northern corner, and the floor is paved with clay brick and asphalt. Heinrich hypothesises that there was a passage from room 5 to the unpreserved area on the south of the building, but this is not certain (Heinrich 1931, 13). On the southern side there were the remains of at least three rooms. Martin considers room 5 to be a reception room (Martin 1988, 96). Room 6 is a large space opening towards the court, and is also connected with room 7, a narrow space like a sort of corridor that gives access to space 8. Martin considers rooms 7 and 8 to be storerooms (Martin 1988, 96). Room 9 on the western corner of the building is a small, square space and no openings were found to access to it. It is alleged to have been a storage room accessed seasonally and closed for part of the year, as was the case in some ethnographic examples (see chapter two). Room 4 is a small space connecting room 3 to the court. In room 4 were the remains of a floor and because of this it has been hypothesised that the space was a washroom. The court floor had baked bricks in some areas covered with asphalt. There was a brick base in situ at the lower part of the walls. A fireplace was present in the court along the eastern wall. Heinrich hypotheses that the hearth may have been in the open court or in a covered room that is no longer preserved (Heinrich 1931, 13).

258

Chapter Five

Fig. 5.222. Fireplace along the eastern n wall of the couurt in House VIIIs-u (Starzmann 20 007, Fig. 41)

The layout oof House VIIss-u has similarrities with the layout of Hou use VIakal, with twoo large rooms and many sm maller rooms aaround a courtt. For this reason, Starzzmann, follow wing the Ming glus classificaation of houses, assigns the functionn of reception room r to spacee 5, believing that, in Fara, reception rooms were situated on thhe southern sides of dwelliings (Starzmaann 2007, 174). As foor House VIakk-al, the iden ntification of a specific “reeception” room has been made onn the basis off the classificcation of the layout of houses withoout other featuures to confirm m it. It seems possible to suuggest that rooms r 5 andd 6 could haave been multipurposse rooms in which w many different houusehold activiities took place. On the souuthern side, Starzmann S reeconstructed rooms 11 an nd 10 as probable coonnecting rooms. She sugg gested that, iif there was a second entrance opeening onto rooom 10, the so outhern area ccould have beeen a shop or a separatee dwelling (Sttarzmann 2007 7, 175).

South Region Case Study: Tell Abu Salabikh and Shuruppak

259

Starzmann built a flowchart for House VIIs-u, and from that and comparisons with other houses she suggested that the main entrance was not from room 3 but from room 2, as suggested in her reconstruction (Starzmann 2007, 175–7). The flowchart shows that the court (space 1) was surrounded by five rooms with the same depth with little difference in RA. The flow chart confirmed the function of the distribution of the court, the most integrated space with the highest permeability. Room 8 was at the other end, the most segregated space, the farthest away from the outside if we do not consider room 9, where no connection was found with the rest of the house. Many objects were found in the building (clay tablets, seals, metal objects, stone and clay vessels), but the excavation reports did not give their location in the different rooms.

5.2.3 House XVa-d The remains of a building were found in area XVa-d, ten metres west of the Tablet House structures. A complete building was not found in the excavation and the plan exposed was hard to interpret as it may have belonged to more than one building. From the exposed plan, it is possible to observe a layout that can be recognised as a central court (space 1) surrounded by rooms, two of which are large and could have been multipurpose rooms (2 and 4); room 2 connected the court with the southern area of the building.

260

Chapter Five

Fig. 5.23. Excaavation of area XVa-d X (Heinricch 1931, 16)

In Fig. 5.244, Starzmann added her hypothesis h off the reconstrruction to Heinrich’s oone.

Soouth Region Caase Study: Tell Abu A Salabikh an and Shuruppak

(a)

261

(b) House XV Va-d

35

30

Surface area (m2)

25

20 surface (m2) basin 15

10

5

0 2

4

8

1

7

9

10

5

6

3

Space numberr

(c) Fig. 5.24. (a)) Plan of Housee XVa-d (Starzm mann 2007); (bb) Flowchart; (cc) Internal space sizes s and featurees of House XV Va-d

262

Chapter Five

Room 4 is connected to room 5, which I small and also connected to the court. Room 5 could have been a facility or a storage room because of its dimensions and position. Room 6 was described as a bathroom because of the presence of basin (sickerschacht) (Starzmann 2007, 179). The different archaeologists who studied House XVa-d thought that the group of rooms 8, 9, and 10 belonged to the same building as the previously described rooms. Room 2 on the southwestern corner was open towards court 1 and also on the other side. It is not clear if on the other side the opening was towards the outside. It is possible that it was opening onto an internal courtyard with a well, which is the space present on the eastern side of the excavated area. Room 2 is also connected with the small space 3 that appears as a storage room because of its position, small dimensions, and high RA (Starzmann 2007, 181). Room 10 was defined as a “vestibule” and leads to spaces 9 and 8. Space 9 has been identifying as a room with a placement for a doorman because the “vestibule” was not near the court but leading to other transitional rooms (Starzmann 2008, 181): there are no elements to back up this interpretation other than the German classification of house layouts. At other sites, for example, cooking facilities have often been found in spaces near the entrances, but no features have been recorded in the excavation reports in this case and it is not possible to know if facilities were present or if there were features that are no longer preserved. Starzmann calculated the RA of each space in House XVa-d, but given the uncertainty of some elements of the plan, such as the presence of a secondary entrance trough room 2, it is not possible to calculate the permeability value of each room in the building exactly.

5.2.4 House XVad Along the same trench on the eastern side another cluster of rooms was found. The area was occupied by several buildings flanking a street following a north-south direction (Starzmann 2007, 183). Archaeologists supposed that the partial remains of four buildings had been uncovered (Starzmann 2007, 183). For the best preserved two, an attempt at reconstruction has been made. They are located on the northwestern and eastern sides of the street.

Soouth Region Caase Study: Tell Abu A Salabikh an and Shuruppak

263

Fig. 5.25. Excaavation of area XVad (Heinricch 1931, 16)

5..2.4.1 House XVad X West The buildingg on the northhwestern side of the excavaated area had the t aspect of a house with a square central cou urtyard. The pplan of the ho ouse was found incom mplete and hass been reconsttructed by diffferent archaeo ologists in a building occupying 110 metres2 (see Fig. 5.26).

264

Chapter Five

(a)

(b)

South Region Case Study: Tell Abu Salabikh and Shuruppak

265

House XVad west 35

30

Surface area (m2)

25

20 surface (m2) 15

10

5

0 10

2

5

6

1

3

11

4

8

9

7

Space number

(c) Fig. 5.26. (a) Plan of House XVad west (Starzmann 2007); (b) Flowchart; (c) Internal space sizes and features of House XVad west

The entrance to the building is on the west, through space 1, which is described as an anteroom (Starzmann 2007, 183). The room is spacious and could have had other functions, such as cooking, as was the case for other houses from the same period (see chapters three and four), but features to help in the assessment of the use of the room are not preserved. Room 1 had a doorway opening towards the northern part of the building leading to rooms 3 and 4. Because rooms 3 and 4 could be reached from the outside without passing by the court, Heinrich hypotheses that they could have been shops (Heinrich 1931, 15). On the southern side, room 1 opened towards a large space that Heinrich supposed might have been stables (Heinrich 1931, 15). On the southeastern corner of the room a small wall of protection has been found, the structure making the archaeologist think that a toilet could have been set behind it, but no remains of water installation have been found (Starzmann 2007, 184). Starzmann considered space 2 as a possible shop or perhaps a dwelling for a guard or doorman (Starzmann 2007, 186).

266

Chapter Five

Heinrich considers room 10 on the western side of the building to be the main “residential room.” He reconstructed room 10 as a large rectangular space connected with the small secluded room 11 (Heinrich 1931, 15). The reconstruction of room 10 makes it look like a multipurpose room with a small annexed storage space (room 11), but room 2 could also have been another large multipurpose room if not stables. Starzmann considered space 10 to be the “main” room and space 6 the “reception” room following Heinrich’s interpretation of the use of space, according to the model that assigns these functions to two and only two rooms in a house. As has been said, there is insufficient information to make more than a hypothesis on the use of space in these rooms. On the northern side of what was probably an open internal court, there was a row of small spaces: rooms 7, 8, and 9. Archaeologists consider them to be storage spaces (Starzmann 2007, 184). No doorway was found for room 9, only foundations that could have belonged to a small entry point (Heinrich 1931, 16). Starzmann calculated the RA of each room for House XVad west, and in this case the house plan is also not complete, but it is not probable that the building had a second entrance. Consequently, the RA values could be reliable. The courtyard was the most accessible space and room 11 the least (Starzmann 2007, 186–7). 5.2.4.2 House XVad East The house analysed on the western side was probably 120 metres2 (see Fig. 5.27). The entrance is from the street on the opposite side of House XVad west. The dividing street is two metres wide. The main doorway lead to a relatively large space (room 1) has been dismissed as an antechamber by the archaeologists (Starzmann 2007, 187), but, because of its dimensions and position, it could have been a place for cooking facilities as in other houses from the same period, or a place for other household activities. Unfortunately, no semi-fixed features were found anywhere in the dwelling.

Soouth Region Caase Study: Tell Abu A Salabikh an and Shuruppak

(a)

(b)

267

Chapter Five

268

House XVad east 35

30

Surface area (m2)

25

20 surface (m2) 15

10

5

0 7

5

1

6

8

9

4

2

3

Space number

(c) Fig. 5.27. (a) Plan of House XVad east (Starzmann 2007); (b) Flowchart; (c) Internal space sizes and features of House XVad east

Space 1 was connected with two small rooms, 2 and 3, that could have been storage places because of their tiny dimensions and position away from the main circulation route. Room 1 also had a doorway on the northeastern corner leading into room 4, a narrow space that seems to have had the function of a passage to space 5, which was probably an internal court (Starzmann 2007, 188). Room 5 had two other preserved doorways, one leading into space 6, a relatively large room that could have been a multipurpose space for household activities and that Heinrich defined “main” room (Heinrich 1931, 16). Another doorway opened onto an area on the southeast of the building that has not been exposed by the excavations. Heinrich reconstructed, for analogy with other Fara houses, a large room 7 because his model needed a “reception” room and reconstructed two small rooms at the end of space 5 (Heinrich 1931, 16).

South Region Case Study: Tell Abu Salabikh and Shuruppak

269

Heinrich’s reconstruction is one among several possible solutions and has been adopted in this thesis to build the flowchart and the bar chart of House XVad east (see Fig. 5.27). Also, Starzmann built a flowchart for House XVad east and calculated the RA. In her opinion, the flowchart suggests a similar layout of the building where the majority of rooms are directly connected with the court. She calculated that House XVad east could have had a minimum of nine rooms (Starzmann 2007, 191). Applying the spatial analysis, Starzmann was able to make some general considerations on Fara houses. She observed that the flowcharts show that the circulation in all the houses had three shapes, and in two cases rings were present (houses VIak-al and VIIs). As explained in chapter two, there is a ring in the flowcharts when one space is accessible in more than one way. Starzmann also noted that the spaces identified as courts never opened directly to the outside and were the centre of the distribution of the circulation in the houses (Starzmann 2007, 191). Starzmann noted that, generally, Fara rooms were well integrated and only rooms near the entrance had a high accessibility (Starzmann 2007, 192). Following the spatial syntax theory that considers the degree of permeability connected to the function of a room, Starzmann considers less-integrated rooms as possible shops, as mentioned in the spatial analysis of House XVad west, and not very integrated rooms also had low degrees of permeability and RA as storage rooms (Starzmann 2008, 193). From the analysis of the bar charts (see Fig. 5.28) it appears that the five Fara houses taken into consideration had a large surface area between almost 87 and 234 metres2, and maybe more, because the smallest House (XVad east) presented an incomplete ground floor and might have been larger. The five Fara houses studied were composed of many spaces (probably between nine and eleven). It has already been mentioned that the archaeologists hypothesised the presence of an open internal courtyard giving access to many rooms around it for all the five houses. Starzmann excluded the possibility of a second storey because the wall rarely reached one metre of wideness that she judged insufficient to support a second floor, and moreover no remains of staircases were found (Starzmann 2007, 165).

Chapter Five

270

Shuruppak/Fara 250

200

150

surface area (m2)

100

50

0 VIIs-u

Viak-al

Xva-d

Xvad west

Xvad east

Fig. 5.28. Surface area of houses in Fara

Another characteristic that seems to be common in Fara’s houses is the presence of two large rectangular rooms perpendicularly arranged in 4 houses out of 5 (House VIak-al is an exception although there are at least two large rectangular rooms but arranged parallel). Those are the rooms that the archaeologists previously analysing Fara had identified as “main” and “gathering” rooms. Although the attribution of such specific functions is questionable, because there is not enough evidence to make the distinction, it is evident that there is a common pattern to Fara’s house plans. This peculiar layout is comparable to that of two houses in Titris Höyük (see chapter three, section 3, Building Units II and IV), although the two large rectangular rooms perpendicularly arranged are, in the Titris Höyük cases, specifically located in the corner opposite the entrance of the houses. A general lack of facilities and semi-fixed features gives the impression that they have not been preserved. This might have occurred because of bad excavating techniques (the investigation was started at the beginning of the 1900s) or peculiar environmental conditions not apt to preservation.

South Region Case Study: Tell Abu Salabikh and Shuruppak

271

Obviously, there were not micro-artefact data or micro-debris data such as that collected by Rainville and Matthews for other sites, because at the time of Fara’s excavation those techniques of analysis had not been developed. To estimate household size, it is necessary to rely on the hypotheses analysed in chapter two, section 4.7. As explained there, different archaeologists propose different figures (Watson 1979; Stone 1981; Postgate 1994; Schloen 2001) between four and ten metres2 of roofed area per person, none of them based on solid data. If we consider an average of seven to eight metres2, we have figures of around eight to more than twenty people per household. Following this estimation, the majority of Fara’s houses had an average of ten to fourteen people, and allows us to think of extended households and not nuclear families. It has already been explained that there is not clear information about the topography of the site, the canalisation of the water, and the roads (Starzmann 2007, 30). The little information available comes from observing the plan of the small portion of neighbourhoods exposed in the excavations. In the area exposed it is difficult to see signs of urban planning such as wall alignments or regular measures of house spaces. The only sign of organisation is the straight street with the doorways of the houses opening onto it in area XVad. Having only few houses excavated it is also impossible to understand any intra-site differentiation due, for example, to status. The five analysed houses seem quite homogeneous, and none of them seems to belong to poor households because they occupy extensive areas and were articulated with many rooms.

CHAPTER SIX DISCUSSION

In this chapter, the analyses carried out region by region are brought together. The chapter is organised in four sections, which are named after the four research questions stated at the beginning of the thesis.

6.1 How Large were Mesopotamian Residential Groups? To estimate household size it is necessary to rely on the hypothesis analysed in chapter two, section 4.7. Different archaeologists proposed different figures, none of them based on solid data. The summary of the estimations analysed in chapter two is: • • • • • •

Kramer estimates ten metres2 of roofed space per person based on ethnoarchaeological data (Kramer 1982b) Naroll estimates ten metres2 per person based on cross-cultural data (Naroll 1962) Watson estimates 7.3 metres2 of roofed space per person based on ethnoarchaeological data (Watson 1979) Stone estimates five to six metres2 per person based on textual and archaeological data from Old Babylonian Nippur (Stone 1981) Postgate estimates between four and seven metres2 per person based on data from Abu Salabikh (Postgate 1994) Schloen estimates of eight to ten metres2 per person based on demographic Mediterranean data (Schloen 2001)

Taking into consideration the above studies and the fact that the larger figures are often estimations of roofed space, for this analysis an average of between six and ten metres2 will be used, because the house area figures utilised also include the likelihood of an unroofed internal courtyard.

Chapter Six

274

500.0

400.0

300.0

200.0

100.0

0.0

surface area (m2)

XLVI (K) Building 6H82 IV (Ti) Building 5G65 Burned XLV (K) Yr (Ta) XXXII (K) Vx (Ta) 1 (Ti) XL (K) Southern Unit XXV Va (Asm) VIIs-u (F) II Vb (Asm) B1 (M) II Vc (Asm) Tcc (Ta) Qaa (Ta) B7 (M) XXXIV (K) L (K) I (Ti) I Vc (Asm) II (Ti) QbbII (Ta) I Vb (Asm) I Va (Asm) Viak-al (F) B4 (M) XX Va (Asm) XXII Va (Asm) Ucc (Ta) G1 (M) Py (Ta) QbbI (Ta) Xva-d (F) LI (K) XXXIII (K) XLVIII (K) Xvad west (F) XLVII (K) Xvad east (F) XLIX (K) VI Vb (Asm) XXI Va (Asm) VI Va (Asm) B2 (M) XIX Va (Asm) V Vb (Asm) III Vc (Asm) XVII Va (Asm) XI Va (Asm) Aar (Ta) LIII (K) X Va (Asm) XII Va (Asm) XV Va (Asm) XIII Va (Asm) V Va (Asm) XVI Va (Asm) XIV Va (Asm) VIII Va (Asm) XVIII Va B6 (M) XXXXIX (K) C2 (M) G2 (M)

Fig. 6.1. Surface area of the sixty-eight buildings analysed

Discussion

275

Fig. 6.1 shows the surface areas (internal floor areas) of the sixty-eight buildings analysed in this project. The range varies from the twenty-five metres2 of the tiny House G2 at Tell Melebiya to the 528 metres2 of Building XLVI at Khafajah. It is possible to group houses by the surface area of the ground floor. To the first group belong buildings of around eighty metres2 or less. These buildings usually have between two and six spaces and no internal courtyard, with only very few exceptions. To the second group belong buildings between one hundred and two hundred metres2. They generally have the appearance of a large house with an internal courtyard surrounded by rooms. Nine buildings out of the sixtyeight analysed were larger than two hundred metres2, two of them larger than five hundred metres2. For the first group of houses, the estimation of the households’ inhabitants is between three and four people in the smaller houses and between nine and fifteen in the largest (eighty-nine metres2). Fifteen people living in an eighty-nine metres2 household seems too many, so perhaps the estimation of six metres2 per person is too low. The number of nine people per house in houses up to eighty-nine metres2 seems to indicate that nuclear families, composed for example of parents and their children, inhabited the smaller houses. For the second group of houses the estimation of the households’ inhabitants is between eleven and eighteen people for a house of 108 metres2 (such as House XLVIII at Khafajah), and between twenty and thirty-four for a house of 203 metres2 (such as House B1 at Tell Melebiya). Again, the minimum estimation of people per metre seems reasonable if we imagine a large extended family with servants as part of the household. The maximum estimation of people per metre appears too large, and that of thirty-four people seems to be too many, even in a large dwelling of 203 metres2. To the third group belong buildings of 232 metres2 up to 528 metres2, although it is difficult to say if it is a real homogeneous group that shares the same characteristics or a list of exceptions. The estimation of people per metres2 is between twenty-three and thirty-nine for House II Vb at Tell Asmar (232 metres2) and between fifty-three and eighty-eight for building XLVI at Khafajah (525 metres2). Eighty-eight people seems a very large number even for a very wealthy household with many servants. Archaeologists often suggest that large buildings with evidence of domestic use could also have been utilised at the same time for commercial activities,

Chapter Six

276

Number of people 100

90

80

70

60

50

40

30

20

10

0

Number of people per house

House

XLVI (K) Building 6H82 (A.Sa) IV (Ti) Building 5G65 (A.Sa) Burned Building (A.Sa) XLV (K) Yr (Ta) XXXII (K) Vx (Ta) 1 (Ti) XL (K) Southern Unit (A.Sa) XXV Va (Asm) VIIs-u (F) II Vb (Asm) B1 (M) II Vc (Asm) Tcc (Ta) Qaa (Ta) B7 (M) XXXIV (K) L (K) I (Ti) I Vc (Asm) II (Ti) QbbII (Ta) I Vb (Asm) I Va (Asm) Viak-al (F) B4 (M) XX Va (Asm) XXII Va (Asm) Ucc (Ta) G1 (M) Py (Ta) QbbI (Ta) Xva-d (F) LI (K) Xvad east (F) XXXIII (K) Xvad west (F) XLVIII (K) XLVII (K) XLIX (K) VI Vb (Asm) XXI Va (Asm) VI Va (Asm) B2 (M) XIX Va (Asm) V Vb (Asm) III Vc (Asm) XVII Va (Asm) XI Va (Asm) Aar (Ta) LIII (K) X Va (Asm) XII Va (Asm) XV Va (Asm) XIII Va (Asm) V Va (Asm) XVI Va (Asm) XIV Va (Asm) VIII Va (Asm) XVIII Va (Asm) B6 (M) XXXXIX (K) C2 (M) G2 (M)

Max. num. people/house

Min. num. people/house

Fig. 6.2. Estimation of the number of people per house

Discussion

277

productive activities, or administrative activities. In the following paragraphs, hypotheses on the use of space in buildings will be presented. As a general comment, the figure of six metres2 per person produces houses that seem too crowded, but it is impossible to verify with certainty with the information in our possession at the moment. Perhaps only future archaeological discoveries of texts with detailed information about household composition could help in that direction.

6.2 How was Space used in Mesopotamian Houses? 6.2.1 Open Courtyard To understand the use of space in Mesopotamian houses, it would be useful to know with certainty which spaces were covered by a roof. Excavation reports are not exhaustive about this topic. For the sixty-eight buildings analysed in this thesis, hypotheses have been made gathering all information available. Lebeau considers the buildings in Tell Melebiya to be without any open internal courtyards (see chapter three, section 1, and Lebeau 1993). Lebeau himself was unsure for a long time about the issue, as different excavation reports give different interpretations, and in the final publication he decided on the absence of an open internal courtyard, even though the layout of the largest buildings appears very similar to buildings with open courtyards (central spaces surrounded by smaller rooms opening towards them). For Titris Höyük, Algaze hypothesises that there was at least one open central spaces in each of the four analysed buildings (see chapter three, section 2, and Algaze et al. 2001), and this can also be suggested on the basis of the layout that shows a large central space surrounded by smaller rooms. Moreover, the four analysed buildings in Titris Höyük are all larger than 150 metres2 and have large central spaces (between twenty and forty metres2) which would have been difficult to cover with a roof. Tell Taya was, among the sites in this thesis, the one with less available information. The site has not been excavated and only the layout of the buildings on the surface of the tell has been recorded. As a consequence, based on this hypotheses about the presence of an open internal courtyard have been made in this thesis (see chapter three, section 3).

278

Chapter Six

Delougaz, who directed the excavation at Khafajah and Tell Asmar, was of the opinion that the houses on the sites possessed no internal courtyards. As proof of his theory he pointed to a pier later added to the main room in stratum Vb of the “Arched House.” The pier should have sustained a roof over the central space, but Henrickson, who studied the material later, identified several internal courtyards in houses on the site (see chapter four and Delougaz, Hill, and Lloyd 1967). The position of many central rooms in the layout of houses was the exact position where an open courtyard could be expected to be. Abu Salabikh is the site among the seven analysed in this work in which the archaeologists produced the most scientific proof to support the identification of open internal courtyards. Matthews considered that the squared central spaces present in Abu Salabikh buildings lacked a roof, “because of the difficulty of acquiring sufficient sturdy timber for spanning widths greater than 4 m or so” (Matthews and Postgate 1987, 115). Matthews’s micro-debris analysis gave more supporting information, and she found more bone remains in courtyards and streets than in roofed spaces (see chapter five, section 1). The several archaeologists who excavated and studied Fara’s houses agreed on the identification of open internal courtyards in all five houses analysed in this thesis (see chapter five, section 2 and Starzmann 2007). Taking into consideration all the above information, hypotheses have been formulated about the presence or absence of an internal courtyard in the sixty-eight houses analysed in this thesis. The result is shown in Fig. 6.3. The graph shows that the vast majority of the buildings of more than one hundred metres2 have an internal courtyard (yellow bar), while the majority of the buildings of less than one hundred metres2 do not have an internal courtyard (light blue bar). The bars corresponding to the buildings with not enough information to determine whether there was an internal open space or not were left grey. A characteristic that seems common to open internal courtyards is that, in the majority of cases, the internal courtyard was the largest space in the dwelling. Out of thirty-three buildings with an internal courtyard, in twenty-five cases the courtyard was the largest space in the ground floor layout. Among the eight remaining buildings, four belong to Tell Melebiya where Lebeau hypothesised an absence of an internal courtyard, and the fact that the central space was not the largest might support the thesis that

Discussion

Fig. 6.3. The presence p and ab bsence of open courtyards

279

280

Chapter Six

it was not an open space. The other four examples belong to the south region (two in Abu Salabikh and two in Fara). In the south region, an internal square courtyard seems to have been common, surrounded by rooms among which was the presence of at least one or two long rectangular rooms sometimes larger than the courtyard itself. A hypothesis for this peculiar layout can be obtained from social factors. For example, Brusasco considered that solidarity among different groups of people or families is expressed in the reproduction of the same plan (Brusasco 1999–2000, 10). If talking about social solidarity may seem too much of a leap, it can be safe to affirm that the similar and formalised layout in houses is an indicator of high social interaction (Matthews and Postgate 1987, 118). In this analysis, buildings from Tell Taya have not been taken into consideration because of the uncertainty of the evidence coming from surface remains exclusively, as no buildings have been excavated at Tell Taya. Fixed features present in open courtyards could help the interpretation of the use of space, but very few have been found in them, and open courtyards do not seem to have many features. The more common feature is a bench along one or more walls which was present in only three open courtyards out of thirty-three, if Tell Melebiya buildings are excluded. At Tell Melebiya, in fact, four central spaces were equipped with benches and the fact can confirm Lebeau’s interpretation of them as central roofed rooms (see chapter three and Lebeau 1993). In House VIIs-u at Fara a hearth was found, but the excavator was not sure if the hearth belonged to the open courtyard or if it had a protective wall (see chapter five and Starzmann 2007). In the courtyard of House XX in the Va stratum at Tell Asmar an altar was found (see chapter four and Delougaz, Hill, and Lloyd 1967). Other features have been not found among the thirty-three buildings with a hypothesised or confirmed open internal courtyard. The absence of features in the excavation reports is not an assurance of the absence of features at all, as structures may have existed that were not preserved. But the presence of semi-fixed features in open courtyards appears to be so small that it seems safe to hypothesise that open courtyards in third-millennium Mesopotamia were generally empty of features. Similar activities to those of multipurpose rooms were probably going on in open courtyards, varying according to season and changing throughout the twenty-four hours of the day, such as: social activities, the

Discussion

281

preparation of food, craft activities, and possibly also sleeping during the hot season.

6.2.2 Features in Rooms Sometimes, fixed features have been found in the excavations of private houses. As it is possible to see from Fig. 6.4, the percentage of features found in the analysed sites is low. As has already been said, the condition of excavation and preservation may vary a lot, and in fact on some sites many fixed features were found and in others hardly any were found. hearths

ovens/ graves platforms kilns tannours 2.00% 5.01% 4.41% 0.40% 0.20%

benches drains bins jars altars 9.02% 1.40% 1.20% 2.40% 1.00% cooking basins silos/pits pots 0.20% 4.61% 0.20%

Fig. 6.4. Percentage of features found in 467 analysed spaces

The ratio between the size of a space and the surface of its house versus the total number of features in a given space produced the plot in Fig. 6.5. All the features analysed were plotted together and the trend emerging from the plot is that larger quantities of features are more likely to be found in small spaces. To further analyse the relation between space and features in order to understand the use of space, the plans of houses where features have been found have been grouped. Figs. 6.6, 6.7, and 6. show the plan of analysed buildings where domestic ovens or hearths have been found. A clear pattern indicates that, when domestic ovens are found, they are often in small rooms not far from the entrance of a house, perhaps in order to easily evacuate the cooking smoke through the door. In the case of smaller dwellings with only three or four rooms, domestic ovens were found in the first room—even if it was the largest room of the building—but they were still near the entrance.

Chapter Six

4 3 0%

5%

10%

15%

20%

25%

30%

35%

0

1

2

Total number of features found per space

Density of features in rela on to space size

5

6

ra o space(m2)/house(m2)

7

282

Percentage of space area (m2) /house area (m2)

Fig. 6.5. Number of objects for a given space on the abscissa and the percentage of the space relative to the total house area on the ordinate

Discussion

283

Fig. 6.6a. Locations of heaarths and domesstic ovens from Matthews and Postgate (1987), Algazze et al. (2001),, Delougaz, Hill, and Lloyd (19967), Lebeau (1 1993), and n (2007) Starzmann

284

Chapterr Six

Fig. 6.6b. L Locations of heearths and domeestic ovens, from m Algaze et al. (2001), Delougaz, Hill, and Lloyd (1 1967), and Lebeeau (1993)

Discussion

Fig. 6.7aa. Locations of hearths and dom mestic ovens, ffrom Delougaz, Hill, and Lloyd L (1967), an nd Lebeau (19993)

285

286

Chapterr Six

Fig. 6.7bb. Locations off hearths and domestic ovens, ffrom Delougaz, Hill, and Lloyd L (1967) an nd Lebeau (199 3)

Discussion

287

Fig. 6.88. Locations of hearths and dom mestic ovens, fr from Lebeau (19 993)

Where fixedd features werre found, it was possible too make hypoth heses: for example, doomestic ovens were the obv vious indicatorrs of cooking activities in the room ms where theyy were found, and if the rooom was very y small it could not acccommodate many other activities—per a rhaps food prreparation and some ffood storage. It seems safe fe to affirm thhat these spaaces were mainly usedd for cooking. In specific ccases, other types t of analy ysis could havve been carried out to further the iinvestigation: for two build dings, micro-ddebris analysis carried out by Mattthews enableed her to iden ntify a patterrn of a smalll cooking facility room m in Abu Saalabikh adjoin ned by an anncillary room with the remains of ffood storage and a food prepaaration (Matthhews et al. 199 94, 201). Algaze, annalysing Titriis Höyük, considers c thee evidence of food preparation, such as the presence of hearths, blacckened plasteered floor patches, coooking pots, andd grinding sto ones (Algaze eet al. 2001). Henrickson analyses thee distribution n in Khafajahh and Tell Asmar A of working toools such as grinding stonees and needle s, and notes that they were found tthroughout thhe houses and not in specificc working areas, which indicates thee fluidity of thhe use of spacee (Henricksonn 1981, 66). Unfortunateely, this kind of o information n is available ffor only a few w sites. Archaeologiists who excaavated Titris Höyük H hypothhesise that larg ge houses accommodaated extended families and d that the num mber of cook king areas

288

Chapter Six

may give an indication of nuclear families within the household, although they also report Henry Wright’s observation that these, “may also reflect the separation of cooking by socio-economic or ritual status” (Matney, Algaze, and Pittman 1997, 65). The information gathered from the sixty-eight houses analysed in this thesis does not seem sufficient to support this hypothesis. Often, only one cooking facility has been found in a building. Where more than one has been found, this could indicate two different phases of the building. In some buildings, a domestic oven and a hearth have been found and the two different facilities could have had different functions; for example, one for cooking and one for heating. In particular for the sixty-eight buildings analysed in this thesis, the locations of hearths vary too much to see a pattern. The number of hearths is too small to support affirmations such as that a hearth can help to identify a space as a living room (see, for example, Postgate 1994, 59–60), and even less to support that the number of hearths is in proportion to the number of nuclear families present in a household, as explained in chapter two, section 2. In Fig 6.9, the buildings where basins and benches have been found are grouped together. It is evident that the majority of these features have been found in Tell Melebiya. Many houses on this site had several rooms with one or more basins and up to three benches, usually along walls, but there does not seem to be a statistical correlation or pattern evident between benches and basins. In the attempt to recognise a pattern it is possible to notice that basins were often in rooms on the sides of the buildings, although in House B1 a basin was present in room 7, which was large and central. Lebeau identified these rooms as workshops, adding that the floors were often plastered (Lebeau 1993, 106). In a final report, Lebeau recognises a more complex function for those rooms perhaps related to their “private” position in the layout (Lebeau 1993, 104). He seems to suggest that these rooms were not utilised as workshops only.

Discussion

289

Fig. 6.9a. Loocations of bennches and basins, from Matthew ws and Postgatee (1987), Algaze ett al. (2001), Deelougaz, Hill, an nd Lloyd (1967 ), and Lebeau (1993) (

290

Chapterr Six

Figg. 6.9b. Locationn of benches an nd basins, from (Lebeau 1993))

Discussion

291

It is possible to affirm that some craft activity was conducted in Melebiya houses, given the large number of basins found on the site (Fig. 6.9). A basin located in a small room means that it is possible the room was used for craft activity exclusively, but when found in a larger room this means it might be possible that multiple activities were taking place in the same space. For the several basins found in Titris Höyük (only one in the buildings analysed for this project), the archaeologists suggest production activities such as the washing and processing of wool or fleece and the processing of grapes for wine production because of the presence of tartaric acid in the plaster (see chapter three, section 2 and Matney, Algaze, and Pittman 1997, 65). Henrickson, who analysed the distribution of water-related facilities such as basins and drains in Tell Asmar and Khafajah houses, could not deduce a functional interpretation for the spaces where the facilities were found (Henrickson 1981, 64). It has already been explained that Lebeau identified “gathering rooms” as the largest spaces with benches and that the concept of “benches” should be reconsidered in light of the ethnographic records (see chapter three). Hansen describes activities that take place almost exclusively on the floor, which was the place for work, sleeping, and eating. She describes a specific use for “benches along walls,” which was to store bedding and other household objects (Hansen 1961). Aurenche confirmed that benches were used for storage purposes in their ethnographic observations (Aurenche, Bazin, and Sadler 1997). A similar use of “benches” in Melebiya houses appears to be possible, and probably a more appropriate name for these “benches” would be “shelves.” If the benches were used as storage places, all the spaces with benches could have been at least partially used as store rooms, in contradiction of Lebeau’s interpretation of the use of space. There was in fact an absence of storage areas in Lebeau’s identification of room functions. He suggests that big jars could have been placed on the terraced roofs (Lebeau 1993, 105). On other sites, very few benches were found (one in Abu Salabikh and one in Khafajah), and only one basin was found in Titris Höyük (see Fig. 6.9). A possible explanation for this fact is that other sites were not so well preserved or well excavated, or that the abundance of basins and benches in houses was a peculiarity of Tell Melebiya related to some specific cultural trait or craft activities carried out at the site.

292

Chapter Six

Other specific fixed features found in the analysed buildings are two altars identified in House B2 and House B7 in Tell Melebiya (see chapter three and Lebeau 1993). The two altars were found in large spaces where benches were also found, and in the case of House B2 were a basin was also present. These rooms could have been used exclusively for cultic activities as chapels or could have been multipurpose spaces with domestic altars. The method for understanding the use of space in the houses was comprised the analysis of specific semi-fixed or fixed structures that indicate specific functions (e.g. ovens, hearths, bins, and basins), and the analysis of objects and other materials (e.g. ceramics and ashes). The next step was the analysis of the size and position of spaces (i.e. a small room farther from the entrance was more likely to be a storage space, while a large room was more probably a multipurpose living space as explained before). This interpretation of spaces seems more realistic than the interpretations that identified bedrooms and assigned areas to different nuclear families within an extended family. As it is possible to see from chapter two, in traditional houses in Iran, Iraq, and Turkey the spaces had very flexible uses, and there were no modern Western concepts such as bedroom or dining room. The sleeping and eating patterns were constantly changing, there was no privacy as intended in contemporary Western dwellings, and there was no nuclear family division when different generations were present in a house. The lifestyle described above produces a material culture very similar to the one we see evidence of in archaeological excavations from sites in the same geographical area and dated to the third millennium BC. As a consequence, it has been hypothesised that spaces without a specific function were probably used for household activities such as sleeping, eating, craft production and all the other activities that did not require specific features. Henrickson’s analysis of the distribution of serving vessels goes in the same direction. She noticed that in Tell Asmar they were more often found in rooms closer to the outside and in Khafajah more often in less accessible rooms (Henrickson 1981, 62). In general, she found that serving vessels were used throughout the house. Her interpretation of that evidence is that serving vessels were easily moved and perhaps different

Discussion

293

groups could have eaten in different rooms at different times, possibly based on sex and age (Henrickson 1981, 63). If we combine this information with the interpretative models from the ethnographic records analysed in chapter two and mentioned above, in the ethnographic analysis (Hansen 1961, 42), it is possible to hypothesise that the eating patterns were constantly changing and people arranged themselves in different groups and in different rooms for each meal. The excavations of the sites of Tell Asmar and Khafajah brought to light many objects that allowed Henrickson to further the understanding of the use of space (Henrickson 1981). Her research gave the same result as this thesis regarding the cooking areas that were mostly located in spaces near the entrances. She also found that most long-term storage was relegated to small rooms at the backs of the buildings and that the other larger spaces were multipurpose areas where serving and eating food, temporary storage, relaxation, and sleeping all took place (Henrickson 1981, 69). The exceptionality of the findings of objects at Tell Asmar and Khafajah such as seals and seal impressions in buildings allowed Henrickson to hypothesise that business meetings and transactions were carried out in the multipurpose rooms of the houses (Henrickson 1981, 69). Finally, small rooms, which were often relatively private, have been interpreted as storage places.

6.3 What Evidence is there of Variation in the Wealth of Mesopotamian Houses? We saw in chapter four that Henrickson (1981, 69–74) took various elements into consideration as indicators of wealth: • • • •

the dimensions of the dwelling (a large house could have belonged to a wealthier family) the presence of luxury objects in the rooms and/or the burials associated with the house the presence of built tombs (for example vaulted tombs) in contrast to simple pit graves more often associated with small houses the presence of water-related facilities

As we saw in chapter one, section 3.4, other archaeologists agree on the identification of larger houses with wealthier families because richer

294

Chapter Six

households could afford polygamous marriages or more servants (see for example Schloen 2001, 181). The four points above have been applied to the evidence collected when looking for variations in the wealth of the houses analysed. Houses in Tell Melebiya have been divided into two groups in chapter three, section 1 on the basis of the surface area. houses B1, B2, B4, B7, and G1 were the largest and were organised with ten to nineteen rooms around a central space. They were well planned and had the characteristics of wealthy households. The other three Melebiya houses analysed were small with three to four rooms and probably belonged to non-wealthy households. The few objects found in Melebiya houses are not of help in understanding the variation of wealth among them. Henrickson considered water facilities as evidence of wealth in houses. At Tell Melebiya, many basins and drains were found in houses, mainly in the large ones, but in one small house a basin was found as well. It appears from the Tell Melebiya information that water facilities were generally correlated with large houses, but were not sufficient evidence of wealth. Evidence in Titris Höyük came from two different excavated areas (see chapter three, section 2 and Algaze et al. 2001). In the Lower Town, three buildings were analysed which were large houses, well-built, and with signs of planning. Algaze considered them non-elite houses on the basis of comparison with similar houses found in contemporary neighbourhoods. They have the characteristics of buildings occupied by wealthy households. Algaze found houses in the Outer Town area to be smaller and not well planned. Moreover, in houses from the Lower Town area, richer intramural burials and a jewellery mould were found. As has been said, the site of Tell Taya did not undergo excavations (see chapter three, section 2; Reade 1971; 1973). The information in this thesis was extracted from the site plan drawn on the basis of the surface remains of building foundations. Among the nine analysed buildings, one was a small house of less than sixty metres2, while the others were larger than 120 metres2, and five of them comprised a central space surrounded by rooms. The buildings seem to be planned and there are signs of wall alignments in the neighbourhoods. These could have been inhabited by wealthy households but there is no other information useful in helping the analysis. The excavations of the sites of Tell Asmar and Khafajah brought to light many objects, among them several luxury goods that allowed Henrickson

Discussion

295

to produce a study of the variation of wealth in the houses of the two sites (Henrickson 1981, 69–74). Her study revealed that luxury objects such as maceheads, stone vases, beads, and amulets were found in houses I and XX at Tell Asmar, and houses XXXII, XXXVIII, and L at Khafajah, which were all large houses (over 130 metres2). Another characteristic of these dwelling was that they were all built to the “ideal” plan, as Henrickson calls it (Henrickson 1981, 71). They were in fact spacious and regular buildings with a main large central space surrounded by rooms. As a consequence, she considered other houses with the same characteristics to have belonged to wealthy families, such as houses XXV and XXII in Stratum Va, Houses I and II in Stratum b-c, and houses XLVI and XLV. Henrickson attributes the lack of valuable objects in the houses to the removal by the occupants or robbers (Henrickson 1981, 71). Henrickson also gives the type of grave as a criterion to identify wealthy households, but in her exhaustive study she warns that the typology of graves is not sufficient evidence of identification of a wealthy household because even if there is a statistical correlation between size of house and type of grave, with large houses usually having built tombs, the opposite also occurred with built tombs in small houses and simple pit graves in large houses (Henrickson 1981, 71–3). Abu Salabikh is a peculiar case where it is possible to analyse only four very large buildings, among the largest analysed in this thesis (Fig. 6.1). The high standardisation in the pattern of spaces in the buildings made Matthews think that the buildings were fulfilling some social function (Matthews and Postgate 1987). Moreover, the absence of cottage industries can contribute to the thinking that the Abu Salabikh buildings belonged to the highest sector of the population and perhaps administrative functions were carried out in them. In the five analysed houses of Fara, no burials or objects have been found and there was also a general lack of facilities and other features. The only element to take into consideration is the surface of houses. The five analysed houses seem quite homogeneous, and none of them seems to belong to poor households because they occupied an extensive area and were composed of many rooms. On the basis of the previously stated criteria it is possible to see that among the sixty-eight houses analysed on the sites, two groups were present: one group of wealthy households generally occupying large houses and owning luxury objects, and many less-wealthy households owning small houses often composed of three to four rooms. The

Chapter Six

296

interesting fact to be noted is that, at some sites (Tell Melebiya, Tell Taya, Khafajah, and Tell Asmar), both groups coexisted in the same areas, and often very wealthy houses shared walls with non-wealthy ones. In the four sites mentioned, a differentiation of the neighbourhood on the basis of wealth as often happens in modern contemporary cities does not seem to be present. In the other three sites, only wealthy houses have been found (Titris Höyük, Abu Salabikh, and Fara), but they are also the sites were small numbers of houses per site were found (four in Titris Höyük, four in Abu Salabikh, and five in Fara), and they are usually isolated houses not belonging to a complete neighbourhood.

6.4 What was Private and Public for the Occupants of Mesopotamian Houses? For this project, sixty-eight buildings have been analysed and flowcharts to represent the circulation patterns and quantify the degree of permeability have been built whenever possible. The principle applied is that a space was more “public” when it was closer to the entrance and more “private” if it was further from the outside world. What emerged by overlapping the flow charts with the interpretation of the use of space is that the spaces identified as storage rooms had the highest degree of privacy in third-millennium houses in Mesopotamia and that the spaces closest to the outside were rooms with cooking facilities. The two domestic altars that have been found were in what looked like large multipurpose rooms towards the backs of the houses in a relatively private area. Larger spaces with a “medium” level of privacy were multipurpose rooms and courtyards where eating and sleeping took place. All the evidence, from fixed features to objects found in the buildings, to architectural organisation, to the analysis of the circulation, points towards a fluid use of space. It is easier to enumerate what there is no evidence of, in comparison to some traditional ideas of archaeologists working on private houses in the Near East: • •

•

There is no evidence of a distinction between “gathering rooms” (for the family) and “reception rooms” (for visitors) There is no evidence of a division of space on the basis of different nuclear families comprising one household (one hearth = one living room = one nuclear family, as in Kramer [1982a] and Brusasco [1999–2000], etc.) There is no evidence of specific private places for sleeping as intended in Western bedrooms

Discussion

•

297

Theree is no evidennce of genderr-specific segrregated areas or rooms such as in later Islaamic tradition n

Fig. 6.10. L Locations of coooking facilities,, internal courtyyards, and strod de rooms (part 1)

298

Chapterr Six

Fig. 6.11. Locations of coooking facilities,, internal courtyyards, and strod de rooms (part 2)

Discussion

299

Fig. 6.12. Location of coooking facilities, internal courtyyards, and strodee rooms (part 3)

300

Chapterr Six

Fig. 6.13. Locations of coooking facilities,, internal courtyyards, and strod de rooms (part 4)

Discussion

301

Fig. 6.14. Locations of coooking facilities,, internal courtyyards, and strod de rooms (part 5)

302

Chapterr Six

de rooms Fig. 6.15. Locations of coooking facilities,, internal courtyyards, and strod (part 6)

Discussion

303

Fig. 6.16. Locations of coooking facilities,, internal courtyyards, and strod de rooms (part 7)

304

Chapterr Six

Fig. 6.17. Locations of coooking facilities,, internal courtyyards, and strod de rooms (part 8)

Discussion

305

In Figs. 6.10, 6.11, 6.12 6.13, 6.14, 6.15, 6.16, and 6.17, the flowcharts of the largest buildings analysed in this thesis are regrouped. They are obviously more complex because, in contrast, the internal circulation in small houses of three to four rooms is extremely simple and gives little information about the concepts of private and public. In the figures above, the rooms in which cooking facilities have been found in excavations have been marked with red circles. The spaces marked with a yellow circle are those identified as internal courtyards by archaeologists or the result of hypotheses made in this work. The same applies to the green circles that identify storerooms. In Fig. 6.10 it is possible to see that houses at Tell Melebiya have a treelike shape that in the spatial analysis theory (see chapter two) is associated with a more-accessible house and a less-hierarchical household. The circulation of houses in Titris Höyük produced a different type of flowchart with a more elongated shape. This is due to the division of the houses into two sectors (a main house with a central courtyard surrounded by rooms and an annex structure connected by a single door). Algaze proposed two possible explanations for this peculiar layout. The first hypothesis is the separation of private space from public space (the main house being private, and the annex being public). The second hypothesis is a gender separation in the space (Algaze et al. 2001). The first hypothesis seems more probable. Archaeologists made several attempts to demonstrate the existence of gender-specific areas in Mesopotamian houses but it seems very difficult to find archaeological evidence of this. In the best cases there is evidence of craft or commercial activities which are not demonstrated to be gender related and that could had taken place in specific separated areas. If the ethnographic models are useful in indicating a flexible use of space and an absence of privacy as intended in contemporary Western society, there is in those models the big difference of the presence of a guest room immediately accessible from the outside, allowing women to remain unseen by male visitors, which seems to be a very Islamic characteristic. There is little information about the relationship between genders and the position of women in Mesopotamian society, but the information we do have does not seem to point to a condition identical to the traditional Islamic one. From textual evidence we know that Mesopotamian women could own and manage properties. There were even cases of female

306

Chapter Six

scribes who received the same education as their male counterparts and were able to write all kinds of documents, including contracts and wills (Lion and Robson 2005). The second peculiar characteristic is the presence of “rings” in the circulation of the houses. There is a ring in a flowchart when one space is accessible in more than one way (see chapter three, section 2). Hillier and Hanson associate “rings” in circulation with a more-accessible house and a less-hierarchical household. It is probably not possible to demonstrate if this is true for Titris Höyük houses, but it is important to observe the presence of at least one ring or more in the four houses analysed at Titris Höyük when often none are present in other sites. Observing the other flowcharts built to show the circulation in the houses, it is possible to notice that they generally have a tree shape except for a few exceptions, such as the Southern Unit in Abu Salabikh, for example. Few cases of “rings” were present. The spaces identified as courtyards never opened directly onto the outside, and they were the centre of the distribution of the circulation in the houses. It is possible to make some general considerations applying the spatial analysis theory: generally, the rooms in Mesopotamian houses were well integrated and only rooms near the entrance had high accessibility. Following the spatial syntax theory, which considers the degree of permeability connected to the function of a room, less-integrated rooms could possibly have been used for commercial activities, and/or in the larger buildings for administrative activities related to private and even government business.

CHAPTER SEVEN CONCLUSIONS

For this project, sixty-eight houses from seven sites dating to the mid-late third millennium BC have been analysed. The intent was to identify common aspects and differences to compose a picture of Mesopotamian houses and households. The value of this work is in its large scope, as many of the relevant publications that address domestic buildings have focused on only one or two sites. Moreover, an integrated method of analysis has been developed that can be applied to minimal evidence such as the surface remains of Tell Taya or to rich contexts such as Khafajah or sites excavated with modern techniques such as micro-artefact analysis. The picture that appears from the analysis is that, although coming from different areas of Greater Mesopotamia and divided into different political entities, houses in cities shared similar characteristics: generally, town houses were not freestanding but built contiguously and shared walls. Usually, there were signs of urban planning in the neighbourhoods such as wall alignments, regular measures of house spaces, and straight streets with the doorways of the houses opening onto them. The surface areas of houses could vary greatly. In this thesis, buildings between 25 and 525 metres2 have been analysed. It has been possible to recognise two groups of households: one consisting of large houses with a central courtyard surrounded by smaller rooms and owning luxury objects, probably occupied by wealthy households composed of an extended family, and smaller houses often composed of three to four rooms without a central courtyard probably occupied by less-wealthy households composed of a nuclear family. From the evidence in this work it seems that both groups of households were often coexisting in the same neighbourhood, and very wealthy houses were sharing walls with non-wealthy ones. This seems to go in the same direction as the study of Stone (1987) about Old Babylonian Nippur, where she hypothesised that the population in a neighbourhood belonged to all classes. The topic needs further exploration, and one direction for

308

Chapter Seven

future work will be the application of the methods used in this thesis to the analysis of neighbourhoods to better understand their social composition because of the increasing availability of extensive evidence coming from the use of geo-archaeological techniques and remote sensing where the main information available is the outlines of houses in neighbourhoods. The evidence of the use of space inside houses suggests a fluid use of the floor surfaces of large rooms that were probably places to eat, work, relax, sleep, entertain guests, and perform cultic activities. The open courtyard, when present, was probably the main locus of activities, weather permitting, because it was the main source of light with the houses probably having a few small windows. Similar activities as in a multipurpose room were probably carried out in open courtyards, varying according to season and changing throughout the twenty-four hours of the day, such as social activities, the preparation of food, craft activities, and possibly also sleeping during the hot season. Small rooms appeared to be more task specific: when domestic ovens have been found they are often placed in small rooms not far from the entrance of a house, while storage rooms were also small private spaces, although some storage areas were present in the large multipurpose rooms, and different kinds of goods were probably stored in different places, perhaps on the basis of more long-term goods versus everyday items. Houses were also economic units, some of them being centres of the production of craft activities. This seems to go in the same direction as studies exploring the economic exchange of a number of necessities that were not provided by the temple or the palace (Widell 2005), and hypotheses stating that craft production could have been a part-time activity not exclusively belonging to specialised artisans (Schloen 2001). What emerges is a rather flexible use of space: sleeping and eating patterns were perhaps changing, no nuclear family divisions when different generations were present in a house were found, and there was probably no privacy as intended in contemporary Western dwellings. If there was probably little privacy among the members of a household, there was, on the contrary, privacy from the outside. Generally, the rooms in Mesopotamian houses were well-integrated and only rooms near the entrance had a high accessibility from the outside. Less-integrated rooms could have been used for commercial activities, and/or in the larger

Conclusions

309

buildings for administrative activities related to private and even government business. It has been highlighted that many wealthy households have been found among the sixty-eight houses analysed, a fact that seems to agree with the general trend noted by historians that saw in third-millennium Mesopotamia a population concentrated in towns and a reorganisation of the economy with the presence of wealthier households compared to the fourth millennium, which included members lacking any family attachments, such as servants (Pollock 1999, 117). The database of houses collected in this thesis can be used as a reference for other sites of the same period to analyse houses and households, and moreover in future work it will be useful to compare and contrast the information from mid-late third-millennium households with evidence from different periods in order to understand variation throughout time. For example, the evidence of wealth can be useful to investigate signs of processes described by historians such as an increase in propertyownership among the elites, and a decrease in the number of small familyowned properties (see Pollock 1999). Limitations to this study are: •

• • •

The study of circulation in houses and use of space was limited to ground floors. There is no information about activities on roofs or second floors The heterogeneity of samples made a comprehensive statistical analysis challenging The sample considered was from an urban environment. No rural houses have been analysed The houses analysed may not be representative of the dwellings of the entire society. Sectors of the population might have lived in dwellings of unpreserved materials. It is reasonable to hypothesise that poorer strata of population lived not in brick buildings but in tents or dwellings made of wood or swamp reeds, for example

From the information emerging from this study it is possible to make suggestions to archaeologists who will undertake the excavation of houses in the future. In this thesis, many methods for the analysis of evidence from published reports have been explored. It emerged that it is very important to find the complete layout of the ground floor of a building, where it has been preserved. This might seem obvious, but too often

Chapter Seven

310

private houses are not a priority in excavations, and too often in excavation reports only a few rooms of a house are present. From the layout is possible to extract more information than what we can understand from an incoherent group of rooms, even if they are rich in well-preserved features and objects. The following list enumerates, in order of importance, what to look for in the excavation of Mesopotamian houses: • • • • • • •

A complete layout The presence or absence of an open courtyard The presence of absence of a second floor Fixed features Graves Micro-debris analysis, analysis of chemical and organic remains, etc. Objects found in situ

Objects are at the bottom of the list because, although they can give information about the household they belong to, it is not possible to know if they have been moved, and so they are not a safe indication of the function of a space. After the systematic analysis, many questions remain open. Following the end of this project, the hope is that the effort has contributed to the advancement of the knowledge of private dwellings of third-millennium Mesopotamia.

BIBLIOGRAPHY

Algaze, G., G. Dinckan, B. Hartenberger, T. Matney, J. Pournelle, L. Rainville, S. Rosen, R. Rupley, D. Schlee, and R. Vallet. “Research at Titris Höyük in Southeastern Turkey: The 1998 Season.” Anatolica (2001): 23–106. Algaze, G., P. Goldberg, D. Honca, T. Matney, A. Misir, A. M. Rosen, D. Schlee, and L. Somers. “Titris Höyük: a Small EBA Urban Center in SE Anatolia: The 1994 Season.” Anatolica (1995): 13–64. Algaze, G., J. Kelly, T. Matney, and D. Schlee, “Late EBA Urban Structures at Titris Höyük, Southeastern Turkey: The 1995 Season.” Anatolica (1996): 129–43. Algaze, G. and A. Misir. “Titris Höyük, a Small Mid-Late third millennium urban center in southestern Anatolia, 1992.” Kazi Sonuclari Toplantisi (1994): 153–70. Algaze, G., A. Misir, and T. Wilkinson. “Sanliurfa Museum/University of California Excavations and Surveys at Titris Höyük, 1991: A Preliminary Report.” Anatolica (1992): 33–60. Allison, P. M. (ed.). The Archaeology of Household Activities. London: Routledge, 1999. Andreae, W. “Ausgrabungen in Fara und Abu Hatab.” Mitteilungen der Deutschen Orient-Gesellschaft 17 (1903): 4–35. Aurenche, O. La maison orientale: l’architecture du Proche Orient ancien des origines au milieu du quatrième millénaire. Paris: P Geuthner, 1981. —. “Famille, fortune, pouvoir et architecture domestique dans les villages du Proche Orient.” In Houses and Households in Ancient Mesopotamia, edited by K. R. Veenhof, 1–16. Istanbul: Nederlands Historisch-Archaeologisch Instituut, 1996. Aurenche, O., M. Bazin, and S. Sadler. Villages engloutis: enquête ethnoarchéologique à Cafer Höyük, number 26 in “Travaux de la Maison de l’Orient méditerranéen,” Maison de l’Orient méditerranéen. Lyon, 1997. Battini, L. “Pour une nouvelle classification de l’architecture domestique en Mésopotamie du IIIe au Ier mill. av. J.-C.” Akkadica 127 (2006): 73–92.

312

Bibliography

Battini-Villard, L. L’éspace domestique en Mésopotamie de la IIIe dynastie d’Ur à l’époque paléo-babylonienne. BAR International Series 767. Oxford: J and E Hedges and Archaeopress, 1999. Batty, M. and S. Rana. “Reformulating Space Syntax: The Automatic Definition and Generation of Axial Lines and Axial Maps.” In Working Paper Series, number 58, CASA (Centre for Advanced Spatial Analysis), UCL, London, 1–41. 2002. Bazzana, A. and M. Delaigue (eds.). Etno-archéologie méditerranéenne. Madrid: Casa de Velazquez, 1995. Ben-Shlomo, D. and Y. Garfinkel. “Sha’ar Hagolan and New Insights on Near Eastern Proto-historic Urban Concepts.” Oxford Journal of Archaeology 28 (2009): 189–209. Bernbeck, R. “Review.” Journal of the Economic and Social History of the Orient 1 (2006): 122–6. Biewers, M. “L’ habitat traditionnel contemporain, sa construction et ses racine dans la maison du proche Orient ancien (trois études de villages jordaniens).” In Ethno-archéologie méditerranéenne, edited by A. Bazzana and M. Delaigue, 29–52. Madrid: Casa de Velazquez, 1995. Braemer, F. “Architecture domestique de l’age du fer. En Syrie du nord.” In Les maisons dans la Syrie antique du IIIème millénaire aux débuts de l’Islam. Pratiques et représentations de l’éspace domestique, edited by C. Castel, M. Al Maqdissi and F. Villeneuve, 61–71. Beyrouth: Institut français d’archéologie du Proche-Orient, 1997. Brusasco, P. “Family Archives and the Social Use of Space in Old Babylonian Houses at Ur.” Mesopotamia 34–35 (1999–2000): 1–173. —. The Archaeology of Verbal and Nonverbal Meaning. Mesopotamian Domestic Architecture and its Textual Dimension. Number S1631 in British Archaeological Reports. Oxford: Archaeopress, 2007. Butterlin, P. “Tepe Gawra et le monde proto-urbain de Mésopotamie.” Subartu (2009): 1–14. Castel, C. and D. Charpin. “Les maisons mésopotamiennes. Essai de dialogue entre archéologue et épigraphiste.” In Les maisons dans la Syrie antique du IIIème millénaire aux débuts de l’Islam. Pratiques et représentations de l’éspace domestique, edited by C. Castel, M. Al Maqdissi, and F. Villeneuve, 243–53. Beyrouth: Institut français d’archéologie du Proche-Orient, 1997. Chesson, M. S. “Urban Households in Early Bronze Age Communities of Syro-Palestine.” PhD thesis. Harvard University, 1997. Conkey, M. W. and J. M. Gero. “Programme to Practice: Gender and Feminism in Archaeology.” Annual Review of Anthropology 26 (1997): 411–37.

Archaeological Perspectives on Houses and Households in Third Millennium Mesopotamian Society

313

Dawson, P. C. “Space Syntax Analysis of Central Inuit Snow House.” Journal of Anthropological Archaeology (2002): 464–480. Deblauwe, F. “Spacings and Statistics, or a Different Method to Analyze Buildings. A Test with Mesopotamian Houses from the Late Bronze and Iron Ages.” Akkadica (1994): 1–8. Delougaz, P., H. D. Hill, and S. Lloyd. “Private Houses and Graves in the Diyala Region.” Oriental Institute Publications vol. LXXXVIII. University of Chicago Press, Chicago, 1967. Fahlander, F. “Archaeology and Anthropology: Brothers in Arms?: on Analogies in 21st Century Archaeology.” In Material Culture and other Things—Post-disciplinary Studies in the 21st Century, edited by F. Fahlander and T. Oestigaard, 185–211. Gothenburg: University of Gothenburg, Department of Archaeology 2004. Forest, J. “L’habitat urukien du Djebel Aruda. Approche fonctionelle et arrière-plans symboliques.” In Les maisons dans la Syrie antique du IIIème millénaire aux débuts de l’Islam. Pratiques et représentations de l’éspace domestique, edited by C. Caste, M. A. Maqdissi, and F. Villeneuve, 217–234. Beyrouth: Institut français d’archéologie du Proche-Orient, 1997. Foucault-Forest, C. “Modèles d’organisation de l’éspace dans l’habitat du bronze moyen et du bronze récent en Palestine.” In Les maisons dans la Syrie antique du IIIème millénaire aux débuts de l’Islam. Pratiques et représentations de l’éspace domestique, edited by C. Caste, M. A. Maqdissi, and F. Villeneuve, 151–60. Beyrouth: Institut français d’archéologie du Proche-Orient, 1997. Frankfort, H., T. Jacobsen, and C. Preusser. Tell Asmar and Khafaje: The First Season’s Work in Eshnunna 1930/31, number 13 in Oriental Institute Communications. Chicago: The University of Chicago Press, 1932. Frankfort, H. Tell Asmar, Khafaje and Khorsabad: Second Preliminary Report of the Iraq Expedition, number 16 of Oriental Institute Communications. Chicago: The University of Chicago Press, 1933. —. Iraq Excavations of the Oriental Institute 1932/33: Third Preliminary Report of the Iraq Expedition, number 17 in Oriental Institute Communications, Chicago: The University of Chicago Press, 1934. —. Oriental Institute discoveries in Iraq 1933/34: Fourth Preliminary Report of the Iraq Expedition, number 19 in Oriental Institute Communications. Chicago: The University of Chicago Press, 1935. —. Progress of the Work of the Oriental Institute in Iraq, 1934/35: Fifth Preliminary Report of the Iraq Expedition, number 20 in Oriental

314

Bibliography

Institute Communications. Chicago: The University of Chicago Press, 1936. Hansen, D. P. Inscriptions from Tell Abu Salabikh, vol. XCIX of The University of Chicago Oriental Institute Publications. Chicago: The University of Chicago Press, 1974. Hansen, H. H. The Kurdish Woman’s Life. Field Research in a Muslim Society, Iraq. København: Nationalmuseet, 1961. Hanson, J. “Deconstructing Architects’ Houses." In CASA (Centre for Advanced Spatial Analysis), UCL, Environment and Planning B, number 21, 675–705. London, 1994. Hanson, J. Decoding Houses and Homes. Cambridge: Cambridge University Press, 1998. Harris, R. “The Archive of the Sin Temple in Khafajah (Tutub).” Journal of Cuneiform Studies 9 (1955): 31–58. Heinrich, E. Fara. Ergebnisse der Ausgrabungen der Deutschen OrientGesellschaft in Fara und Abu Hatab 1902/1903. Berlin: Vorderasiatische Abeilung der Staatlichen Museen zu Berlin, 1931. Henrickson, E. F. “Non-religious Residential Settlement Patterning in the late Early Dynastic of the Diyala Region.” Mesopotamia XVI (1981): 43–140. Hillier, B. Space is the Machine: A Configurational Theory of Architecture. Cambridge: Cambridge University Press, 1998. Hillier, B. and J. Hanson. The Social Logic of Space. Cambridge: Cambridge University Press, 1984. Hodder, I. The Present Past. An Introduction to Anthropology for Archaeologists. London: B. T. Batsford ltd., 1982. Horne, L. Village Spaces: Settlement and Society in Northeastern Iran. Smithsonian Series in Archaeological Inquiry. Washington DC, London: Smithsonian Institution Press, 1994. Jameson, M. H. “Domestic Space in the Greek City-state.” In Domestic Architecture and the Use of Space: an Interdisciplinary Cross-cultural Study, edited by S. Kent, 92–113. Cambridge: Cambridge University Press, 1990. Kent, S. “Activity Areas and Architecture: an Interdisciplinary View of the Relationship between Use of Space and Domestic Built Environments.” In Domestic Architecture and the Use of Space: an Interdisciplinary Cross-cultural Study, edited by S. Kent, 1–8. Cambridge: Cambridge University Press, 1990. Kohlmeyer, K. “Houses in Habuba Kabira South: Spatial Organization and Planning of Late Uruk Residential Architecture.” In Houses and

Archaeological Perspectives on Houses and Households in Third Millennium Mesopotamian Society

315

Households in Ancient Mesopotamia, edited by K. R. Veenhof, 89– 104. Istanbul: Nederlands Historisch-Archaeologisch Instituut, 1996. Kramer, C. “An Archaeological View of a Contemporary Kurdish Village: Domestic Architecture, Household Size, and Wealth.” In Ethnoarchaeology: Implications of Ethnography for Archaeology, edited by C. Kramer, 139–63. New York: Columbia University Press, 1979. —. “Ethnographic Households and Archaeological Interpretation. A Case from Iranian Kurdistan.” American Behavioral Scientist 25 (1982a): 663–75. —. Village Ethnoarchaeology. Rural Iran in an Archaeological Perspective. New York: Academic Press, 1982b. Larsen, M. T. “Individual and Family in Old Assyrian Society.” Journal of Cuneiform Studies 59 (2007): 93–106. Lawrence, R. J. “Public Collective and Private Space: a Study of Urban Housing in Switzerland.” In Domestic Architecture and the use of Space: an Interdisciplinary Cross-cultural Study, edited by S. Kent, 73–91. Cambridge: Cambridge University Press, 1990. Lebeau, M. Tell Melebiya. Cinq campagnes de recherches sur le MoyenKhabour (1984–1988), number IX in Akkadica Supplementum. Leuven: Peeters, 1993. —. “Les maisons de Melebiya. Approche fonctionnelle de l’habitat privée au III Millénaire av. notre ère en haute Mesopotamie.” In Houses and Households. 40th Rencontre Assyriologique Internationale, 129–136. Leiden/Istanbul, 1996. Lévi-strauss, C. La voie des masques. Paris: Plon, 1979. Lion, B., and E. Robson. “Quelques textes scolaires paléeo-babyloniennes réedigées par des femmes.” Journal of Cuneiform Studies 57 (2005): 37–54. Liverani, M. Antico Oriente. Storia società economia. Bari: Editori Laterza, 1988. Lloyd, S. “Some Ancient Sites in the Sinjar District.” Iraq 5 (1938): 123– 42. MacEachern, S., D. J. W. Archer, and R. D. Garvin (eds.). Household and Communities. Proceedings of the Twenty-first Annual Conference of the Archaeological Association of the University of Calgary. Calgary: The University of Calgary Archaeological Association, 1989. Manning, S. W., B. Kromer, P. I. Kuniholm, and M. W. Newton, “Anatolian Tree Rings and a New Chronology for the East Mediterranean Bronze-Iron Ages.” Science (2001).

316

Bibliography

Margueron, J. C. “Les maisons syrienne du néolithique au premier millénaire. Quelques remarques sur la documentation et son interpretation.” In Les maisons dans la Syrie antique du IIIème millénaire aux débuts de l’Islam. Pratiques et représentations de l’éspace domestique, edited by C. Caste, M. A. Maqdissi, and F. Villeneuve, 3–8. Beyrouth: Institut français d’archéologie du ProcheOrient, 1997. —. “La maison oriental.” In Houses and Households in Ancient Mesopotamia, edited by K. R. Veenhof, 17–38. Istanbul: Nederlands Historisch-Archaeologisch Instituut, 1996. Martin, H. P. Fara: A Reconstruction of the Ancient Mesopotamian City of Shuruppak. Birmingham: Chris Martin & Assoc., 1988. Matney, T. and G. Algaze. “Urban Development at Mid-Late Early Bronze Age Titris Höyük in Southeastern Anatolia.” BASOR (1995): 33–52. Matney, T., G. Algaze, and H. Pittman. “Excavations at Titris Höyük in Southeastern Turkey: A Preliminary Report of the 1996 Season.” Anatolica (1997): 61–84. Matney, T., G. Algaze, and S. Rosen. “Early Bronze Age Urban Structure at Titris Höyük, Southeastern Turkey: the 1998 Season.” Anatolica (1999): 185–201. Matthews, R. J. The Archaeology of Mesopotamia: Theories and Approaches. Approaching the Ancient World. London: Routledge, 2003. Matthews, R. J. and J. N. Postgate. “Excavations at Abu Salabikh, 198586.” Iraq (1987): 91–119. Matthews, W. Excavations at Tell Brak, vol. 4 of McDonald Institute Monographs. McDonald Institute for Archaeological Research and British School of Archaeology in Iraq. 2003. Matthews, W., J. N. Postgate, S. Payne, M. P. Charles, and K. Dobney. “The Imprint of Living in an Early Mesopotamian City: Questions and Answers.” In Whither Environmental Archaeology?, number 38 in Oxbow Monograph, edited by R. Luff and P. Rowley-Conwy, 171– 212. Oxford: Oxbow Books, 1994. McClellan, T. L. “Houses and Households in North Syria during the Late Bronze Age.” In Les maisons dans la Syrie antique du IIIème millénaire aux débuts de l’Islam. Pratiques et représentations de l’ éspace domestique, edited by C. Caste, M. Al Maqdissi, and F. Villeneuve, 29–59. Beyrouth: eds, Institut français d’archéologie du Proche-Orient, 1997. Meskell, L. Private life in New Kingdom Egypt. Princeton: Princeton University Press, 2002.

Archaeological Perspectives on Houses and Households in Third Millennium Mesopotamian Society

317

Naroll, R. “Floor Area and Settlement Population.” American Antiquity (1962): 587-589. Nishimura, Y. “North Mesopotamian Urban Space: A Reconstruction of Household Activities and City Layout at Titris Höyük in the Third Millennium BC.” PhD thesis. University of California, 2008. —. “The Life of the Majority: A Reconstruction of Household Activities and Residential Neighborhoods at the Late-Third Millennium Urban Settlement at Titris Höyük in Northern Mesopotamia.” In New Perspectives on Household Archaeology, edited by B. Parker and C. Foster, 347–72. Winona Lake, IN: Eisenbrauns, 2012. Pfälzner, P. Haus und Haushalt. Wohnformen des Dritten Jahrtausends von Christus in Nordmesopotamien, vol. 9. Meinz am Rhein: Damaszener Forschungen, 2001. Peponis, J., J. Wineman, S. Bafna, M. Rashid, and S. H. Kim. “On the Generation of Linear Representation of Spatial Configuration.” In Environment and Planning B 25, 559–76. CASA (Centre for Advanced Spatial Analysis), UCL, London, 1998. Peponis, J., J. Wineman, M. Rashid, S. H. Kim, and S. Bafna. “On the Description of Shape and Spatial Configuration Inside Buildings: Convex Partitions and their Local Properties.” In Environment and Planning B 24, 761–81. CASA (Centre for Advanced Spatial Analysis), UCL, London, 1997. Pollock, S. Ancient Mesopotamia: the Eden that Never Was. Case Studies in Early Societies. Cambridge: Cambridge University Press, 1999. Pollock, S., M. Pope, and C. Coursey. “Household Production at the Uruk Mound, Abu Salabikh, Iraq.” American Journal of Archaeology 100 (1996): 683–9. Postgate, J. N. The West Mound Surface Clearance, vol. I of Abu Salabikh Excavations. London: British School of Archaeology in Iraq, 1983. —. “Excavations at Abu Salabikh, 1988–89.” Iraq (1990): 95–106. —. Early Mesopotamia: Society and Economy at the Dawn of History. London: Routledge, 1992. —. “How Many Sumerians per Hectare?” Cambridge Archaeological Journal 4 (1994): 47–65. Postgate, J. N. and J. A. Moon. “Excavations at Abu Salabikh, 1975.” Iraq (1994): 133–70. Rainville, L. Investigating Upper Mesopotamian Households using MicroArchaeological Techniques. British Archaeological Reports. Oxford: Archaeopress, 2005. Rapoport, A. “Systems of Activities and Systems of Settings.” In Domestic Architecture and the Use of Space: an Interdisciplinary

318

Bibliography

Cross-cultural Study, edited by S. Kent, 9–20. New Directions in Archaeology. Cambridge: Cambridge University Press, 1990. —. “Archaeology and Environment-behavior Studies.” Archeological Papers of the American Anthropological Association 16 (2006): 59– 70. Reade, J. E. “Tell Taya (1967): Summary Report.” Iraq 30 (1968): 234– 64. —. “Tell Taya (1968–9): Summary Report.” Iraq 33 (1971): 87–100. —. “Tell Taya (1972–73): Summary Report.” Iraq 35 (1973): 155–87. Sanders, D. “Behavioral Conventions and Archaeology: Methods for the Analysis of Ancient Architecture.” In Domestic Architecture and the Use of Space: an Interdisciplinary Cross-cultural Study, in S. Kent, 43–72. New Directions in Archaeology. Cambridge: Cambridge University Press, 1990. Schloen, J. D. The House of the Father as Fact and Symbol: Patrimonialism in Ugarit and the Ancient Near East. Harvard Semitic Museum Publications. Winona Lake, IN: Eisenbrauns, 2001. Seeden, H. “Aspects of Prehistory in the Present World: Observations Gathered in Syrian Villages from 1980 to 1985.” World Archaeology 17 (1985): 289–303. Sillar, B. Shaping Culture: Making Pots and Constructing Households: an Ethnoarchaeological Study of Pottery Production, Trade and Use in the Andes, number B883 in BAR. International series. Oxford: J. & E. Hedges, 2000. Starzmann, M. T. Archäologie des Raumes. Soziale Praxis und kulturelle Bedeutung am Beispiel der Wohnarchitektur von Fara. Wien: LIT, 2007. —. “Use of Space in Shuruppak: Households on Display.” In Proceedings of 5 ICAANE. Madrid: Universidad Autonoma, 2008. Steadman, S. R. “Recent Research in the Archaeology of Architecture: Beyond the Foundations.” Journal of Archaeological Research 4 (1996): 51–93. Stiles, D. “Ethnoarchaeology: A Discussion of Methods and Applications.” Man, New Series 12 (1977): 87–103. Stone, E. C. “Text, Architecture and Ethnographic Analogy: Patterns of Residence in Old Babylonian Nippur.” Iraq (1981). —. Nippur Neighborhoods, number 44 in Studies in Ancient Oriental Civilization. Chicago: Oriental Institute of the University of Chicago, 1987. Tringham, R. “Archaeological Houses, Household, Housework and the Home.” In The Home: Words, Interpretations, Meanings, and

Archaeological Perspectives on Houses and Households in Third Millennium Mesopotamian Society

319

Environments, edited by D. N. Benjamin, 79–107. Aldershot: Avebury, 1995. Vallet, R. “Habuba Kébira Sud, approche morphologique de l’habitat.” In Les maisons dans la Syrie antique du IIIème millénaire aux débuts de l’Islam. Pratiques et représentations de l’éspace domestique, edited by C. Caste, M. Al Maqdissi, and F. Villeneuve, 105–19. Beyrouth: Institut français d’archéologie du Proche-Orient, 1997. Verhoeven, M. An Archaeological Ethnography of a Neolithic Community: Space, Place and Social Relations in the Burnt Village at Tell Sabi Abyad, Syria. Istambul: Nederlands HistorischArchaeologisch Instituut, 1999. —. “Ethnoarchaeology, Analogy, and Ancient Society.” In Archaeologies of the Middle East, edited by S. Pollock and R. Bernbeck, 251–70. Malden, MA: Blackwell Studies in Global Archaeology, 2005. Watson, P. J. Archaeological Ethnography in Western Iran, number 57 in Viking Fund publications in anthropology. Tucson, AZ: The University of Arizona Press, 1979. Wattenmaker, P. Household and State in Upper Mesopotamia, Washington DC: Smithsonian Institution Press, 1998. Widell, M. “Some Reflections on Babylonian Exchange during the End of the Third Millennium BC.” Journal of the Economic and Social History of the Orient 48 (2005): 388–400. Wilkinson, T. J. “The Structure and Dynamics of Dry-Farming States in Upper Mesopotamia.” Current Anthropology 35 (1994): 483–520. Wright, H. T. “Problems of Absolute Chronology in Protohistoric Mesopotamia.” Paleorient 6 (1980): 93–8. Yon, M. and O. Callot. “L’habitat à Ougarit. A la fin du Bronze recénte.” In Les maisons dans la Syrie antique du IIIème millénaire aux débuts de l’Islam. Pratiques et représentations de l’éspace domestique, edited by C. Caste, M. Al Maqdissi, and F. Villeneuve, 15–28. Beyrouth: Institut français d’archéologie du Proche-Orient, 1997. Zimansky, P. “Archaeology and Texts.” In Archaeologies of the Middle East. Critical Perspectives, edited by S. Pollock and R. Bernbeck, 308–26. Malden, MA: Blackwell Publishing, 2005.