Lithic Technology of Neolithic Syria 1841710458, 9781841710457

Citation preview

Lithic Technology of Neolithic Syria Y oshihiro Nishiaki

BAR International Series 840 2000

Lithic Technology of Neolithic Syria Y oshihiro Nishiaki ,;:;·

BAR International Series 840 2000

This title published by

Archaeopress Publishers of British Archaeological Reports POBox920

Oxford OX27YH

England

BARS840

Lithic Technologyof Neolithic5-yria

© Y Nishiaki 2000

Printed in England by The Basingstoke Press ISBN 1 84171 045 8

All BAR titles are available from: Hadrian Books Ltd 122 Banbury Road Oxford OX27BP

England The current BAR catalogue with details of all titles in print, prices and means of payment is available free from Hadrian Books

ABSTRACT

This study attempts to examine the interpretive possibility of stone tools for understanding cultural development in the Levantine Neolithic. The method employed is a dynamic approach known in French as chatne operatoire.This has been accompanied by the exhausted use of quantitative data from lithic analyses. It puts more emphasis on bebavioral aspects of lithic industries such as core reduction and raw material use rather than on typological study of particular tool forms. By this means it is hoped that a more comprehensive and dynamic picture is provided for the use of lithic material by the Neolithic people in Syria. The raw material procurement, core reduction, and tool manufacturing processes at Pre-Pottery to Pottery were analyzed Neolithic sites in Syria (Douara Cave II, Tell Damishliyya, Tell Nebi Mend and Tell J{asbkasliok in detail to examine their inter-relations, so that the complex factors creating the technological variability of Neolithic assemblages were isolated. The results suggest that one important factor was availability, kinds and procurement methodsof raw material, and that the change from the Pre-Pottery to the Pottery Neolithic was characterized by the increasing use of locally available raw material, as well as a move from specialized blade to more expedient flake production.

m

The period with which this study is concerned is a period when profound changes were occurring in various aspects of human life-styles, when the then still primitive farming community evolved to the more advanced one

that characterizes the later Neolithic period. This study suggests that the change in lithic technology closely reflects those changing patterns in raw material economy, settlement system, subsistence economy, and social organization, and that detailed analysis of lithic technology using pertinent methods can contribute to a better understanding of important developments within the Neolithic communities of the Ancient Near East.

3

ACKNOWLEDGEMENTS

This study refers to numerous lithic collections from Neolithic sites not only in Syria but also in Turkey, Iraq, and Palestine. I wish, first of all, to thank the following institutions and individuals who so generously made their collections available to me: Prof. Takeru Akazawa (1be University Museum, Tokyo) and Mr. KhaleelAssaad (Palmyra Museum) for Douara; Dr. Andrew M.T. Moore (Yale University), Dr. Roger Moorey (Ashmolean Museum) and Dr. Jill Cook (The British Museum) for Abu Hureyra; Dr. P.R.S. Moorey for Jericho; Dr. Peter M.M.G. Akkermans (Rijilcsmuseum van Oudheden, Leiden) and Mrs. Lorraine Copeland (Grand Brassac) for Damishliyya and the Balikh sites; Mr. Peter Parr (Institute of Archaeology, London) for Nebi Mend; Prof. Toshio Matsutani (Institute of Oriental Culture, Tokyo) and Mr. Jean Lazar (Department of Antiquities, Hassake) for Kashkashok; Prof. T. Matsutani for Thalathat; Mlle Bertille Lyonnet (CNRS, UPR35) for the Khabur sites; Prof. Ufuk Esin and Dr. Mahmet Ozdogan (Istanbul University) for Soutbeastern Turkish sites; Dr. David French (British Institute, Ankara) for Central Anatolian sites; Dr. Joan Oates (Girton College, Cambridge) for Choga Mami.

My field work and artifact analysis in Syria have been greatly facilitated by the Syrian authorities, particularly, Dr. Sultan Muhesen and Dr. Adnan Bounni, Directorate General of Antiquities and Museums, Damascus, to whom I am most grateful. This monograph largely represents my doctoral dissertation that was written up in 1991 at the Institute of Archaeology, London. Due to a number of reasons it has been possible to update the original document only partially. Nevertheless I hope that it still makes some contribution to the Neolithic research of the Near East Research for the dissertation was carried out under the supervision of Dr. Ian Glover, whose continuous interest in my work was of great help. Drafts were read and commented on by him. Dr. Andrew Garrard helped me to formulate the outline of the research. Discussions with Mrs. Lorraine Copeland on the Levantine Neolithic lithics have been always enjoyable and instructive. Dr. Mark Newcomer helped me to study obsidian technology. Dr. Christopher Bergman, 3D Environmental Services, Ohio, provided me with specific comments on lithic technology. Dr. Roger Grace, Dr. Norah Moloney and Ms. Yoko Ogawa gave continuous encouragement during my study in London. My study at the University of Tokyo was painstakingly supervised over several years by Prof. Tsuyoshi Fujimoto, Department of Archaeology. Prof. T. Akazawa has been a major source of information about the prehistory of Syria since the 1984 season at Douara. Prof. T. Matsutani, who introduced me to Neolithic research, has offered every possible help for my work in the Khabur region. Prof. Minoru Utada, Department of Mineralogy, 4

kindly conducted an X-ray diffraction analysis of flint samples. Prof. Katsuhiko Ohnuma, Kolmsbikan University, bas given me continuous encouragement as well as technical advice on litbic analysis. Financial support for this study was provided from the Great Britain Sasakawa Foundation, Central Research Fund of London University, Gordon Childe Fund of Institute of Archaeology, a fellowship from the Japan Society for Promotion of Science, and the Scientific ResearchFund of the Ministry of Education, Japan. Yoshihiro Nishiaki The UniversityMuseum The Universityof Tokyo

5

PREFACE

The Neolithic habitation sites of the first fanning communities in the Near East have been extensively studied in the past several decades. They are located in temin which forms a wide arc on the rim of the Arabian Desert known as the Fertile Crescent and finds unearthed by excavations have provided insights into the "Neolithic Revolution" which took place at the end of the Ice Ages, around nine thousand years B.C. At this time populations alteied their ways of life from the (often nomadic) bunting and gathering regimes to ones based on the raising of domesticated plants and animals while living in settled villages or caves. Most of the research projects dedicated to the Neolithic periods were concentrated in two zones: the Levant (Palestine, Jordan, western Syria) and Mesopotamia (eastern Syria, northern Iraq and some areas of Anatolia).

However, there was one area which seemed to have been largely neglected: the central zones of the Fertile Crescent known as the Jazirah, the "island" between the Syrian Euphrates and Tigris Rivers in northern Mesopotamia. This region was exceedingly rich in proto-bistoric and historic remains, but very little was known of the folk who had populated it in the preceding Sth-7th Millennia, while knowledge of 8th-9tb Millennia communities was almost entirely lacking. Furthermore, when early sites were found, the lithic artifact component of the inhabitants' material culture was often not published in detail, or it was studied from only the typological point of view, with little attention given to the technology or to the process of raw material reduction. Thus, important potential sources of understanding of the activities carried out by the Neolithic inhabitants of the site, including their probable relationships with other cultures, were not sufficiently brought out

In this book the author addresses both these problems. He has made the central Fertile Crescent zone particularly bis own focus of research. Not only has be carried out field surveys, during which be discovered new sites, but he has also participated in excavations in different pans of the region (for example at Kasbkashok and Douara Cave). He has also personally examined, in museums and institutions, the collections of Neolithic chipped stone artifacts deriving from the Levant (Jericho) to Iraq (Tbalatbat) as well as Anatolia (Hayaz). Not surprisingly, while at London University in 1992, be chose "Litbic Technology of Neolithic Syria" as the subject for bis PhD dissertation. This book incorporates the same material together with additional data. Dr. Nisbiaki's work in the region coincided with a revival of interest by other teams in the prehistory of the central Fertile Crescent. It is, after all, from the geographical as well as from the archaeological viewpoint, a "zone of ttansition" between two major river basins and between the region's two most influential culture zones : the 6

Levant and Mesopotamia, already mentioned. Several new research projects, mainly in the Balikh and Khabour drainages, but also to the north in EasternAnatolia, have since taken place. As new information has become available, these areas are seen to be far from devoid of Neolithic occupation ; not only have diverse cultures (some not heretofore expected to be present, e. g. Hassuna) been found, but also those not previously known to exist (e.g. at Nevalla Cori; IWlan Cemi). Thus light is at last being thrown on an erstwhile 'cultural gap'. The village communities can now be studied from various aspects: the siting and form of the habitations, the artifact repertoires, trade contacts and other relationships- all in the context of the particular environmental constraints of a region which, although rich in alluvium, vegetation and wildlife in some areas, is plagued by the absence of rainfall in the summer. Through time, as the post-glacial climate fluctuated between more humid or more arid regimes, the populations practicing rainfed agriculture in this zone were obliged to adapt their lifeways to suit the changing conditions. This necessitated the solving of many problems and the invention of new ranges of tools and weapons. These adaptations are examined and discussed by Nishiaki who, by paying meticulous attention to detail as regards the stone artifacts, is able to propose convincing explanations for the diversity of lithic industries in the region during the 8th-5th Millennia B.C. He has opened up a rich methodological vein and his methods could well be acted on in other areas, where they should produce similar valuable results. Lorraine Copeland

7

CONTENTS

LIST OF FIGURES ................................................................................................................................................ 10 LIST OF TABLES ................................................................................................................................................... 11

CHAPTER 1. INTRODUCTION ................................................................................................................ 13 1.1 Aims ............................................................................................................................................ 13 1.2 Natural environments ............................................................................................................... 14 1.3 PPNB and Pottery Neolithic cultures of Syria ..................................................................... 17

CHAPTER 2. LITHIC STUDIF.S OF THE LEV ANTINE NEOLITHIC:

A METHODOLOGICAL REVIEW............................................................................ 22 2.1 2.2 2.3 2.4 2.5 2.6

Introduction................................................................................................................................ 22 Typological study ...................................................................................................................... 22 Technological study .................................................................................................................. 24 Functional study ........................................................................................................................ 26 Contextual study ........................................................................................................................ 28 Conclusion .................................................................................................................................. 29

CHAPTER3. METHOD................................................................................................................................ 31 3.1 3.2 3.3 3.4

Introduction................................................................................................................................ 31 Terminology ............................................................................................................................... 32 The operational chain and archaeological data ....................................................................38 Conclusion .................................................................................................................................. 52

CHAPTER4. PRE-POTTERYNEOLITHICB: DOUARACAVE D ...................................... 53 4.1 Introduction ................................................................................................................................ 53 4.2 4.3 4.4 4.5 4.6

Material ······································································································································· 54 Research history of the Navifonn method ............................................................................55 Analysis ....................................................................................................................................... 57 ProxiJnalscars ............................................................................................................................ 81 Conclusions ................................................................................................................................ 83

CHAPI'ER 5. PRE-POTTERYNEOLITHIC B TO EARLYPOTTERY NEOLITHIC: TELL DAMISIILIYY A ...................................................................... 96 5.1 Introduction ................................................................................................................................ 96 5.2 Material ....................................................................................................................................... 96 8

5.3 Analysis ....................................................................................................................................... 97 5.4 Conclusions .............................................................................................................................. 124

CHAPTER6. EARTLYPOTIERY NEOLITHIC:TELLNEBI MEND (WESTSYRIA).................................................................................................................... 132 6.1 6.2 6.3 64

lnttoduction .............................................................................................................................. 132 Material ..................................................................................................................................... 134 Analysis ..................................................................................................................................... 134 Conclusions .............................................................................................................................. 158

CHAPTER 7. EARLYPOTIERY NEOLITHIC:TELL KASHKASHOKll 162 (NORTHEASTSYRIA) ..................................................................................................

7.1 7.2 7.3 7.4

Introduction .............................................................................................................................. 162 Material ..................................................................................................................................... 164 Analysis ..................................................................................................................................... 166 Conclusions .............................................................................................................................. 181

CHAPTER8. OBSIDIANTECHNOLOGY:TELLKASHKASHOKll AND 189 OTHERSITES ..................................................................................................................... 8.1 8.2 8.3 8.4 8.5

Introduction ................................... ............................................ ............................................... 189 Material ································································· .................................................................... 189

Analysis ..................................................................................................................................... 191 Analysis of side-blow blade-flakes ...................................................................................... 199 Conclusions .............................................................................................................................. 205

CHAPTER9. CONCLUSIONS:CHANGEFROMTHE PRE-POTIERYTO 214 THE POTTERYNEOLITHIC..................................................................................... 9.1 9.2 9.3 9.4 9.5

Introduction ............................................................................................................................. 214 214 Raw material procurement system ....................................................................................... Lithic production system ........................................................................................................ 218 221 Lithic technology and the society ......................................................................................... Conclusion ................................................................................................................................ 222

REFERENCES...................................................................................................................................................... 224

9

LIST OF FIGURES Map showing the Near Eastern Neolithic sites 14 mentioned in the text ------···· F1pre 1.2 Map showing the physiographic divisions of 15 Syria (Suzuki 1973) .......... _____ Flame3.1 Core types ___ ......................................... 34 Flame3.2 Crested piece types .............................................. 35 Flpre 3.3 Core tablet types .................. ___ ............35 Flame3.4 X-ray spectra showing the mineralogical composition of flints from Douara andTell 40 Damishliyya ............. ________ Flpre 1.1

Flame 4.16 Schemati.:: presentation of typology for flint Flpre4.17 Flame 4.18

Flame4.1' F1pre 4.20 F1pre 4.21

Flame3.5 Measurementsof flakes and blades ............... 42 Flame 3.6 Measurements of cores ....................................... 43 Flame4.22 Flame 3.7 Exterior-butt-shape types ................................... 44 Flame4.23 Flame3.8 Dorsal scar panems___ __44 Flpre 3.9 Distal shape types ................................................. 45 Flpre 3.10 Transversal section types ................................... 45

Flame5.1

Flame3.11 Profile types ............................................................ 45 Flame 3.12 Distal end types ................................. ___ 45 F1aare 5.2 Flame 3.13 Schematicpresentationof cbange-of-orienta___ ............. 46 tion IDCthods....................... Flame 5.3 Flame3.14 Tnmcation types ................................ ___ 49 Flpre 5.4 Flame 3.15 Patterns of heating processes ........................... so Flame4.1 Map of the Douara area (F.ndo et al.1978) ..54 Flaare 5.5 Flame 4.2 Suggested model of the reduction sequence of the Navifonn method by Suzuki andAkaFlame 5.6 Flame 4.3 Flame 4.4 Flame 4.5 Flame 4.6 Flame 4.7 Flame 4.8 Flpre 4.9 F1pre 4.10 F1gare 4.11 Flpre 4.12

Flpre 4.13

Flame 4.14

Flpre 4.15

tools from DouaraCave II ..,___ ...._ .._ 1s Points and burins from DouaraCave II ........77 End-scrapers andotherretouchedtools from Douara Cave II ....................................................... 78 Proximal-scarredpieces and their retouch spalls from Douara Cave II ......................... - ... 79 Suggested model showing the technological systemat Douara Cave ll ................................... 85 Flint artifacts from Tell Abu Hureyra, Trench D (Level 1) .............................................................. 'tr1 Flint artifacts from Jericho, Area M .............. 88 Comparison of technologicalattributes of debitage from Douara Cave ll, Tell Abu HureyraandJericho ............................................. 92 Map of Tell Damishliyya (Akmmans 1990) ...................................................................................... 97

Correlation between flint types andblank types at Tell Damishliyya .................................. 98 Cores from Tell Damishliyya ----· 101 Core-trimming pieces and hammentone flakes from Tell Damishliyya ........................ 106 Schemati.:: presentation of typology for flint tools from Tell Damishliyya ........................... 109 Points and sickle elements from Tell Damishzawa (1971) ............................................................ 56 liyya ·--.................. ___ ......... - .. 112 Suggested model of the reduction sequence Flame5.7 Schematir.presentation of sickle elements of the Naviform method by Calley (1986b) from Tell Damishliyya. and their possible 116 ways of hafting .................................................... Navifonn core types ........_____ sa Flame 5.8 Burins from Tell Damisbliyya,Types 1-6 and 8 ..._ ............... ___ .................................... 117 Navifonn cores from Locality 35 (Akazawa 1979b) ....................................................................... 59 Flpre 5.9 Burins, Type 7, andburin spalls from Damishliyya ....___ .................................... 11s Navifonn cores from Douara Cave ll, Types 1-4 ................. _________ 60 Flame5.10 Other retouched tools from Tell Damishliyya Navifonn cores, Types S and 6, andother ----------.. 120 cores from Douara Cave II ............ ___ .. 61 Flame5.11 Correlations between tool types andflint types at Tell Damishliyya _____ 123 Core-trimming pieces from DouaraCave II .....................·--......................................... 62 Jllgare 5.12 Correlations between tool types and flint Scar-pattern analysis of a Naviform core types at Tell Damishliyya ................................ 123 from DouaraCave II (Nishiaki 1994a) .........66 Flpre 5.13 Lengths of tools and debitage from Tell Blade types for Douara Cave II ....................... 67 Damishliyya: Coarse-grainedflints .............123 Unretouched blades from Douara Cave II .. 68 Jllgare5.14 Suggested model showing the technological 125 system at Tell Damishliyya .......___ Suggested positions on the core from which various types of blades were probably Jllgare 5.15 Frequency of blank types in the sequence at detached ____ ......................................... 69 Tell Damishliyya .......................... ___ 126 Butt size of blades from Douara Cave II .....69 Jllgare 5.16 Frequency of scar patterns of debitage and Suggested model of the reduction stages for tool blanks in the sequence at Tell Damishthe Douara type Naviform method (Nishiaki liyya ......................................................................... 127 1994a) ....................................................................... 71 Figure 5.17 Frequency of butt types of debitage and tool Angle of skewness of the core tablet and the blanks in the sequence at Tell Damishliyya core platform .......................................................... 72 .................................................................................... 127 10

Flpre 5.18 Frequency of flint types in the sequenceat

Tell Damishliyya ................... ____ 121 Flpre 5.19 Flint artifacts from Tell Assouad (the University of Amsterdam's survey collection) .....129 Flpre6.1 Map of Tell Nebi Mend (PaJT 1983) ...........133 Flpre 6.2 Correlation between flint types and blank types at Tell Nebi Mend ................................... 136 Flpre 6.3 Cores from Tell Nebi Mend ............................ 138 Flpre 6A Core-trimmingpieces and • barnrneritone flalccfrom Tell Nebi Mend ·······--139 Flpre 6.5 Frequency of tbennally altered pieces from Flpre 6.6 Flpre 6.7 Flpre 6.8 Flpre 6.9

Flpre 6.10

Tell Nebi Mend ··················--······· ..········142 Schematic presentationof typology for flint 148 tools from Tell Nebi Mend ............................ Points from Tell Nebi Mend ......................... 1.50 Sickle elements from Tell Nebi Mend ........151 Schematic presentationof sickle elements from Tell Nebi Mend and their possible way 152 ofbafting .................................. ,----······ Burins and burin spalls from Tell Nebi Mend

.................................................................................. 183

Flpre 7.13 Butt types by contexts at Tell KasbkashnkIl

,____

Flpre 7.14 Dorsal scarpatterns by contexts at Tell

KasbkasbokIl ................... ___ .............. 184 Obsidian cores and core-edgepieces from Tell KasbkashokII ......................................... 190 Flame8.2 Histogram showing the width of obsidian blades and tools from Tell Kasbkasholcn ....... 191 Flpre 8.3 Schematicpresentationof typology for obsidian tools from Tell KashkashokIl ·······-·· ..··· 192 Flpre 8.4 Various obsidian tools from Tell Kasbkasbok Figure 8.1

Flpre 8.5

Comer-thinnedblades from Tell Kuhkasbk Il, Pit 9 Fill (Nishiald 1990) ..........................196

Schematicpresentationof the obsidian sideblow blade-flalcc(L Braidwood 1961) ..... 197 Flpre 8.7 Terminologyfor the side-blowblade-flalcc Flpre U

Flpre 6.11 Other retouched tools from Tell Nebi Mend .................................................................................... 154

Flpre 6.12 Suggested model showing the technological system at Tell Nebi Mend _____ 159

...............................................................................200

Flpre 8.8

Map of Tell KasbkashnlcIl (Matsutani 1991) ____

........................................... 184

n ............................................................ 19S

------········153

Flpre7.1

Flpre 7.10 Suggestedmodel showing the technological 112 system at Tell KasbkasbolcIl ........................ Flpre 7.11 Lengths of debitage and tools from Neolithic 183 contexts at Tell Kubkasbnlc Il ...................... Flpre 7.12 Con! types by contexts at Tell KasbkasbnkIl

,•.........163

Flpre 8.9

Comparison of the replicated and the archaeological materials ............................................. 202 Suggestedsequence producing "pseudo-retouch" on a side-blow blade-flalcc............ 203

Plan and section of Pit 9 of Tell Kasbkasbok (Matsutani 1991) ........,______ 163

Flpre 8.10 Width of side-blowblade-flalccsfrom Tell

Flpre 7.3 Cores, core trimming pieces, anda barnrner165 stone from Tell Kasblcasbokn ___

Flpre 8.11 Height of side-blow blade-flalccsfrom Tell

Flpre 7.2

n

KasbkasbokIl ----····-·········-·-·

203 KashkasbokIl ......................................................

Schematic presentationof typology for flint 172 tools from Tell Kasbbsbnk Il ....................... l'lpre 7.5 Points from Tell KashkasbokII .................... 174 Flpre 7.6 Width and thickness of points from Tell KasbkasbokIl .............................. ____ 174 l'lpre 7.7 Sickle elements from Tell KasbkasbokII Flpre 7.4

Flpre 8.12 Dorsal ridge number of side-blowblade204 flalccsfrom Tell Kasbkasb()kII .....................

Flpre 8.13 Suggestedmodel showing the obsidian technological system at Tell KashkasbokIl .....206 Figure 8.14 Comer-thinnedblades from Neolithic sites on the Taurus footbills ............................................ 208

.................................................................................... 175

Flpre 7.8

l'lpre 7.9

..........203

Reconstructionof a sickle at Tell Hassuna (Level ill) by Lloyd and Safar (1945) ........11s Retouched tools from Tell KasbkasbokII ---·····················---·········178

showingNeolithic sites with cornerthinnedblades -------···· 209 Flpre 8.16 Side-blowblade-flalccsfrom from Neolithic sites on the Taurus foothills ____ 210

Figure 8.15 Map

LIST OF TABLES Table3.1 Tallle3.2 Tallle4.1 Table4.2 Table4.3 Table4.4 Table4.5

Attributes examined for cores .......................... 39 Table4.6 Attributes examined for flakes and blades .. 39 Flint artifacts from DouaraCave II ................ ss Tallle4.7 Cores from Douara Cave Il ............................... 59 Crested pieces from Douara Cave II ............. 62 Table4.8 Technologicalattributes of blades from Table4.9 64 Douara Cave Il ....................................................... Core tablets from Douara Cave n .................. 12 11

Estimatedflaking modes of debitage from 74 Douara Cave Il (%) .............................................. Correlationbetween tool types and blank types at

Douara Cave Il ······································ 76

Proximal-scarred pieces from Douara Cave II

···································---··························

Flint artifacts from Tell Abu Hureyra,Trench D(Level I) ·······················--·······················84

Table 4.10 Cores from Tell Abu Hureyra. Trench D ...................................84 (Level 1) ..............___ Table 4.11 Crested pieces from Tell Abu Hureyra. Trench D (Level 1) ...............................................86 Table 4.12 Crested pieces from Jericho, Aml M ............ 86 Table 4.13 Core tablets from Tell Abu Hureyra. Trench

D (Level 1) -------····· Table 4.14 Core tablets from Jericho,Ama M __

TableU

Flint and obsidian artifacts from Tell Nebi Mend ___________ 135

Table 6.2

Correlation betweenflint types and blank

Table 6.3

types at Tell Nebi Mend ··---··············· 135 Technological attributes of debitage and tool blanks from Tell Nebi Mend: F"'me-grained

Table6.4 Table6.5

flints ............................ ·----······· ..·········140 Cores from Tell Nebi Mend ...................·-·-· 143 Technological attributes of debitage and tool blanks from Tell Nebi Mend: Coarse-grained

.. 86 86

Table 4.15 Technological attributes of complete blades from Tell Abu Hureyra. Trench D (Level 1) ............·-·······-···· .............._ ...................................... 89

Table 4.16

Technological attributes of complete blades from Jericho, Aml M .................... ___ 90

Table6.6

Table 5.1

Flint and obsidian artifacts from Tell Damisbliyya ................... ______

Table 7.1

Table5.2

98

Table 7.2

Flint types by S1ratUm at Tell Damisbliyya ............................................ _ ...................................... 98

Table 5.3 Table 5.4 Table 5.5

Table 5.6

Table 5.7

Table 5.8

Table59

Correlation betweenflint types and blank types at Tell Damisbliyya _____

Kashkasholt II, Pit 9 Fill ·-···-..·····-···-·--·· 166 Technological attributes of debitage and tool blanks from Tell KadJkasbolc II, Pit 9 Fill

Table 7.4

Tools from Tell KasbkashokII (Nishiaki

Table7.5

Correlation betweenblank types and tool types at Tell Kashkashnlc ll: Neolithic conteJtts .......- .................................................... ·--·-·· 179 Correlation betweenblank types and tool types at Tell Kashkashnlc ll: OChercontexts (a selected sample) ............-. ·--..·· 179 Platformtypes of cores from Tell Kashkashok II: Layer 4, Pit D8, and Layer 3

..·-·-···· ..··-·---·

types at Tell Damisbliyya ................................ 103 Technological attributes of debitage and tool blanks from Tell Damisbliyya: Fme-grained flints ....- .................................................................. 104 Technological attributes of cores from Tell Damisbliyya ..................................... _ ..................105 Technological attributes of debitage and tool blanks from Tell Damisbliyya: Coarse..... ;....A flints &'......... -------·················· 107 Tools from Tell Damisbliyya ____ 111

Table 5.10 Correlation betweenflint types and tool types at Tell Damisbliyya ................... ____ 113 Table 5.11

Table 7.3 99

Cores from Tell Damisbliyya ····-···-··· ..······ 100 Correlation betweencon, types and flint

Table 7.6

Table 7.7

Table 7.8

Correlation betweenblank types and tool

Table8.1

Sickle elements from Tell Damisbliyya ..... 115 Retouch spalls from Tell Damisbliyya ....... 119 Truncation types of retouched blades from Tell Damisbliyya ___ .............................. 122

----····--··

1991)----···········

types at Tell Damisbliyya ................................ 114

Table 5.ll Table 5.13 Table 5.14

flints-----·---.. 144 Correlation betweenflint types and tool types at Tell Nebi Mend ............................................... 149 Flint and obsidian artifacts from Tell Kashkashnlt II (Nishiaki 1991) ...................... 164 Technological attributes of cores from Tell

Table 8.2

Table8.3

167

.................................... 173

-------------180

Technological attributes of debitage and tool blanks from Tell Kashkasholc II, Layer 3 .. 181 Obsidian artifacts from Tell KasbkashokII (Nishiaki 1991) .................................................... 189 Obsidian tools from Tell Kuhkasht>k (Nishiaki 1991) ______ ....193

n

Attributes of side-blow blade-flakes from

LIST OF ABBREVIATIONS USED IN TABLES Asymm. : Asymmetric Bi. : Bicfirectional

Cross. Cx. Dihed. Expand.

Fa. Facet. IDB. Max.

Mecl. Min.

: Median : Minimum

:Crossed

N.

:Number

: Cortex : Dihedral :Expanding : Faceted :Faceted : Imitating a Dihedral Burin Butt :Maximum

Pl. Proxim. Pt.cx.B. Pt.cx.F. Rectan. Ret. S.D.

: Plain : Proximal : Pan-cortex blade : Pan-cortex flake : Rectangle : Retouched : Standard deviation

12

Side b. Str. Surf. Symm. Thinn.

Triangl. Uni. Uniden.

: Side-blow : Stratum : Surface :Symmetric :Thinned : Triangle : Unidirectional : Unidentified

CHAPTER!

INTRODUCTION

on the development of farming / herding communities, and the development of scientific analytical techniques. Traditional, often descriptive, studies of the material culture have become replacedby more varied approaches in archaeology. These new trends are accelerated by new discoveries from the increasing number of excavations mainly due to rescue work. An immense body of data on various aspects is now in band, which have documented profound changes that occurred in human life-styles in the Neolithic period. These trends are welcome since they partly reflect that the cultures of the Levantine Neolithic have been studied in broader perspectives than ever before. On the other band, the increasing information from other disciplines and more varied view-points recently held by the archaeologists appear to have created, in tum, a need for the material cultures be studied in a more detailed and comprehensive way.

1.1 AIMS This study analy1.CSlithic assemblages from several PrePottery to Pottery Neolithic sites in Syria. so as to document full details of the litbic manufacturing activities of early farming communities. Through those analyses, an attempt is made to interpret the technological change in the context of the overall cultunl changes in this period. Four flaked stone collections obtained from more or less

contemporaneous sites in various geographic areas of Syria arc intensively studied. These are from Douara Cave II in the Palmyra basin, Tell Damishliyya on the Balilch, Tell Nebi Mend in the Orontes valley, and Tell Kasbkashok II in the Upper Khabur basin (Fig.1.1). In order to make a comparison of their characteristics in a wider geographical context, numerous lithic collections from other areas of the Levant were also examined. These collections include those from sites in Turkey, Israel, Iraq as well as in Syria All the four sites intensively studied here belong to a rather limited period spanning for several centuries around 8000 BP, that is a transitional period betweenthe Pre-Pottery (PPNB) and the Pottery Neolithic. Preliminary assessment of cultural change between these periods is attempted from a view point of lithic analysis.

This study focuses on one aspect of material culture; lithic tools and their manufacturing technologies. It analy1.CS details of the use of stone material by the Neolithic people, from its procurement and reduction, to use and discard, so that the Neolithic lithic technology can be understood in relation to other aspects of human behavior.

In the following sections of this chapter, a brief account of the environmental and cultural backgrounds of Neolithic Syria is given. Chapter 2 examines previous lithic studies of the Levantinc Neolithic from a methodological view-point. Chapter 3 presents the analytical method employed in this study. Detailed analyses of the lithic assemblages are then followed in chronological order from the earliest material. Chapter 4 is devoted to the Pre-

Recent studies of the Levantine Neolithic are characterized by a couple of new trends, namely interdisciplinary approaches (including archaeo-zoological, -botanical, geographical and ecological studies), modeling theories 13

Chapurl

2e 3

•

Mediterranean Sea

37

•

•38

•36

Q Flpre LI Map showing the Near EasternNeolithic sites mentionedin the text. I: Cllall1 HUyflk,2: Asikli HOyflk,3: Can Hassan III, 4: Sakce GOzil, 5: Cafer HOyflk,6: 86y Tepe. 7: nmight have functioned to regulate social problems. Moreover, this kind of huge settlements of the PPNB, unrivaled by other contemporary settlements, could have been a regional center.

1.3.5 Society

Objects made of exotic or locally unavailable materials became even more popular, which suggests the greater goods flow along the exchange network in the PPNB than in earlier periods. The materials obtained through exchange could include obsidian, marine shell, turquoise, greenstone, basalt, bitumen (cf. Roaf 1990: 34-5), and 1986; even flint (Mellaart 1975), and ceramic (Le Miere Le Miere and Picon 1987). It is probable that development of exchange networks was partly responsible for the rapid diffusion of inventions like the floor plastering and the ceramic technologies.

Ceremonial aspects also show a marked development in the PPNB period. At the PPNB settlements of Ain Ghazal and Jericho, for instance, caches of magnificent human statues made of lime plaster over the core of reeds and twine have been discovered (Simmons and Rollefson 1984; Rollefson et al. 1992; Garstang 1935). Plastered skulls and masks, also considered relating to the ceremonial use, have been widely discovered in the southern Levant such as at Ain Ghazal, Beisamun (Lechevallier 978), and Nabar Hemar (Bar-Y osef 1985). Similar materials have been known at Ramad in Syria (de Contenson 1967). Another evidence of the development of ceremony is the presence of specialized buildings within the settlement. Possible shrines have been reported from Ain Ghazal (Rollefson 1998a). In the northern Syria, the recent discovery of floor paintings of female figurines in

Redman (1978: 205) states that the early farming communities of the PPNB and the early Pottery Neolithic such at Jericho and Jarmo were basically an egalitarian society. Analysis of mortuary practices of this period supports this view. There is little difference between burials of different sexes and ages (Fiedel 1979). However, the difference tends to have become clearer in the later PPNB (Rollefson 1989: 171). In addition there are indications of a "complex society", if not a ranked society existed in the communities of the period. The elaborated craft workmanship represented by various ornaments and flint artifacts, for example, would suggest that some form of part-time craft 20

the Middle PPNB context of Tell Halula may suggest the

cultural development in the late PPNB period of Syria. The cultural growth that occurred in that period represents a continuity from the earlier periods, but shows a rather rapid development in diversified aspects ranging from subsistence,settlement pattern, technology and perhaps to social organization. The cultural aspects of the following early Pottery Neolithic, which certainly share many elements with those of the late PPNB,are also consideffil a continuous one. At the same time, however, differences are evident particularly in the socio-economic terms. Changes related to ideological and ceremonial aspects are also remarkable. New survival strategies under the changing climatic conditions, as indicated by the movement of settlements into the previously less occupied environmental zones and the transformation of subsistence to that with more emphasis on domesticated food resoun:es,we,e increasingly employed in this period. As a whole, the developments in the early Pottery Neolithic, some of which began already in the late PPNB period, are regarded those representing a transition toward a new society and culture that characterize the rest of the period of the Syrian Neolithic. Moore (1978, 82) designated the Neolithic after the outset of the Pottery Neolithic as "Advanced Neolithic", while the previous one as "Archaic Neolithic". The late PPNB and the early Pottery Neolithic (ea. 8600 - 7500 BP), to which all the four sites intensively analyzed in this study belong, represent an importantperiod of transition leading to the new fanning society.

use of that building for ceremonial practice (Molist 1998b). Further in the north, the series of remarkable carved stone objectsand statues at CxSbe1di Tepe, Turkey, indicates a unique development in the Anatolian plateau (Schmidt 1997). Comparable objects and features are rarely known in the Pottery Neolithic period, perbaps showing important ideological changes. Ceremonial or ideological elements characteristic of the Pottery Neolithic are"mothergoddess"or clay figurines offemales (J. Cauvin 1972b, 1978). A diagnostic type representinga seated female, thought to be a

symbol of fertility and reproduction for agriculturists, in the Levant during a came to be widely manufactured period of the early Pottery Neolithic (Hori 1980; J. Cauvin 1994). Mortuary customs characterizing the PPNB period include the habit of skull cult and the custom of burying the dead under the floor (Molist 1998a). However, they were not practiced in the Pottery Neolithic (Gopher and Orrelle 1995). Most of the dead except infants we,e buried outsidethe settlement at a communal

,..,,...,ezy.

1.3.6 Summary

The brief survey of the literature shows a considerable

21

CHAPTER2

LITIIlC STUDIES OF THE LEVANTINE NEOLITIIlC: A MEIBODOLOGICAL REVIEW

specific analysis of artifacts (Deetz 1967: 49). Typological study of stone tools bas beendeveloped mainly in the field of Palaeolithic research, particularly of Europe. The best known example is that established for Lower and Middle Palaeolithic by Fran~ois Bordes (1950, 1961). Bordes constructed a standard type list consisting of over 60 tool types based on morphological attributes taking into consideration their manufacturing techniques. The type list was intended for systematic classification of each assemblage, and for comparison in a statistical way with the aid of cumulative graph showing the frequency of each tool type.

2.1 INTRODUCTION This chapter presents a review of the literature to examine how flaked stone artifacts have beentreated with and how their analytical possibility has been exploited in previous Levantine Neolithic studies. 1bis review will hardly be exhaustive, however, for most of lithic studies have been published as part of an excavation report, and therefore they are scattered in numerous publications so widely that it is practically difficult to recover all of them. Fmtber the increase of the articles in the 1990s is especially remarkable (Gebel and Kozlowski 1994; Kozlowski and Gebel 1996; also see Neo-Lithics).

This analytical method has been applied to study of later Palaeolithic and Bpi-Palaeolithic industries as well. Examples include type lists for Upper Palaeolithic of France prepared by de Sonneville-Bordes and Perrot (1954, 1955, 1956a, 1956b) and by Laplace (1964), lists for Bpi-Palaeolithic industry of North Africa designed by Tixer (1963) and for the Near East by Bar-Yosef (1970), Hours (1974), Goring-Morris (1987) and Byrd (1989), and Late Palaeolithic industries of Japan devised by Akazawa and bis collaborators (1980). Each tool type of these lists was defined, in general, using the same kind of attributes as used in the Bordean typology, i.e. morphological attributes. Unlike the Bordean list, however, they include many site- or area-specific tool types that can not be expected to occur in assemblages from different industrial contexts, and give such tool types specific names after the sites at which they were first found or described, e.g. Noaille burin (de Sonneville-Bordes and Perrot 1956a) and Columnata point (Tixer 1963). The applicability of these lists are thus inevitably restricted to much smaller chronological and geographical units than Bordes• list

For the sake of convenience, I shall confine the following discussion mainly to the studies using the northern Levantine material. The studies are classified into four categories: typological, technological, functional and contextual studies. Practically, however, some of the studies can not be classified into one of these categories. They are instead overlapped in many cases at least to some extent Accordingly, the same article may be discussed repeatedly in more than one section.

2.2 TYPOLOGICAL STUDY 2.2.1 Type list and type fossil

Typology is a study based on the type concept, which is central to morphological analysis, and it has been a most useful tool to the archaeologists both for description and

22

Bar-Yosef (1981a), small points with notches at the base or lateral edges like Khiam points, Abu Madi points and Helwan points are characteristic to the PPNA periods, and larger points often retouched with pressure technique such as Amuq, Byblos and Jericho points are more commonly found in the PPNB contexts. In addition, some of them show a rather restricted distribution, and thus are to be considered a space marker as well. Jericho points, for instance, are densely found in the southern LevanL More recently, an extensive survey of point types from Neolithic sites in the southern Levant has been presented by Gopher (1985, 1989a, 1989b, 1994).

On the other hand, no standardized type list has been established for the Levantine Neolithic, in spite of the effort to establish by some analysts (Burian and Friedman 1979; Rollefson 1989: 172; Gebel and Kozlowski 1994). As already pointed out by some authors (Noy 197S: 171; Ataman 1988b: 63), it seems to be actually not very useful to establish a single type list, if possible, for the Neolithic of the whole Levant when considering rather strong cultural localintion in this period. Some authors have employed lists constructed for earlier periods but with considerable modification (e.g. Roodenberg 1986: 4). Others have been preparing type lists specifically designed for their own site or a group of sites in a restricted uea showing certain similarity (e.g. Mortensen 1970a, 1982). One result of this procedure is the difficulty in comparing one site with another in a wider geographical context. There appears to be no useful alternative approach (Ataman 1988b: 61), however, in view of the distinct nature of Neolithic stone tools. The absence of this son of a standard list for the Levantine Neolithic obliges one to employ another, more traditional method for comparison. that is the type fossil approach.

Sic/ck ekments The sickle element has been another well studied tool type in the Levantine Neolithic. There is a marted difference in the approach to the analysis of points and sickle elements. The sickle element is defined in terms of 1\mc· tion, while the point is morphologically isolated from other tool classes. Sickle elements are identified if they exhibit so-called sickle gloss that is believed to arise from their use for cutting plants. Hence they are often termed glossed pieces. In the course of recent development of microwear studies, however, it has become clear that pieces without gloss visible with the naked eye may also have been used as sickle elements (Unger-Hamilton 1988). Thus, pieces showing morphological similarities to those with visible gloss are also to be considered as sickle elements, termed "Shape-defined sickle elements" (Copeland n.d.). In any case, the functional trace has been of primary imponance for identification of sickle elements. As a result, they comprise a very heterogeneous group from a morphological point of view, consisting of tools that might have been traditionally classified as, for example, denticulates, truncations, backed blades, and even unretouched flalces.

2.2.2 Flint tools

Points The tool class most frequently used in the type fossil approach by the Levantine Neolithic researchers has been the point or arrowhead. The point represents one of the most developed and ubiquitous tool types found in the Levantine Neolithic contexts, and is therefore considered useful to examine their chronological and geographical distribution (Burian and Friedman 198S; Gopher 198S, 1994). More than ten specifically named point types or "type fossils" have been proposed, including the following: Khiam point (Neuville 1951), Helwan point, Abu Madi point (Bar-Yosef 1981a), Harifpoint (Marks 1973), Amuq point, Byblos point, Jericho point (J. Cauvin 1963), Nemrik point (Kozlowski and Szymczak 1989), Ugarit point (de Contenson 1977-78), Abu Gosh point (M.-C. Cauvin 1974), Bouqras point (Roodenberg 1986) and so on. In addition to these points named after their eponymous sites, a number of often seemingly exhaustive typological classifications have been attempted for points (e.g. M.-C. Cauvin and Stordeur 1978; Mortensen 1970a).

In addition, the morphological heterogeneity may also have been created by the repeated retouching during the tool's use-life. An extensive series of experimental use of modem copies of flint sickles demonstrates the easy exhaustion of use-edges (Anderson 1994). It is quite possible that exhausted sickle elements of the prehistoric times received secondary retouching either on the same working edge or on a different area of the tool blank, which must have helped changes of their original tool

forms. Another notable tendency in sickle element studies is that typological classification often takes into consideration ways of hafting, i.e. shape of the handle, number of sickle elements per handle, and way of insertion of each element

Some of these point types have been proved to be useful as time markers, hence type fossil. Burian and Friedman (1985) and Bar-Yosef (1981a) present a summary of chronological succession of point types. According to

23

Chapter2

1990) suggests, on the other hand, that the facets on those burins were in fact not working edges but scars that produced blanks for drill-bits.

(M.-C. Cauvin 1983; Fujii 1983; Stordeur 1987). Although other tool classes such as points and end-scrapers might also have beenused hafted, their hafting methods have rarely been considered in typological study (cf. Burian and Friedman 19TI-78). The chronological change of the hafting method is interpreted to be relating to development of plant domestication and its harvesting efficiency (M.-C. Cauvin 1983).

2.2.3 Obsidiantools

1nthe above, I have mentioned tools mainly made of flint Obsidian was also utilized as an important raw material for flaked stone tools of the Levantine Neolithic, its proportion varying by assemblage, from a few percent to more than half of the total artifacts. Apart from studies of obsidian-rich industries found in Anatolia and the Zagros mountains (e.g. M.-C. Cauvin and Balkan 198S; Mortensen 1970b), typological study of obsidian artifacts bas been obviously less developed than that of flint (Nishiaki 1993b). This lack of attention to obsidian probably relates to the fact that obsidian is represented by only regular blades with little, if any, retouch at most of the sites, apparently showing little morphological variability. Thus, most of the site reports have referred to obsidian artifacts simply as blades or retouched blades, paying no serious attention to their morphological characteristics.

Many morphological types have beendefined to describe sickle elements found at each site. Unlike points, however, no specific nomenclature such as the Byblos point bas been given to sickle element types. Chronological and spatial significance of each type bas not been examined at least as in the same detail as for the point types. This is perhaps partly because of the morphological heterogeneity of sickle elements. M.-C. Cauvin (1983), however, suggests that morphological patterning in sickle elements does emerge if·carefully examined, and some of them could have chronological and spatial significance. She recognizes five groups of sickles in the Near Eastern Neolithic: unretoucbed pieces, pieces with retouched ends, truncated pieces, backed pieces, and denticulates. She suggests that pieces with retouched ends appeared in the earlier chronological contexts, while denticulates were used in the later ones, 9th millennium BP afterwards. She also remarks that sickle elements with denticulate edges have been densely found in the Levant, while sparsely in the Mesopotamia and the Zagros.

1n addition

to tool types common in flint, however, a cou-

ple of distinct tool types unique to obsidian have been revealed by recent close i-.xaminatinns. "Side-blow bladetlakes" (L. Braidwood 1961), "

c:!> ,

·-

····£°... ·. •.. .

.

,'

.

(

-

f

h 3

2

.II/

.'' ·r•.

M

R

t Proximllscaning

5

6

Flpre 4.14 Suggested model of the reductionstages for the Douaratype Naviformmethod (Nisbiaki 1994a) cussion point from the core edge is markedly different between Type Land Types MIR blades (Fig.4.13). The percussion point for detachment of Type L blades was chosen well away from the core edge, those for Type M and R blades were selected very closely to the edge. The deep percussion point for Type L blades was a major sourcefor giving the butt an appearancesimilar to a burin (IDB butt). Furtbcrmorc, the percussion point for production of Type L blades was quite consistently chosen at the right end of the butt to be produced (Fig.4.11: 1-3; Type 2, Table 4.4.5), while it was more frequently chosen at the middle of the butt for detachment of other types of blades. This unusual practice at Douara Cave II resulted in frequent production of a twisted profile of those blades (Table 4.4.6).

blades were detached most likely from the left part of the core, then Type M blades from the middle of the flaking surface, and Type R blades were from the right side of the core. The acute platform angle of the core and biasedexploitation of the core surface were responsible for creating this distinct pattern in the blade morphology. This interpretationis supported well by a couple of morphological features. First, the position of cortex on blades shows a strong right-left bias (Table 4.4.1); Cortex on Type L bladesis exclusively situated along the left edge, and that of Type R blades is along the right, which clearly reflect the position of cortex on the core. Another distinct feature is the position of crossed scars on the dorsal surface with the biface pattern (e.g. Fig.4.11: 3). Such scars arc always found left on Type L blades, while right on Type R blade (e.g. Fig.4.11: 7, 8). Type M blades show, not surprisingly, rather mixed features of Type L and R blades (Table 4.4.2).

The above analyses suggest that Type L blades can better be isolated from the other two types of blades. What was the intention of producing such a distinct type of blade? The general size (length, width, and thickness) of the blades show little difference by type (Table 4.4.10). Notable differences arc, however, found in the transversal section and the relative thickness of the butt. As already noted, Type L blades retain thicker butts than others. Further, Type L blades exhibit a much higher frequency of a

Interestingly enough, these three types of blades were produced not only from different parts of a core, but also with different technological strategies. Firstly, Type L blades were almost always detached after careful abrasion on the platform edge (Table 4.4.4), while Type Mand R blades were often unabraded. Secondly, the depth of per71

Chapter4

2

Figure 4.15 Angle of skewness of the core tablet and the core platform. 1: Core tablet, 2: Core platform.

Table4.5 Core tablets from Douara Cave II 1 Type (cf. Fig. 3.3) No %

2 Measurements(mm) Length Width

Thickness

uw

wrr Angleof skewness(Type 2 only)

Type I 24

Type2 35

36.36

Uniden.

7

Tollll 66

53.03

10.61

100

N

Mean

S.D.

Min.

Max.

Median

31 66 66 31 66 20

52.548 29.121 6.715 1.855 4.834 43.05

17.968 8.140 3.237 0.672 1.987 14.479

20

85 58 23.9 3.182 12.80 69

so

72

11

2.5 0.641 1.667 24

30 6.1 1.963 4.522 38.5

Douara Cave II

trapezoidal section than other blades do (Table 4.4. 7). These morphological differences suggest that Type L blades were detachedin view to obtaining blades with a robust butt, trapezoidal transversal section and more or less parallel lateral edges; these might well have been suitable as blanks for certain tool types.

blades (18.6 %; Table 4.4.9) , and '215(43.2 %; 67/155) of the platforms of Naviform cores show binge-terminated ends. In sum, serious accidents caused by plunging and binge-termination arc observed on more than half of the core platforms (89/155 or 57.4 %), which partly explains the core abandonment

The consistent production of Type L blades was apparently encouraged by the common use of twisted Naviform corcs as described before. A careful t!xamination of the Douaracorcs reveals that their platforms arc not set up at right angles to the longitudinal axis of the core. The angle a, as shown in Fig.4.14, is always larger, and the angle b is smaller than 90 degrees on Douara cores. Keeping those angles at about 90 degrees should be essential to produce blade regularly along the long axis of the core (Nisbiaki 1994a). The cores from Douara, nevertheless, rarely show the "correct" angle, but maintainthe relationship of a> 90•> b so that one side (i.e. b's) side of the core are more easily exploited. We cannot determine whether this strategy was conducted on purpose or in error, but the remarkable consistency strongly argues that it wasan important habit of the Douara knappers, and that one possible advantage of the use of this technology was pemaps to allow consistent production of Type L blades.

Core lfl4inurumce(Suzukiand Akavzwa 's Stage 6) Suzuki and Akazawa's "platform rejuvenating flakes" or core tablets in this study arc the best indicators of this stage of reduction. There arc 66 examples of core tablets in the collection from Douara Cave (Fig.4.8: 6-8).

n

Although such pieces arc undoubtedly derived from the platform of cores, Suzuki and Akazawa (1971: 122) pointed out an interesting discrepancy between core tablets and core platforms in their surface collection. They compared angles formed by the butt and the longitudinal axis of the tablet, with their supposed counterparts on the core (Fig.4.15), so as to examine the nature of the "biased distribution of blade detachment scars". They found the angle on the core tablet is closer to 90 degrees on average (ea. 75 degrees), while a narrower angle is represented on the core platform (ea. 60 degrees). This result led them to suggest that core tablets were mostly removed in earlier stages of reduction before the cores wereheavily exploited.

Finally, one more remarkable feature should be mentioned here on blades Douara Cave ll. Upsilon blades defined by Ataman and Calley (1988) at contemporary TurkishPPNB sites arc extremely rare at Douara Cave These blades, which show a symmetric Y-shaped dorsal ridge patlCm, arc found on a few specimens only (8/451 or 1.8 %). In fact, blades with a symmetric dorsal ridge pattern themselves are rare at Douara (Table 4.4.8); instead, a distal tip curved to right is very common here (e.g. Fig.4: 11: 1-3; 197/451 or43.7 %). There is no doubt that this anomaly in the distal tip shape is related to the cwious biased exploitation of the core surface, or the use of twisted cores.

n.

Careful examination of the DouaraII collection revealed at least two types of core tablets (see Fig.3.3 in Chapter 3; Table 45.1), which Suzuki and Akazawa (1971) did not distinguish. Comparison of the angles should be made between core platforms and Type 2 tablets (Fig.4.8: 7, 8) alone, since Type I tablets (Fig.4.8: 6) quite possibly include pieces produced in the course of initial core preparation. The results at Douara Cave II are presented in Tables 4.2.3 and 4.5.2, which show that the angles on core tablets are significantly smaller (ea. 43 degrees) than those on core tablets (ea. 56 degrees), suggesting that the core tablets were removed from highly skewed (i.e. exhausted) cores.

Knappingaccuknts (Suzuki and Akavzwa's Stage 5) Knapping accidents actually do not represent a "stage" of the reduction sequence, but rather could have occurred throughout Stages 1-4. Suzuki and Akazawa's ''fisbbooksbaped flakes" correspond to plunging blades in this study (Fig.4.8: 5). Traces of removals of plunging blades arc recognizable on cores as well. Navifonn cores of Type 5 arc examples showing such traces (Fig.4.7: 3, 4). About one seventh of the platforms of Navifonn cores (221155 or 14.2 %) show this sort of accident

Heat treatment

The possibility of beat treatment of flint was not examined, since flint pieces in the present collection show very poor surface conditions.

Flaking mode Seven bammerstones were recovered from Douara Cave Il. All of them arc re-used cores (7/121 or 5.8 %), whose

Another notable accident is binge-fracture. About 1/5 of

73

Chapter4

Table 4.6 Estimated flaking modes of debitage from Douara Cave D (%)

Caregories

CJeStedpieces (N=73) Care tablets (N=66) Plunging pieces (N=9) Core-edge pieces (N=27) Cortex flakes (N=33) Part-cortex flakes (N= 145) Flakes(N=346) Blades: Type L (N=253) Blades: Type M (N=l7l) Blades: Type R (N=l 11) Blade: D-sbaped Type (N=70) Tolal (N= 1304)

Soft 67.12 72.73 77.78 48.15 2727

Hard Uniden.

Tolal

28.77

8.19 11.71 0.00

100 100 100 100 100 100 100 100 100 100 100

10.58

100

4.11 10.61

22.22

25.93 54.55 55.86 33.10 65.32 21.68 97.23 0.79 87.13 4.68 86.49 l.80 100.00 0.00 76.23

13.19

16.67 0.00 25.93 18.18 11.03 13.01

l.98

ther processing. The present analysis of the Naviform method at Douara Cave II confirms Suzuki and Akazawa's reconstruction of the reduction sequencein general terms. Both of the two important traits noted by them on materials from the Palmyra basin were recognizedat Douara Cave II. These traits are 1) biased distribution of blade detachment scars on the cores, and 2) frequent production of debitage with an IDB butt

In addition to these, the present study revealed a couple of previously unrecognized features as well: 1) manufacture of a core-preform with an asymmetric shape; 2) core preparation with removal of D-shaped blades; 3) selection of a percussion point at the right end of the butt to be detached; 4) common production of blades with a tip curved to the right as well as a twisted profile.

original core types are:five Naviform cores (Figs. 4.6: 6; 7: 1, 4), and two change-of-orientation cores. They are badly battered at the end(s) or along lateral margins. Cores with accidents that could not be corrected such as plunging (Fig.4.7: 4), seem to have been reused as bammerstones.

4.4.3 Tool manufacturing

Flaking mode can be evaluated by means of examining debitage and tool blanks (Ohnuma and Bergman 1982). The result is shown in Table 4.6, which suggest the very frequent use of soft(er) b&mmer'!Iat Douara Cave II, while showing cortex-flakes were mostly detached with the bard(er) hammer mode (cf. Newcomer 1971). This result may be contradicted by the presence of relatively many hamOJfflltones in the flint collection. H the identification of flaking mode on dehitage is reasonably secure,a possible explanation would be that the hammerstones were kept at the cave with a view to use them at factory sites, probably, within the Douara basin. As Akazawa (1979b) states, flint artifacts from the factory site of Locality 35 include many cortex-flakes perhaps produced with the hard hammer mode during the initial stage of preform

About 340 flint artifacts were identified as retouched "tools" in the present collection. Two major difficulties were encountered in dealing with ''tools" at Douara.One is involved with the physical condition of the material; many of the flint artifacts from Douara exhibit extensive post-depositional retouch or natural retouch, which are often difficult to be distinguished from those by intentional modification (cf. Miller 1982). Under these circumstances, identification of certain miscellaneous tool classes such as denticulates and notches would be suspect, and was therefore avoided in this study. The second problem is concerned with a curious "burin" technique diagnostic of the PPNB in Syria that has been noted by various authors (e.g. Moss 1983; Roodenberg 1986; Fujii 1986a; Copeland n.d.). Many of the blades and flakes from Douara Cave II exhibit a "pseudo-burin facet" at the ventral surface of their bulbar end (Fig.4.19). This type of facet is situated almost parallel to the longimdinal axis of the tool blanks. Unlike the ordinary burin facet, however, it slants onto their ventral surface, resembling the "burin plan" or the obsidian "comer-thinned blade" (Nishiaki 1990). This type of retouch, or "PPNB proximal scars" of Copeland (n.d.) and "Palmyran retouch" of Fujii ( 1986a) was first reported, as far as I know, by de Contenson and van Llere (1966b: 183) on blade tools from Tell Bouqras. Pieces with the same retouch type have since been reported from many other PPNB sites particularly in inland Syria, for instance, from Palmyra (Suzuki and Kobori 1970), Tell Assouad (M.-C.

making. Suzuki and Akazawa (1971: 120) write that "parallelsided flakes" or our Type L blades were produced by pressure flaking. 1be observed attributes suggest that this is not necessarily to be the case at least at Douara Cave II. Similar attributes can also be produced by direct or indirect percussion with a soft hammer.

Summary Flint was perhaps collected and / or mined in the nearby Douara basin. It is likely that the very initial stages of reduction such as test flaking and roughout / preform manufacture were mainly performed at the flint source; the bifacial preforms were brought into the cave for fur74

Do110TtJCavell

,l,

~

"l

[re G] {rn 2

a

0

..,

t

4

3

2

!

POINTS

~ G>

' 2

"'

d--®

~~

....i

i

f)-fJ~ t

6

5

8

7

3

SCRAPERS

BORER

l

BURINS

OBLIQUE TRUNCATION

SPLINTERED PIECES

RODS

2

2

RETOUCHED BLADES

RETOUCHED FLAKES

Figure 4.16 Schematic presentation of typology for flint tools from Douara Cave II

75

Chapter4

Table 4.7 Correlation between tool types andblanktypes Categories

Core-trimmingpieces CT PL

CR* Poinls Type I Typc2

0 0 0

Bwins

4

I

1 1 1 0 1 0 0 0 0 0 0 0 0 0 0 0 0 11 0 11 0 0 0

0 0 0 0 0 1 0 0 0 3 3 0 0 0 0 0 0 9 0 9 0 0 0

Type 1

Typc2 Typc3 Typc4 Types

Typc6 Typc7 Types

Others Scrapers Typcl Typc2 Typc3 Borer

Truncation Rods Splintcffll Rei.blades Type 1 Typc2 Rei.flakes Type 1 Typc2 Total

... Dcbitagc ...

0 0 0

15 4.40

3.81

163 6.01

2.43

13 66

0 0 0 2 2 0 0 0 0 0 0 0 0 2 I

0 1 0 0 0 0 0 0 0 0 0 0

at

DouaraCave II

Flakes

Blades L M

CE

ex

PCX

FL

DS

0 0 0 1 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 0 2

0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 0 0 0 0 0 0 0 0 0 0 0 0

0 0 0 3 2 0 0 0 1 0 0 0 0 5 1 1 3 0 0 0 0 0 0 0 7 1

0 0 0 9 0 4 1 0 0 0 0 1 3

0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

1 1 0 0 0 1 0 1 59 1 58 0 0 0

4

2 0 2 0 0 0 0 0 0 0 23

8 15

6

1 1 0 15 3 1 3 0 3 3 0 0 2

4 1.17

3 0.88

0.29

15 4.40

10.56

0 0.00

78 22.87

14 0.52

27 0.99

62 2.28

258 9.51

814 29.99

74 2.73

253 9.32

I

36

0 0 0 7 3 0 0 0 2 2 0 0 0 2 2 0 0 1 0 0 0 21 2 19 0 0 0

Total R

0 0 0

3 0 3 14 4 0 3 1 3 2 1 0 0 3 1 2 0 0 0 2 1 73 3 70 0 0 0

9.09

49 1437

28.15

171 6.30

4.09

31

0 0 0 9 2 1 0 1 1

Unid

4

0 0 0 3 3 0 0 0 0 1 0 36

0 36

Ill

96

701 25.83

No 4

... 1.17

1 3 65

19.06

17 7 9 2 11 12 1 1 5

24 15 3 6 1 1 3 2 209 6 203 32 9

7.04

0.29 0.29 0.88 0.59 61.29 938

23

341 100.00 100 2714 100

*CR: Cmred pieces, CT: Core tablets, PL: Plunging pieces, CE: Corc-«tge pieces, CX: Cortex flakes, PCX: Pait-cortex flakes, FL: Fakes, DS: D-lhlped blades, L: Type L blades, M: Type M blades, R: Type R blades, Unid.: Unidentified blades

Cauvin 1972) and Tell Abu Hureyra (Moore 197S).

A schematic typology for tools from Douara Cave Il is presented in Fig.4.16.

Since close examination of the examples from Douara Il shows that presence of this retouch scar type cross-cuts conventional tool types (see below), it is inadvisable to regard as representative of an independent tool type. Nevertheless, blades and flakes with "proximal scars" also exist without a conventional piece-modifying retouch along edges. This rather puzzling situation certainly needs further clarification. whose detailed analysis will be therefore attempted later.

Points Four points were recovered at Douara Cave II all in a broken fonn. Three of them are proximal fragments, and the last example is a medial fragment (Fig.4.17: 1-4). These points can be generally described as Byblos points; however, they may be subdivided as follows. Type I: Points with a retouch-covered tang ( 1 piece; Fig.4.17: 1). The tang is formed by bifacial invasive flake scars. The transversal section of the tang is semi-circular. Type 2: Points with an abruptly retouched tang (3 pieces; Fig.4.17: 2-4). The tang is made by either direct (Fig.4.17: 2, 3) or inverse abrupt retouch (Fig.4.17: 4). The rest of the body is only marginally retouched.

Tools thus identified at Douara correspond to 9.S % of the total flint artifacts (341/3S84). The type composition is rather simple, consisting of points (1.2 %), burins (19.l % ), end-scrapers (7 .0 %). a borer (0.3 %), a truncation (0.3 %), rods (0.9 %), splintered pieces (0.6 %), and retouched blades (61.3 %) and retouched flakes (9.4 %). The last two categories are actually dominated by proximal-scarred pieces.

The blanks used for points are all blades, including at least one Type L blade (Fig.4.17: 2). Three of them exhib76

Douara CavttII

3

8

7

6

5

4

5cm

~ I

&-

9

I

r

12

13

0

~-

',,,/

\

~

~

14

16

Figure4.17 Points and burins from Douara Cave II. 1-4: Points (I: Type I. 2-4: Type 2). 5-11, 14-16: Burins (5-6: Type I. 7-8: Type 2, 9: Type 5, 10, 14: Type 6, 11: Type 4, 15: Type 7. 16: Type 8), 12-13: Burin spalls.

77

Chapter4

3

.~~

~-i

~~

6

9

10

5cm

0

12

13 Figure 4.18 End-scrapers and other retouched tools from Douara Cave II. 1-8: End-scrapers (1-5: Type I, 6-7: Type 2, 8: Type 3), 9: Borer, I0: Splintered piece, 11: Oblique truncation, 12-13: Rods.

78

Douara CaveII

3

s

w 1

7

5cm

0

11

12

13

14

Figure 4.19 Proximal-scarred pieces and their retouch spalls from Douara Cave 11. 1-9: Proximal-scarred pieces, 10: Schematic presentation of proximal-scarring retouch, 11-14: Spalls of proximal-scarring retouch.

79

Chapter4

and a scar of D-shaped blade removal. Its burin facets themselves correspond to Type 6, but this type is distinguished by the tang made by steep direct retouch. Similar pieces have been abundantly recovered from the PPNB levels of Tell Abu Hureyra (Moss 1983). Type 8: A multi-burinated piece (I piece; Fig.4.17: 16). It exhibits several "burin facets" covering one or both surfaces of the blank. It resembles a splintered piece but exhibits transversal facets as well as longitudinal and oblique ones.

it a proximal scar at their left venttal end. Two of the points show impact fractures at the tip (Fig.4.17: 2, 3), suggesting that the points were actually used as projectiles (Bergman and Newcomer 1983).

Burins Burins represent one of the most common tools at Douara Cave II. Apart from the problematic proximal-scarred pieces, they constitute more than half of the total flint tools (65/100 or 65.0 %). Although these pieces exhibit at least one "burin facet" at part of their blank, as we discussed in Chapter 3, they also include pieces whose assignment as "burins" may be questionable at least in a functional sense.

The blank forms of the burins are shown in Table 4.7 by type. Most of the burins were made on blades, but a few flakes were also utilized as blanks for, in particular, Type 2 pieces (truncation burins). There are six retouch spalls that were produced in manufacturing burins. Four of them are from angle burins (Fig.4.17: 12), and the remainings are from transversal pieces (Fig.4.17: 13). These pieces indicate that manufacture of burins was conducted at the site.

Burins are characterized by two important variables (Newcomer 1972): trace of a burin blow (burin facet), and a platform for the blow. Based on observation of these two variables, I developed the following typology for burins from the DouaraCave II (5 pieces were unclassifiable because of bad preservation):

Scrapers

Type 1: Angle burins on an unmodified surface or on a broken/ snapped surface (17 pieces; Fig.4.17: 5, 6). Type 2: Angle burins on a retouched truncation, or truncation burins (7 pieces; Fig.4.17: 7, 8). Concave truncations are rather common. The concave truncation burins exhibit a similar morphology to that of Mejalla burins that have been known from the late 9th to the 8th millennium BP contexts in the inland Levant (Betts 1986; Rollefson et al. 1989: 101). Type 3: Transversally burinatedpieces on an unmodified lateral edge (9 pieces). The burin facet could either be perpendicular to or oblique to the longitudinal axis of the blank. Type 4: Transversally burinatedpieces on a retouched lateral edge (2 pieces; Fig.4.17: II). When the facet slants onto the dorsal or the ventral surface of the blank, the piece appears similar to a "chamfered piece" or tranchet. Like examples found in Palaeolithic contexts (Newcomer 1966-69; Goring-Morris and Rosen 1989; Nishiaki and Copeland 1992), chamfered pieces are given a status as a tool type in the current Levantine Neolithic typology (e.g. Crowfoot-Payne 1983: 693-5; Gopher 1989c: 74). Type 5: Burins whose facets intersect perpendicularly at a comer of the blank (I I pieces; Fig.4.17: 9). Type 6: Dihedral burins (12 pieces; Fig.4.17: 10, 14). The burin facets intersect at one end of the blank. Type 7: A tanged burin. There is only one example of this type in the present collection (Fig.4.17: 15). This piece is made perhaps on a Type L blade with a proximal scar,

There are three general forms of end-scrapers at Douara Cave II. Type l: "Typical" end-scrapers (15 pieces). A round working edge is made at either the distal (Fig.4.18: 3, 5) or proximal end (Fig.4.18: 2, 4) of the blank. One example shows working edges at both ends (Fig.4.18: 1). Type 2: Composite tools of end-scrapers / burins (3 pieces; Fig.4.18: 6, 7). One end retains a round scraping edge, and the other end shows burin-facets. One of those pieces (Fig.4.18: 7) shows a similar morphology to that of "V-shaped end-scrapers" from the early PPNB levels of Tell Mureybet (M.-C. Cauvin et al. 1987). According to M.-C. Cauvin and others (1987), the burin-facets at the proximal end could serve to facilitate hafting. They mention comparable examples also from later Neolithic contexts such as < Sc~ ~ ·c,. o rl o.,

=5 vJgvn.,

5 t:

!.:.;

i!

g,

-t!

! u

:iuuu ~5.f£.f]88888u:C~ =i--i-,i-,i-,i-,~u:c

~ C

g- u

'f

u

~

=•

Artifacts made of the fine- and coarse-grained flints exhibit a great contrast in many technological aspects including blank morphology and core reduction strategics (sec below). Table 5.3 shows correlation between flint types and blank types. While coarse-grained flints include

-= : ... e! ~ !.c ~:;; g

t:

rl

·c

:in c.,.

CCCCC.ccrl

o o o o o u • ~ 'f u ., ., ., ., ., g

s.

! 99

Chaptt!r5

Table 5.4 Cores from Tell Damishliyya Categories

Tow

Slrala

Str. 5

Str. 6

Str. 7

Pit

0 0 0 0 1 2 0

0 0 0 1 0 0 0 0 0 0 0 0

0 0 0 0 0 1

0 2 0 2 0

3 14 6 7

I

I

0 0 0 0 0

0 0 0 0 0 0

14 8 1 1

I

0 1 1 0 0 2 2 0 0 0 2 2

33

10

7

2

5

Str. 1

Str. 2

Str. 3

Exhausted Unflaked Heated fragments

0 0 0 0 0 0 0 0 0 0 0 0

0 3 2 2 0 1 2 0 0 1 1 0

3 8 3 2 3 7 3 0 0 0 3

Tow

0

12

Semi-chipped Single-platform/flat Single-platform/prismatic Multiple-platform/uniface Multiple-platform/biface Change-of-orientation/simple Change-of-orientation/globular Opposed-platform/uniface Opposed-plal1'orm/Naviform (?)

Str.4

I I

0 I I

4

I

7 4

70

site in the form of already prepared cores or even finished tools from as yet tmlocated soun:e(s).

a large number of cores and cortical flakes, the finegrained flints are characteri7.ed by blades and tools on blades. Fig.5.2 is a barchart demonstrating the contrast as well as similarity between flint types within the same group. Miscellaneous, or other flints exhibit characteristics similar to both of the two major groups.clearly showing its mixed nanue.

lnitialjlaking (and test flaking) There are no semi-chipped cores made of fine-grained flints. Core reduction of this stage was perhaps not performed in the site.

These contrast and similarity strongly suggest that the fine- and coarse-grained flints were processed according to different operational sequences. In the following analyses, therefore, the two group of flints are dealt with as two separate analytical units.

Core preparation There are two crested pieces in the present collection: one in the form of a distal fragment (Type 5; Fig.5.4: 3), and the other as a blank on which a burin was made (Type 4; Fig.5.18: 3).

5.3.2 Core-reduction sequence: Ihle-grained flints

Thereareonly a small number of fine-grained flints in the present collection (204 pieces in total), and most of them are represented by tools and retouch spalls. Technological information available on the fme-grained flints is accordingly limited.

Raw malerial procure-nt

Sourcesfor the fine-grained flints have not been located to date. Akkermans (1988:26) mentions one possible flint outcrop in the vicinity of Damishliyya but it is unknown if the same type of flint is available there. A sample of flints from that outcrop, which was made available to me, does not include the fine-grained ones. The fine-grained flint collection from Tell Damishliyya contains only a few cores and cortex-flakes, each constituting less than 3 % of the total (Tables 5.3 and 5.6.1). Instead, it consists of a large number of tools made on non-cortical blanks. It is most likely that flints of this group were brought into the

Table 5.5 shows types of the cores made of fine-grained flints. They consist of one single-platform / flat, one multiple-platform/ biface (Fig.5.3: 1), three change-of-orientation / globular (Fig.5.3: 2), and one exhausted type of cores (Fig.5.3: 3). Platforms of these cores are generally plain. Butts of debitage and tool blanks are shown in Table 5.6.2. The most common ones are plain butts which constitute nearly 70 % of the total butts which can be typed. Platform preparation by faceting was rarely carried out (8.3 %). The frequency of overhang removal is different between blades and flakes (Table 5.6.3). While nearly half of the blades were detached from finely abraded platforms (8/19 or 42.1 %), most of the flakes were not (2/17 or 5.9 %).

Core reduction Dorsal scar patterns of debitage and tool blanks are shown in Table 5.6.4. More than half of them represent the unidirectional pattern (65.5 %). The presence of a substantial proportion of the bidirectional (26.6 % ) and the biface

100

Tell Damishliyya

5 0

5cm

7

Figure 5.3 Cores from Tell Damishliyya. I: Multiple-platform type, 2: Change-of-orientation type, 3: Semi-chipped type, 4-5: Single-platform type, 6: Naviform type (?), 7: Opposed-platform type, 8-9: Change-of-orientation type.

101

Chapter5

cracks on the surface (Table 5.6.5). Intentional heat pretreatment can not be demonstrated, however, since there is no example showing the part-heated pattern.

patterns (3.6 %), however, suggests that some sort of opposed-platform flaking technology was utilized for core-reduction. although no such cores are preserved in the present collection.

Flaking

The main products of the core reduction strategy of fine. grained flints were blades, which outnumber flakes in the blank. forms (71 %; Table 5.6.1). The flake scars left on the cores of fine-grained flints are something inconsistent with this trend, however, representing all the flake shape. This is perhaps a result of recurrent exploitation of the same core; blades appear to have been produced in earlier stages of the core reduction. In any case, the flakes in the assemblage are perhaps by-products of core preparation for blade removal.

nuxu

A markeddifference is found betweenthe flakingmodes of flakes and blades (Table 5.6.6). Almost all of the flakes were probably detached with the hard(er) hammer mode (16/17 or 94.1 %), while more than half of the blades were produced with the soft(er) hammer mode (10/19 or 52.6 %). A change of flaking tool is suggested for detachment of flakes and blades.

Summary The remarkable difference between technological features such as platform preparation, flaking directionality, and perhaps flaking mode suggests that flakes were by-products of core preparation for blade removal. Blades were often detached from opposed-platform cores, which were subsequently so intensively exploited to change their form to globular flake cores.

Percussion points for blank. production were often chosen about o-6 mm away from the platform edge. Comparison of the butt depth (distance between the percussion point and the platform edge) hasrevealed an interesting pattern. Tables 5.6.7. and 5.8.11 indicate that the percussion points for blades were generally chosen closer to the edge than for flakes irrelevant to flint groups. The butt depth of fine-grained flint blades is similar to that of coarsegrained flint blades. On the other hand, the butt depth of fine-grained flint flakes is smaller than that of flakes made of coarse-grained flints, which may suggest the use of a different flaking mode.

5.3.3 Core-reductionsequence: coarse-grainedftints

Raw material procurement Coarse-grained flints represented by grayish brown flint are the main raw material used for flaked stone tools at Damishliyya. The dominance of grayish brown flint at Tell Damishliyya has been already noted by the excavator. Akkermans (1988: 26) states in a preliminary report that a flint outcrop possibly used by the inhabitants has been found on top of calcareous rocks within a distance of 3 to 5 km from Tell Damishliyya. Flint artifacts do occur on this outcrop, but their typological characteristics differ from those of flint tools found at Damishliyya and more resemble earlier industries. He suggests, taking into consideration the presence of numerous flint debris at Damishliyya, that flint was taken from this outcrop in the form of nodules and brought into the site for subsequent knapping.

Core abandonment It is evident that cores of fine-grained flints were intensively exploited. They are all very small with the maximum length between 24 mm and 43 mm, clearly showing considerable reduction. Cores of very pale brown flints (3 pieces) still retain part of cortex, while others (two very dark brown flints and one dark brown flint) are wholly decorticated (Fig.5.3: 1, 2). Comparison of the sizes between fine- and coarse-grained flint cores is made in Table 5.3.4, demonstrating the smaller size of the finegrained flint cores. This might well reflect the more precious nature of this raw material for the knappers.

Core maintenance The change-of-orientation method was utilized to increase productivity of the same core. Cores exhibit their surfaces extensively flaked from different directions.

This interpretation is supported by a result of the present study. The coarse-grained flint collection indeed includes a certain number of unflaked flints (7/64 or 10.9 % of the core assemblage; Table 5.5) which consist of both flint nodules and chunks. Their size ranges from 57 x 36 x 30 mm to 37 x 35 x 17 mm. The relatively small size of these non-worked flints may well reflect that they were rejected pieces for knapping due to their small size. These non-

Heat treatment Nearly half of the fine-grained flints exhibit traces of thermal alteration in the form of reddish appearance and

102

Tell Damishliyya

worked flints consist of grayisb brown flints (5 pieces) and dusky red flints (2 pieces).

Ill

!

ff"):!-.or--..-:!oo---r--•

Initial flalcing

The coarse-grained flint collection includes three semichipped cores, all madeof grayish brown flinL Two of them areon nodules,and one on a thick flake. One of the pieces on nodule (Fig5.3: 3) bas a main flaking surface slightly exploited from a cortex-covered platform. The other example was exploited using a plain surface as a platform.

~a&:ia;:;a~ ~~ ~a;:; i •~ao2"'~=---gV\

Ill

fll~~a;f!l~oii~~aa;:cii N"'""--og'°

• ~°'2•

!

~

ff'\~

\0

r--M~

ff\2

\0

tnN

-

-

0

r-,,. "llt

~

fP\ -

-

0

tn

-"'

tn

:I

=

iS

1=

~, .,g

A large number of cortical flakes and blades are included in the coarse-grained flint assemblage (Table 5.8.1). Cortex and part-cortex pieces constitute 31.0 % (126/406) and 45.6 % (185/406) of the total flakes respectively; thus, nearly 80 % of the whole flakes are cortical. Blades of this flint group also include many cortical pieces (22/43 or 51.2 %). These cortical pieces demonstrate that coarse-grained flints were reduced on the site from the very beginning stage of the reduction process.

:!

-

000000000000

1111 0"'

0

N -

0

0 NO O C

N ...

!

Cor~ pr~paration

There is only one crested piece madeof coarse-grained flint (dusky red flint), which is a crested blade of Type 1 (Fig5.4: 4). The cresting technique appears to have been used very limitedly for core preparation of this type of flints. As shown on the semi-chipped cores, core reduction was most commonly commenced either by creating a large plain platform with a single blow or by using a suitable cortex-covered surface as a ready-made platform (Fig.5.3: 3) rather than employing such a formal preparation technique as cresting.

Ill

!~!!~!~!~~~!

i

I

e-oc-Cf'l"'ICC-OQ

"'

1111 c-ccccNCCOOO

,~

"'

1

e

1~

occococccccc

0

cccccc-ccooo

-

·,> i II:

~

=

!

... occccccccooo j

0

o:i cccc-ccco-cc

N

=

Cores of coarse-grained flints consist of several types (Table 5.5). The most common cores are of the singleplatform (Fig.5.3: 4, 5) and the change-of-orientation types (Fig.5.3: 8, 9), each representing about 30 % of the core assemblage. Multiple-platform cores are less common (about 15 %), and opposed-platform cores are very rare (about 3 %; Fig.5.3: 7). Generally speaking, these cores can be described as amorphous flake cores with little trace of formal core preparation. One possible example of a bladecore of the Navifonn type exists but its exhaustion prevents from reliable identification (Fig.5.3: 6).

Q

>

I~ 1 11111~ 1-2; _§_g;J •s: e ·2~11 ·2• 2 11 ·=·=1j

=

j

I

The cores have a shon and thick morphology in general. The width / thickness ratios are mainly between 1.4 and 1.8 (about 40 %). The length and the width are almost equal in most cases, and the ratio of length / width is aboutone (Table 5.7.3).

ii '-'-

; Cc. 1a ·

en 00 ~

:!5

:J

1l l ~ ~ -,-,..~ g.1l

11·.! j E lililaa :I

.,_.,_.!!.!!

rl

·15

·-Y ~.!!'?-~"'?-1·-a. ·-"E. ~,' .c

~.!!

c,C

~ u

;;

00 00

::, 0

t!

~j

j:3~

103

Chapter5

Table 5.6 Technological attributes of debitage and tool blanks from Tell Damisbliyya:Fme-grained flints 1 Flint types

DarkB

V.D.B.

Black

V.P.B.

0 4 IS 2

0 0 8 0 16

0 0 0 0 7

17 7 17

9

0.00 3.82 26.75 5.73

100

63.69 100

WeakR

Total

CJI,

Cortex flakes Pan-cortex flakes Flakes Pan-cortex blades Blades

48

0 1 2 0 12

Total

69

15

24

7

42

157

Plain

Tbinn.

Dihed.

Facet.

Cortex

10 14

1 1

4 1

2 1

0 2

17 19

24 66.67

2 5.56

5 13.89

3 8.33

2 5.56

100

Abnsion

Facet.

Absent

2 8

5 5

10 6

17 19

10 27.78

10 27.78

16 44.44

100

Biface

Cross.

Odien

Total

2 Butttypes Flakes

Blades Total CJI, .

3 Overbang removal Flakes

Blades Total 'I,

0

I

0 6 42

Total

36

Total

36

Uni-

Bi-

Flakes

28 63

5 32

2 3

4 1

1 0

40

Blades Total

91 65.47

37 26.62

5 3.60

5 3.60

1 0.72

139 100

Whole

Pait

Absent

4 11 11 66

0 0 0 0

2 5 5

92 45.TI

0 0.00

54.23

Hard

Soft

Uniclen.

16 6

0 10

1 3

17 19

22 61.11

10 27.78

4 11.11

100

4 Dona1scarpattemS

'I,

5 Thezmalalteratioo

Cores Flakes

Blades Tools Total CJI,

6 Flakingmode Flakes

Blades Total CJI,

7 Measurements (Mean;mm) Flakes (N= 10) Blades (N=8)

97 109

L

(S.D.)

w

(SD.)

T

(SD.)

PW

(S.D.)

PD

24.20 40.88

10.780 14.710

22.20 16.69

7.525 4.183

5.94

4.262 4.295

17.42 6.84

9.898 3.608

4.23 1.83

7.40

99

Total 6 16 16 163 201 100

Total

36

(SD.) 5.400 1.410

relation would confirm that striking platforms of cores were rarely prepared by means of faceting.

The cores have generally more than one platform; 83 platforms were identified on the 57 typable cores. As shown in Table 5.7.l, their platforms are mainly large plain surfaces created by a few rough blows (60.2 %) and cortical ones (34.9 %). Faceted platforms are rare, constituting less than 5 % (4.8 %). A more or less similar trend is found on the butt types of debitage and tool blanks (Table 5.8.2), which consist of mainly plain (60.4 %), conex (17.5 %) and faceted/ dihedral types (11.7 %). This cor-

Traces of overhang removal, as an alternative technique for platform preparation, were recognized on less than one quarter of debitage (Table 5.8.3, 22.5 %). Blades show relatively higher percentages of overhang removal traces but still under 30 %. The overhang removal was conducted almost exclusively by rough faceting rather than by abrasion. This is confirmed by refitted pieces 104

Tell Damishliyya

Table 5.7 Technological attributes of cores from Tell Damishliyya Plain

4 5 6 9

3 0 0

1 8 0 7 2 24 4 2 2

29 34.94

60.24

4.82

83 100

Uni-

Bi-

Crossed

Cortex

Total

0 12 7 0 I 26 3 0 0

0 0 0 0 0 0

0 0 0 7 5 2 4 0 0

5 I 3 0 0 I 0 0 0

5 13

49 61.25

3 3.75

18 22.SO

10 12.SO

80 100

No

Mean

SD.

Min.

Max.

Median

53 53 53 53 53

34.96 34.77 22.SO 1.06 1.63

9.452 9.010 6.375 0.334 0.428

18 22 12.4 0.346 0.800

58 59 38.0 2.091 2.700

34 34 20.7 1.044 1.545

No

Mean

SD.

Min.

Max.

Median

6 6 6 6 6

28.83 24.58 14.48 1.31 1.93

13.23 9.489 5.755 0.857 1.046

16 17 5.4 0.667 1.237

49 43 19.4 2.882 3.981

25.5 22.8 15.0 0.842 1.539

Semi-chipped cores (N=3) Single-platform/flatcores (N=13) Single-platform/prismaticcores (N=6) Multiple-platform/unifacecores (N=7) Multiple-platform/bifacecores (N=3) Change-of-orientation/simplecores (N=14) Change-of-orientation/globularcores (N=S) Opposed-platform/unifacecore (N= I) Opposed-platform/Naviform core (?) (N=I)

I I

.,

Total (N=53)

2 Scarpatterns(Coarse-grainedflints) Semi-chippedcores (N=3) Single-platform/flatcores (N=I 3) Single-platform/prismaticcores (N=6) Multiple-platform/unifacecores (N=7) Multiple-platform/bifacecores (N=3) Change-of-orientation/simplecores (N=14) Change-of-orientation/globularcores (N=S) Opposed-pltafonn/unifacecore (N= I) Opposed-platform/Naviform core (?) (N=1)

.,

Total (N=53)

3 Measurements (Coarse-grained flints) (mm)

Length W'idlh 1bidaJess l.JW

wrr

4 Measurements (Fme-grainedflints) (mm) Length Widlh Tbiclaless l.JW

wrr

Facet.

Cortex

1 Platform types (Coarse-grainedflints)

shown in Fig.5.4: 8. The large gap between the overlapped butts represents the part removed by faceting. In short, platform preparation for coarse-grained flint cores was carried out to very limited extent at Damishliyya

Core reduction The percussion point for blank detachment was selected most commonly behind a central dorsal ridge (Type 3, 26.4 %; Table 5.8.4). No lateral biases (Type l or Type 2) are identified. Percussion points were also selected behind a flat surface (Type 5, 21.8 %), on a cortical surface (Type 6, 21.4 %), and a surface between two dorsal ridges (Type 4, 12.9 %). Aalcing direction can be examined both on the main flak-

so

0 0 0 0 2 I

I 0 0

4

Total 5 13 6 16 5 26 8 2 2

10

7 6 29 8 1 1

ing surface of cores and on the dorsal surface of debitage. Main flaking surfaces of most cores were flaked unidirectionally (Table 5.7.2, 61.3 %). Multidirectional flaking was also conducted but occasionally (22.5 % ). Opposed bidirectional flaking was even rarely applied to cores (3.8 % ). Dorsal scar patterns of debitage and tool blanks show a more or less similar tendency (Table 5.8.5), where the unidirectional scar pattern is most common (57.1 %); multi- (or crossed) and bidirectional patterns are represented by a small number of examples (8.7 % and 3.5 % respectively). The biface pattern was recognized on a very few examples ( 1.7 % ), indicating little use of systematic core preparation. The main products of the core reduction of coarse-grained flints at Tell Damishliyya were flakes rather than blades. 105

Chapter5

3

5cm

0 ~r---