The growth of early social networks: New geochemical results of obsidian from the Ubaid to Chalcolithic Period in Syria, Iraq and the Gulf

1 2 The growth of early social networks: New geochemical results of obsidian from the Ubaid to Chalcolithic Period in Syria, Iraq and the Gulf 3 4 5 Lamya Khalidia*, Bernard Gratuzeb, Gil Steinc, Augusta McMahond, Salam Al-Quntare, Robert Carterf, Richard Cuttlerg, Philipp Drechslerh, Elizabeth Healeyi, Marie-Louise Inizanj, Damase Mouralisk, Ernst Pernickal, Anne-Kyria Robinm 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 a* Université Nice Sophia Antipolis / CNRS / Cultures et Environnements Préhistoire, Antiquité, Moyen-Âge (CEPAM) / UMR 7264 / 24 avenue des Diables Bleus/ 06300 Nice, France/ lamya.khalidi@cepam.cnrs.fr b UMR 5060, CNRS/Université d'Orléans, Institut de Recherche sur les ArchéoMATériaux Centre Ernest-Babelon (IRAMAT-CEB), 3D rue de la Férollerie F-45071 Orléans, cedex, France / gratuze@cnrs-orleans.fr c Oriental Institute / University of Chicago / 1155 East 58th St. / Chicago, IL, 60637 USA / gstein@uchicago.edu d University of Cambridge / Division of Archaeology / Downing Street / Cambridge CB2 3DZ, UK / amm36@cam.ac.uk e University of Pennsylvania/ Penn Museum/3260 South St, Philadelphia, PA 19104 / USA / salamal@sas.upenn.edu f UCL Qatar (University College London) / PO Box 25256 / Georgetown Building / Education City / Doha, Qatar / robert.carter@ucl.ac.uk g Institute of Archaeology and Antiquity, University of Birmingham, Edgbaston, Birmingham B15 2TT, UK / r.cuttler@bham.ac.uk h University of Tübingen / Institut für Ur- u. Frühgeschichte, Burgsteige 11,· 72070 Tübingen, Germany / philipp.drechsler@uni-tuebingen.de i School of Arts, Languages and Cultures, University of Manchester, Oxford Road, Manchester, UK. M13 9PL / elizabeth.a.healey@manchester.ac.uk j UMR 7055 – CNRS – Préhistoire et Technologie, Université Paris Ouest – Nanterre – La Défense / 21 allée de l‟Université, F-92023 Nanterre, cedex, France / inizan@wanadoo.fr k Université de Rouen, Département de Géographie, UFR LSH et CNRS, UMR IDEES (6266) et UMR LGP (8591) / Rue Lavoisier, 76831 Mont-Saint-Aignan, France / damase.mouralis@univ-rouen.fr l Curt-Engelhorn-Zentrum Archäometrie, D6,3, 68159 Mannheim, and Institute for Geosciences, University of Heidelberg, Im Neuenheimer Feld 236, 69120 Heidelberg, Germany / ernst.pernicka@cez-archaeometrie.de m Université Paris 1 Panthéon-Sorbonne, Laboratoire de Géographie Physique, UMR 8591 – CNRS / 1 Place Aristide Briand, 92195 Meudon cedex, France / anne-kyria.robin@lgp.cnrs.fr 1 © 2016. This manuscript version is made available under the Elsevier user license http://www.elsevier.com/open-access/userlicense/1.0/ 48 Abstract: 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 Obsidian artifacts are geochemically traceable to their geological sources of origin. The results of their analysis provide some of the most accurate testimonies of interaction, exchange and population movement. This article presents results of obsidian analyses of artifacts from twelve sites from the Middle Euphrates to the Arabian Gulf. We demonstrate that the Sıcaksu flow of Nemrut Dağ in eastern Turkey consistently supplied obsidian to the majority of sites across this region from the 7th to 4th millennia BC. This outcrop predominated in analyzed assemblages and as a production material for the region, across all site positions, sizes and periods; this has been argued to be a result of the quality, quantity and accessibility of this flow (Robin et al. this volume). The analyses demonstrate the presence of mainly finished products from a variety of additional sources in eastern Anatolia and Armenia (average > 4 sources) on northern Mesopotamian sites during this time span. We argue that the Nemrut region was a major economic node and chief actor in the establishment and dynamics of networks in the greater region. The diachronic persistence or breaks in obsidian supply from more minor sources are an additional source of information on the inner workings and development of subtle interregional socio-political and economic relations. Obsidian analysis provides a detailed picture of the contributions of increasingly complex networks and channels of communication to intensified adoption of common practices and styles across regions, to intensification of processes leading to urbanization and state formation, and to accentuating periods of stress and conflict. These data nourish and update existing models on social networks during the crucial Ubaid to Late Chalcolithic periods and advance debates on the role and impact of these networks on early state formation. 71 Highlights: 72 73 74 75 76 77 78 79 80 81 82 83  Keywords: Mesopotamia; Arabian Gulf; Ubaid; Chalcolithic; obsidian provenance; LA-HRICP-MS; social networks 84 1. Introduction 85 86 87 88 89 90 Research on the dynamics and modelling of early interaction spheres is increasingly valuable in explaining socio-economic transformations. Within such modelling, developments in methods of geological sampling and obsidian fingerprinting allow us to look beyond simple formulations of raw material exploitation (Robin et al., this volume) and its distribution (Ortega et al 2014) to include potential factors that may have had an impact on the configuration and functioning of social networks (Ibañez et al 2015). 91 92 In this paper, we evoke new data on obsidian source distinctions treated in more detail in Robin et al. (this volume) and discuss these in relation to recent Laser Ablation Inductively    We analyze 168 samples of archaeological obsidian from 12 Halaf to LC3 period sites in Mesopotamia and the Arabian Gulf using LA-HR-ICP-MS We demonstrate that these sites match a great number of sources in eastern Anatolia and Armenia and that varieties increase in the Ubaid and LC2 periods Overall matches show a preponderance of obsidian from the newly studied Sıcaksu outcrop of Nemrut Dağ across periods and regions treated The data suggests that the Nemrut region was a major economic actor in complex networks of interaction that were essential to the trajectories of pre-state societies in the ancient Near East 2 93 94 95 96 97 98 99 100 101 102 103 Coupled Plasma Mass Spectrometry (LA-ICP-MS) analyses of archaeological obsidian recovered from Halaf/Ubaid to Late Chalcolithic period (6th-4th mill BC) sites in Syria (Khirbat al-Fakhar, Tell Brak/Tell Majnuna, Tell Zeidan), Iraq (Kheit Qassim, Khirbet Derek, Umm Dabaghiyah, and Arpachiyah), Iraqi Kurdistan (Surezha) and the Arabian Gulf (asSabiyah, Dosariyah, Wadi Ḍebayʿan) (Fig. 1). We focus mainly on new data from the northern Mesopotamian sites of Tell Zeidan (Halaf-Ubaid to LC2), Surezha (LC1-LC3), Khirbat al-Fakhar (LC1-LC2) and Tell Brak/ Tell Majnuna (LC2-LC3). We then compare these to new obsidian data from sites in northern Iraq and the Arabian Gulf. We discuss implications of these results for our understanding of the development of long-distance interaction spheres in the region on a diachronic scale that spans the crucial period between the onset of the Neolithic and the beginnings of state formation. 104 3 105 Fig. 1 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 Formalized prehistoric interaction networks in the Near East are documented from as early as the Pre-pottery Neolithic A (PPNA; 10000 - 8500 BC), and their growth and impact was increasingly visible thereafter. During the Pre-Pottery Neolithic B (PPNB; 8500 - 7000 BC), social hierarchies develop in tandem with increasingly complex social networks, affecting the reach of cultural expressions and material goods, and encouraging genetic intermixing between regions (Ibañez et al. 2015). PPNB networks traversed most of the Fertile Crescent, and comparable lithic technologies such as the naviform technique can be found in the southern Levant (Khalidi et al. 2013; Burian et al. 1999) and northern reaches of Arabia (Crassard and Khalidi in press). During the early Neolithic, the source of Göllüdağ among others in central Anatolia prominently featured in obsidian assemblages but were more or less restricted to central Anatolia and the Levant, rarely supplying sites east of the middle Euphrates, while eastern Anatolian obsidian was present in smaller quantities throughout the entire region. 133 134 135 136 137 138 139 140 141 The early Ubaid is mainly known from southern Mesopotamian sites such as Ur, Eridu, and Tell al-Ubaid, while sites like Arpachiyah in northern Mesopotamia were characterized by a local Halaf culture. By the Ubaid 2/3, southern Mesopotamian Ubaid ceramics are found amid a local Neolithic culture in the Arabian Gulf (5500-5000 BC), at sites such as asSabiyah (H3) in Kuwait (Carter 2006) and Dosariyah in eastern Saudi Arabia (Masry 1997). The “Northern Ubaid” or Ubaid 3/4 (5300-4500 BC), which overlaps with the Gulf Ubaid for about three centuries, is characterized by the spread of late Ubaid material culture into existing distinct local polities in Upper Mesopotamia and eastern Anatolia and can be found at sites such as Tell Zeidan, Kosak Shamali, and Tepe Gawra. 142 143 144 145 146 147 148 149 150 151 The Late Chalcolithic period (LC1-5 or 4200-3100 BC, northern Mesopotamian regional chronology) that followed also comprises a number of developmental trajectories that vary depending on region and specific site. The LC1-3 periods witness developing forms of social complexity in the form of large public buildings (McMahon et al. 2007), organized administrative systems and specialized production of utilitarian and prestige goods (Al Kuntar and Abu Jayyab 2014; Khalidi 2014; Stein 2012) as well as unprecedented settlement growth at proto-urban scales and formalized site hierarchy (Adams 1981; Algaze 2008; AlQuntar et al. 2011; McMahon and Stone 2013; Ur et al 2007; Wilkinson and Tucker 1995; Wright 2001) and what has been interpreted as the beginnings of organized warfare (McMahon et al. 2011). During the Ubaid period (6500-4200 BC), the southern Mesopotamian ceramic style expanded, reaching the Arabian Gulf, parts of the Caucasus and Anatolia, and Mediterranean shores (Carter and Philip 2010: 1). The Ubaid, like the Middle/Late Uruk (LC4-5), is one peak of southern Mesopotamian influence on surrounding regions, namely on northern Mesopotamia and eastern Anatolia. However, it is now largely accepted that symbiotic relationships formed between regions during this period that were not simply limited to unidirectional influence or to a single dominant political economy, but rather to coexisting regional economies and social identities (Stein 2012: 126; 2010: 25). As a result, the Ubaid spread is commonly referred to as a phenomenon, an „interaction sphere‟ and a „horizon style‟ (Carter and Philip 2010; Stein 2010) to describe observed cultural similarities over long distances and millennia. Coinciding with this expanding cultural sphere, eastern Anatolian obsidian can be found across the Fertile Crescent, and its supply networks overlap over much of the region that had been previously dominantly supplied by central Anatolian sources. 4 152 153 154 155 156 157 158 159 160 161 162 163 164 165 This contribution together with that of Robin et al. (this volume), discusses issues in Near Eastern obsidian studies, including the increasing inter- and intra-source distinction of the peralkaline sources of eastern Anatolia and the increasing archaeological visibility of the „3D‟ source. Through multidisciplinary research, Robin et al. (this volume) propose that the Sıcaksu flow of Nemrut Dağ was preferred by populations from eastern Anatolia to the Persian Gulf and the southern Levant from the PPNB until at least the 3rd millennium BC. The data presented here demonstrate the overwhelming importance of Nemrut Dağ as a source and hub of economic activity across periods and regions. The identification of the major obsidian locale of Sıcaksu provides a plethora of possibilities for future research regarding the reasoning underlying human choice (individual or group-led) among resources of equal accessibility and how those choices were distorted or reinstated through time. The data also provide a window into the waxing and waning in importance of more distant sources, allowing comprehensive discussion on developing political and economic relations of Mesopotamian sites with eastern Anatolia and the Caucasus. 166 2. Previous research and current trends 167 2.1 Introduction to previous obsidian research 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 Since the 60s and 70s when Renfrew, Dixon and Cann (Cann and Renfrew 1964; Renfrew, Dixon and Cann 1966; 1968; Renfrew and Dixon 1976) published their seminal papers on obsidian exchange, the peralkaline obsidian sources of eastern Anatolia have been the source of ambiguous artifact to source attribution as a result of similarities in chemical fingerprints of available source data. At the time, Renfrew and authors had identified a number of obsidian compositional groups based on analyses of a large number of archaeological and geological obsidian samples in western Asia and beyond, providing a framework from which to interpret trade networks on a massive scale. In the last thirty years, many scholars have worked to refine source composition data in the Mediterranean, Anatolia and the Caucasus (Acquafredda et al. 1999; Balkan-Atli et al. 1999; Binder et al. 2011; Blackman et al. 1998; Cauvin and Balkan-Atli 1996; Cauvin et al. 1998; Chataigner 1994; Chataigner and Gratuze 2014a; Chataigner et al. 2014; Frahm 2012; Francaviglia 1984; 1988; 1990; 1994; Gratuze 1998; 1999; Keller et al. 1996; Le Bourdonnec 2006; Lugliè et al. 2006; Montigny et al. 1991; Poidevin 1998; Poupeau et al. 1998; Tykot 1997). As a result of this ongoing work, matches between obsidian artifacts and specific source outcrops are increasingly possible, allowing for a developing picture of diachronic patterns in exchange. Ironically, the gaps in datasets highlight the simultaneous need for finer-tuned sampling and analysis of sources and artifacts and larger spatial and chronological representation of site data. One of the most intractable obstacles to the reconstruction of prehistoric networks in western Asia has been the long-standing challenge of distinguishing the peralkaline sources of Nemrut Dağ and Bingöl A. Recent breakthroughs in their distinction and subsequent reanalysis and reattribution after forty years of uncertain provenance are testament to the gargantuan impediment this issue has been to advancing understanding of exchange routes and interregional relations. The new results presented here and in Robin et al. (this volume) are representative of the scale to which most peralkaline obsidian circulating over the greater part of western Asia can now be attributed to the source of Nemrut Dağ and particularly to its Sıcaksu outcrop. This unprecedented discovery forces us to rethink models of exchange to account for what was a major supply center that likely had an impact on the nature and development of social, political and economic networks in prehistory. This is not to say that 5 200 201 202 the Nemrut region was necessarily at the center of economic activity, but that discussions of interregional relations and network models must now account for its persistent economic power over millennia. 203 3. Obsidian Analyses 204 3.1 Method of analyses 205 206 207 208 209 210 211 212 213 Most of the analyses (149 of 168) of obsidian artifacts discussed here were conducted at the Center Ernest-Babelon, IRAMAT (UMR 5060 CNRS / Univ. Orléans) using LA-HR-ICPMS (Element XR from Thermofisher Instrument) with the protocol described by Chataigner and Gratuze (2014a and b). These were analyzed alongside geological obsidian from the Göllü Dağ (Cappadocia), Pasinler (North Turkey), Bingöl A and Nemrut Dağ (East Turkey) sources. Two different laser sources were used: an Nd YAG pulsed laser, whose frequency is quadrupled, allowing it to operate in the ultraviolet region at 266 nm (VG elemental UV microprobe), and an ArF excimer laser operating at 193 nm (Resolution M50E from Resonetics). 214 215 216 217 218 219 220 221 For three of the sites, some analyses were carried out in other laboratories. Three of nine artifacts from Dosariyah were analyzed by neutron activation analysis (INAA) by E. Pernicka (CEZA, Mannheim; see Pernicka 1992), five by LA-ICP-MS in Orléans, and one (DOS2010257) using both methods. All of the artifacts from Tell Zeidan (35) were analyzed by LAICP-MS, including twelve by H. Neff and J. Jaynes (Jaynes 2009), California State University (Long Beach,) and twenty-three in Orléans. The three artifacts from As-Sabiyah were analyzed by Gilles Chazot (Université de Bretagne Occidentale, Brest, Kallweit and Davies 2010) using ICP-MS. 222 4. The setting, the sites and results of analyses 223 4.1 Geochemical results and attribution to sources 224 225 226 227 228 229 As different methods and analytical protocols were used, all elemental data are not available for all of the artifacts (Table 1, 2), especially for those analyzed by INAA (e.g. Y, Nb, Sr and Ba were not determined). Some systematic deviation between different laboratories due to the calibration protocol used are possible, however in most cases it has been possible to compare analytical datasets obtained by these laboratories. Only one artifact from Dosariyah was difficult to assign with certainty to the Armenian obsidian sources of Gegham or Sjunik. 230 231 232 As demonstrated in Fig. 2, a majority of artifacts exhibit high concentration of iron and zirconium (104 among 168). A large number (91) have an iron content ranging from 19000 to 25000 ppm while that of the 13 other artifacts ranges from 29000 to 35500 ppm. 6 1600 Arteni 2 Arteni 3 1200 Bingöl A Zr ppm Bingöl B Erzincan Gegham 800 Gegham or Sjunik Meydan Dağ Nemrut Dağ 400 Pasinler Sarıkamiş North Sarıkamiş South 3D 0 0 10000 20000 Fe ppm 233 30000 40000 234 Fig. 2 235 236 237 238 Using the dataset of the GeObs project (Robin et al, this volume), the main group, which is peralkaline, can be assigned to the obsidian flows of the Nemrut Dağ volcano (Sıcaksu) while the smaller groups of peralkaline obsidian can be assigned to obsidian outcrops located in the Bingöl A area (Fig. 3). 40 Arteni 2 Arteni 3 35 Bingöl A Bingöl B Rb/Cs 30 Erzincan Gegham 25 Gegham or Sjunik Meydan Dağ Nemrut Dağ 20 Pasinler Sarıkamiş North 15 Sarıkamiş South 3D 10 0,0 239 0,5 1,0 Y/Nb 1,5 2,0 2,5 240 Fig. 3 241 242 243 244 245 246 The 64 remaining artifacts form 9 main chemical groups (Fig. 4) related to the obsidian sources or outcrops of Bingöl B (31), Meydan Dağ (16), Pasinler (4), Arteni (2; one from Arteni 3 and one from Arteni 2), Sarıkamiş (5; 3 related to Sarıkamiş North, 2 to Sarıkamiş South), Gegham (1), Gegham or Sjunik (1 analyzed by INAA), Erzincan (1). Finally the 3 remaining artifacts exhibit high rubidium concentrations and correspond to the source referred to as „3D‟ by Renfrew et al. (1966). 7 247 248 249 250 251 252 The source „3D‟ is characterized by high contents of lithium, boron, rubidium and cesium and artifacts matching this group have consistently been of poor quality and are often found with artifacts from the Lake Van and Taurus areas (Bingöl, Nemrut Dağ, Meydan Dağ). Its exact location remains unknown, however, recent geological work by the GeObs program points to a probable location around Bingöl area as was proposed by C. Chataigner (see Cauvin et al 1998: 317). 253 254 Table 1 255 500 Arteni 2 Arteni 3 400 Bingöl B Ba ppm Erzincan 300 Gegham Gegham or Sjunik Meydan Dağ 200 Pasinler Sarıkamiş North 100 Sarıkamiş South 3D 0 0 100 200 300 400 500 Zr ppm 256 8 257 Fig. 4 258 Obs # per Source Site Nemrut Dağ Meydan Dağ Bingöl A Bingöl B 3 D Sarıkamiş S Sarıkamiş N Pasinler Arteni 3 Arteni 2 Gegham Gegham / Sjunik Erzincan TOTAL Zeidan (Halaf-LC2) 15 1 7 7 1 0 0 3 0 1 0 0 0 35 Khirbet Derek (Halaf) 0 0 0 1 0 0 0 0 0 0 0 0 0 1 Um Dabagyah (Halaf?) 2 0 1 1 0 0 0 0 0 0 0 0 0 4 Arpachiyah (Halaf) 2 0 0 0 1 0 0 0 0 0 0 0 0 3 Kheit Qassim (Ubaid) 4 1 0 0 0 0 0 0 0 0 0 0 0 5 Dosariyah (Ubaid 2/3) 5 0 1 0 0 0 0 1 0 0 1 1 0 9 as-Sabiyah (Ubaid 2/3) 2 0 0 0 0 0 0 0 0 0 0 0 1 3 Wadi Debayʿan (Ubaid 2/3) 0 0 4 0 0 0 0 0 0 0 0 0 0 4 Surezha (LC1-3) 16 6 0 0 1 1 1 0 0 0 0 0 0 25 Khirbat alFakhar (LC1-2) 28 1 0 2 0 0 2 0 0 0 0 0 0 33 Brak (LC2) 7 4 0 6 0 1 0 0 0 0 0 0 0 18 Majnuna (LC3) 10 3 0 14 0 0 0 0 1 0 0 0 0 28 TOTAL 91 16 13 31 3 2 3 4 1 1 1 1 1 168 259 Table 2 260 4.2 Halaf - LC3 Northern Mesopotamian sites 261 Tell Zeidan (Halaf – LC2) 262 263 264 265 266 267 268 269 270 271 272 273 274 Tell Zeidan is a 12.5 hectares located at the confluence of the Euphrates and Balikh rivers in north central Syria. It consists of a large southern mound 15 m in height, a lower town, and two smaller mounds: the northeast and northwest mounds. It was excavated between 2008 and 2010 by the Oriental Institute (University of Chicago) and the Syrian General Directorate of Antiquities and Museums (DGAM). A continuous occupation sequence covers the Halaf to LC2 periods. A number of Ubaid domestic contexts as well as the outer wall of a large public building were excavated. Specialized production is demonstrated by prestige goods made from exotic materials, as well as a clay boat model and ceramic wasters and a large number of spatially concentrated pyrotechnic features – apparently Ubaid pottery kilns. 99% of ceramics were northern Ubaid, combining typically southern Mesopotamian geometric decoration with local animal and human figural motifs. Area D produced domestic architecture spanning the late Ubaid to LC1 transition, while Area E revealed LC1 and LC2 domestic occupation and burials (approximately 40). Area F produced an LC2 pottery kiln 9 278 279 280 281 282 283 284 Obsidian is found in all areas and most contexts at Tell Zeidan (Table 3) with the greatest concentration being in the northeast mound area. Preliminary analysis indicates that obsidian accounts for between 4 and 7% of the chipped stone assemblage depending on the area. Most obsidian artifacts are of green peralkaline obsidian, but transparent greys, browns and reds are also present. Analysis of their geologic origin demonstrates that they match an unusually high number of sources. As discussed below there is some evidence to suggest that some obsidian (almost exclusively the green peralkaline type) was worked at Zeidan. Totals Blades Cores Weight in g Obsidian from Tell Zeidan Samples extracted (not yet classified) 277 Flakes, chips and frags providing evidence of on-site ceramic production during this period. Following the LC2 period, the site was abandoned until the 3rd millennium BC (Stein 2010-2011). Splintered pieces 275 276 South Mound (78% green) Operation 1 18 7 5 24 9 3 Operation 8 2 Operation 9 0 1 6 5 4 Subtotal S mound 2 1 48 21 12 6 2 1 89 20 2 30 7 16 13 2 12 3 1 153 45 15 216 84 145.9 6 207 334.7 11 4 30 77 77 22 321 557.6 Northeast Mound (63% green) Operation 7 0 Operation 11 1 Operation 14 0 Operation 3 0 Operation 4 0 Subtotal NE mound 1 1 1 2 Northwest mound (80% green) Operation 2 0 TOTALS 3 3 10 285 Table 3 286 287 288 289 290 291 292 The assemblage is comprised predominantly of blades (>70%); although rare, there are also cores, core fragments and crested blades suggesting that at least some of the obsidian was reduced at Zeidan. Three splintered pieces were also recorded (Fig. 5: 11-12). Flakes, chips and other debris account for 24% of the assemblage. One large blade core measuring 12.4 cm in length and weighing 81.9 g featuring a crested back and an obliquely set striking platform (not included in preliminary totals; Fig. 6) came from a Late Chalcolithic context in Operation 3 on the northeast mound. 293 294 Fig. 5 295 296 297 298 299 300 301 302 303 304 The other complete core is from an Ubaid context in Operation 1 on the south mound and is much smaller (Fig. 5: 1). It measures 5.6 cm in length and weighs 14.8g featuring a horizontal platform and flat back with blades removed around three quarters of the perimeter. As well as the core, blades exhibit technological attributes associated with pressure flaking (Pelegrin 2012); most are narrow (41% were 10 mm or less and a further 38% between 10 and 15 mm wide), and only 20% are wider (Fig. 5: 7-9; Fig. 7). Modification is rare and almost entirely takes the form of retouch on the edge of the blade; there are also a small number of non-tool items from Operation 11 (not listed in Table 3) including a substantial basal fragment of a bowl of transparent grey obsidian (Stein 2010: 115, Fig. 19), a vperforated button and another flat piece with a heavily ground ventral face. 305 Fig. 6 HERE 306 Fig. 7 HERE 307 308 309 310 311 312 313 The techno-typological aspect of the Zeidan assemblage is different from the broadly contemporary Ubaid assemblage of Kenan Tepe in the Tigris valley near Bismil, where obsidian accounted for almost 25% of the chipped stone. At Kenan Tepe flakes were more usual than blades but this may reflect the type of obsidian being used (some of which is of poor quality and often in the form of small nodules). Even so, two obsidian bowls and a bead were also present in the Kenan assemblage suggesting that it, like Zeidan, may have had other more prestigious connections. (Healey 2006, 2009; Healey forthcoming). 314 Tell Zeidan geochemical results 11 315 Thirty-five obsidian artifacts from Tell Zeidan form seven groups (Table 4). 316 317 318 319 The data comprise the largest number of sources represented for any Ubaid Near Eastern site to date, which suggests that Ubaid Tell Zeidan was linked to a complex, multi-stranded obsidian exchange network that included the eastern source zones of Anatolia and the western Caucasus region. 320 321 322 The most represented source in Ubaid period contexts at Tell Zeidan is Nemrut Dağ (Sıcaksu outcrop), followed by Bingöl B (Aletepe and Catak outcrops; calcalkaline Bingöl) and Bingöl A (peralkaline Bingöl). 323 324 325 326 327 Very few obsidian artifacts were analyzed from later contexts at the site and therefore may not be representative. Nonetheless, the source matches suggest that besides the single object analyzed from the Ubaid/LC1 transition that matched Bingöl A, Nemrut Dağ continued to be a major supply zone until the LC2 period. The only other source represented in Late Chalcolithic contexts thus far is Bingöl B. 328 329 330 331 332 333 334 The results of the geochemical analyses suggest that Nemrut Dağ was a major supplier of obsidian to sites across the Near East and that the Ubaid period marked an increase in the extent of the obsidian exchange network, introducing a larger variety of obsidian to certain sites such as Tell Zeidan and Kenan Tepe (five sources represented; Frahm in Healey forthcoming). Given the diversity of obsidian found at Ubaid Tell Zeidan and Kenan Tepe, this period appears to have witnessed not only exploration of territory but also of obsidian resource acquisition, two things that are likely inextricably connected. 335 336 337 338 Though sample quantities are small, the Tell Zeidan Late Chalcolithic obsidian source matches reflect contemporaneous patterns in obsidian acquisition strategies at Tell Brak and Tell Hamoukar, where Nemrut Dağ, followed by Bingöl B, were the major suppliers in Late Chalcolithic 1/2 contexts. 339 Breakdown of number of artifacts analyzed per source/period – Tell Zeidan SOURCE # OBS PER Nemrut Meydan Bingöl Bingöl 3D Pasinler Arteni 2 TOTAL Dağ Dağ A B PERIODS Halaf1 Ubaid 1 Transition 9 1 5 6 1 2 1 Ubaid 25 2 Ubaid/LC1 1 2 1 LC1 3 3 LC2 3 1 Unknown 1 TOTAL 15 1 7 7 1 3 1 35 340 Table 4 341 Surezha (LC1-3) 342 343 344 345 Surezha is a 31 hectare mounded site located 20 km south of Erbil in Iraqi Kurdistan. The Chalcolithic occupation of the site is concentrated on the 2.8 ha high mound. Investigation of this site by the Oriental Institute began in 2013 and continues today. Systematic survey has revealed occupation from the Ubaid to Uruk periods. Ubaid levels had not yet been reached at 12 346 347 the time of the analyses, therefore all artifacts pertaining to this site date to the LC1-LC3 periods. 348 349 350 351 Excavations revealed intact LC1 (4800-4200 BC) domestic architecture and exterior surfaces. Radiocarbon dates, corroborated by ceramic studies, date the beginning of the LC1 in this part of Upper Mesopotamia 300 years earlier than in the neighboring Khabur region (Stein and Alizadeh 2013-2014: 149). 352 353 354 355 At Surezha, obsidian fragments analyzed were recovered from LC1 levels that include one LC1 pit context, interior surfaces in domestic contexts, and collapse levels, as well as from LC2 mudbrick collapse and detritus levels, LC2/3 transition trash deposits and a variety of LC3 contexts including ashy pits, floors, trash pits and mudbrick detritus levels. 356 Obsidian from Surezha 357 358 359 360 Obsidian is present at Surezha in the form of prismatic blades and knapping debris, but it comprises a relatively small component of the chipped stone assemblage. Preliminary analyses of the chipped stone have so far focused on the chert blade tools (99) with fully preserved widths among which many have remains of bitumen and silica. 361 Surezha geochemical results 362 Twenty-five obsidian artifacts from Surezha form five groups (Table 5). 363 364 365 366 367 368 Lake Van sources dominate the analyzed Late Chalcolithic obsidian assemblage at Surezha, with Nemrut Dağ (Sıcaksu) matching the greatest number of artifacts across levels. The largest number of sources represented is found in the LC1 period, where the elusive „3D‟ appears, as well as a small sample from Sarıkamiş South and North in northeast Turkey. Sarıkamiş North is also represented at the contemporary site of Khirbat al-Fakhar, while Sarıkamiş South is represented in LC2 levels at Tell Brak (TW). Breakdown of number of obsidian artifacts analyzed per source/period – Surezha SOURCE # OBS PER Nemrut Meydan Sarıkamiş Sarıkamiş 3D TOTAL Dağ Dağ South North PERIODS 3 3 1 1 1 LC 1 9 3 3 LC2 6 2 LC2/3 2 7 LC3 7 1 Unknown 1 TOTAL 16 6 1 1 1 25 369 Table 5 370 Tell Hamoukar’s Khirbat al-Fakhar (LC1-2) 371 372 373 374 375 376 Khirbat al-Fakhar is located midway between the Tigris River and Khabur basin in northeast Syria and lies due south of the Tell Hamoukar mound. It was mainly occupied in the late 5 th millennium BC or LC1-2 periods (ca. 4300-3850 BC), before being abandoned. Systematic survey across the site demonstrated dense artifact scatter of LC1-2 date over an area of 300 hectares (Ur 2002: 64; 2010). The site is characterized by a low mounded center surrounded by an extensively plowed area densely strewn with artifacts. 13 377 378 379 380 381 382 383 384 385 386 387 The low mounded area was excavated in three trenches, which revealed three consecutively occupied architectural levels (I-III) that support the presence of sedentary manufacturing households. Soundings (1-10) carried out beyond this area presented poorly preserved remains of LC1-2 architecture and pits. T. J. Wilkinson proposed that these remains were left by seasonal mobile groups (Khalidi and Gratuze 2013; Wilkinson 2002: 99). Site size is significantly bigger than any other site of the period and for millennia thereafter. The morphology of the site is unique and may reflect an atypical version of proto-urbanism whereby population aggregation occurred in the form of a center with dispersed sedentary households hosting large numbers of seasonal mobile groups along its periphery, rather than the more commonly accepted form of a city as a densely populated and physically constrained area. 388 389 390 391 392 393 This site grew in an anomalous way contemporary with regional urbanism: site sizes expanded, site hierarchies became apparent, and populations aggregated into centers, all during a period when relations with southern Mesopotamia had waned. The study of the materials recovered allows us to better reconstruct the functioning of such a site within a larger regional system and to assess what would have attracted populations, enabling such unprecedented site size, albeit at low density. 394 395 396 397 398 399 400 401 402 403 Across the LC1-2 levels and areas excavated, there is evidence of on-site mass production of ceramics. Increased standardization of forms, use of chaff temper, and decline in decoration characterize this local ceramic production and gradually distinguish it from preceding periods (Al Quntar and Abu Jayyab 2014: 99). Obsidian blade production occurred in quantities that reflect intensive on-site activity. The large quantities of obsidian excavated (> 5000 artifacts) from LC1-2 levels are seldom seen at sites in the region or at distances of 200-300 km from source areas (Khalidi and Gratuze 2013). While Khirbat al-Fakhar‟s size and its obsidian quantities are an anomaly for the period and the region, the site likely developed in a unique way in response to growing populations and their needs, and to intensified interaction and exchange. 404 Obsidian from Khirbat al-Fakhar 405 406 407 408 409 410 411 412 413 414 The obsidian discussed here originates from excavations in the mounded area and from a sounding (10) in the non-mounded portions of the site. Obsidian makes up 97% of the raw material used for tool-making at Khirbat al-Fakhar. Every stage of obsidian manufacture is found, from neo-cortical blade core preparation flakes to final products (mainly blades), strongly suggesting on-site production. However, 70% of the obsidian studied from these excavations is made up of blades and another 6% of other tool forms. The remaining 24% is made up of cores and core trimming debris. This implies that more than 3/4 (76%) of the obsidian remains were final products for consumption and/or trade. All blade cores studied were single platform. Blades suggest the use of a pressure debitage technique although percussion techniques are evident in core preparation and core trimming debris. 415 Khirbat al-Fakhar geochemical results 416 Thirty-three obsidian artifacts form four groups (Table 6). 417 418 419 420 421 Twenty-eight obsidian artifacts analyzed match Nemrut‟s Sıcaksu outcrop. Save for two samples from Sarıkamiş North, the great majority of sources represented are located in the Lake Van region of eastern Anatolia. Interestingly, Bingöl A is not represented in the analyzed assemblage of peralkaline archaeological obsidian from this site, despite such obsidian predominating (99%) in the total assemblage (Khalidi and Gratuze 2013: 19). 14 422 423 424 425 426 427 Results point to direct relations between Khirbat al-Fakhar and Nemrut Dağ that were linked, at least in part, to obsidian acquisition. The presence of a few finished elements originating from the Lake Van region in which Nemrut is situated are expected. The representation of a small sample from Sarıkamiş North is enigmatic but may have had more to do with existing networks between northeast Anatolia and Nemrut than with direct links between Khirbat alFakhar and the Sarıkamiş area. 428 Breakdown of number of artifacts per source/period – Khirbat al-Fakhar # OBS PER SOURCE Nemrut Meydan Sarıkamiş Bingöl B TOTAL Dağ Dağ North PERIODS 28 1 2 2 LC1-LC2 33 Phase I or 8 2 later 7 1 Phase II 13 2 Phase III mound extension TOTAL 6 7 28 1 1 1 2 2 10 8 15 7 8 33 429 Table 6 430 Tell Brak (LC2-LC3) 431 432 433 434 435 436 437 438 439 440 Tell Brak is situated in the southern part of the dry farming zone in the upper Khabur Plain, northeastern Syria. The tell is one of the largest sites in northern Mesopotamia with a height of 40 m and a surface area of 130 hectares. This large gateway site, strategically placed on major exchange routes that linked southern Mesopotamia, the upper Tigris and western Syria, had grown to 55 hectares (Ur et al. 2007) in the LC2 period (4200-3800 BC) concentrated on the central mound. The site reached its maximum extent of 130 hectares (Ur et al. 2007) in the LC3 period (3900-3600 BC) when its outer town was settled. This outer town comprised a corona of small sub-mounds and a ring of low-density occupation between the corona and central mound. Tell Majnuna, 450 m north of the central mound, is the largest of the submounds, at 2-3 hectares and 7 m high. 441 442 443 444 445 446 447 448 449 450 The late 5th to 4th millennia occupation (LC2-LC3) at Brak was excavated in several areas on the central mound and outer town. Area TW features a sequence that spans the entire 4th millennium BC. During the LC2 period, the area was an industrial zone adjacent to a monumental building with a large basalt threshold and thick mudbrick walls; its function remains unknown but is hypothesized as secular administrative. Most of the TW obsidian analyzed was excavated from the adjacent industrial zone, which provided evidence of centralized ceramic, stone tool and textile manufacture, as well as the production of elite items such as beads, inlay and unique objects such as an obsidian chalice (Khalidi 2014). The obsidian (18) analyzed from TW pertains to LC2 level 21, which may have had a temporal overlap with occupation of Khirbat al-Fakhar. 451 452 453 During the LC3 period, Area TW‟s function shifted to feasting, and industry may have moved to the Outer Town. Tell Majnuna and adjacent outer town sub-mounds, such as T2, were most intensively occupied during this period 454 455 456 Tell T2 had mixed domestic occupation and intensive industry, mainly manufacture of pottery and possible textile dyeing or leather tanning (McMahon & Stone 2013). However, Majnuna is anomalous to any other excavated mounds in that it served solely as a rubbish 15 457 458 459 460 461 462 463 dump for discard of garbage from the dense occupation on the central mound, accumulating via a series of managed and large-scale episodes. Excavation at Majnuna also revealed a series of mass graves dating between 3900 and 3700 BC. These mass graves are secondary single-episode deposits that have been interpreted as the outcome of warfare based on analysis of stratigraphy, forensics, age-curves and pathologies (McMahon et al. 2011). The graves were covered by layers of rubbish consisting of ash, soil, broken pottery, tools, container sealings, and animal remains. 464 Obsidian from TW, Tell Majnuna and T2 465 466 467 468 469 470 471 472 473 474 Eighteen obsidian samples excavated from LC2 level 21 in area TW were analyzed. Eight of these were re-analyzed (Khalidi et al. 2009) in order to conclusively distinguish their peralkaline origins. This level contained relatively ephemeral and unplanned structures, ovens and other industrial features built alongside a major N-S street leading into the city. Obsidian dominates the lithic assemblage in this level and is used to make pressure blades and prestige objects such as beads and pendants. While knapping of blades and objects occurred on-site, several blades were traded to the site as finished products. The scale and intensity of the industries intensified in the following Level 20, where the same lithic tradition continued but with a gradual increase in larger and thicker chert blade products. Obsidian analyzed came from outdoor (street, pits, courtyards) as well as indoor (floors, fill, bins, ovens) contexts. 475 476 477 478 479 480 481 Twenty-eight obsidian samples excavated from LC3 levels in the outer town mounds of Majnuna and T2 were analyzed. A study of the total lithic material from Majnuna is ongoing, however, preliminary observation has shown a large predominance of chert in the assemblages (both knapping debris and finished tools) and production of wider and thicker chert blades than in previous periods. Among the obsidian recovered were few pendants and blade fragments as well as some debris. The contexts at Majnuna are within single episode LC3 rubbish dumps, while at T2, the context pertains to LC3 mixed industry. 482 TW, Tell Majnuna and T2 geochemical results 483 Forty-six obsidian artifacts form 5 groups (Tables 7, 8). 484 485 486 487 488 489 490 491 492 493 Similar to the situation at Khirbat al-Fakhar, Bingöl A is also absent among the peralkaline obsidian artifacts analyzed from Brak, while Bingöl B is present. Peralkaline obsidian makes up the vast majority (84% in Level 21 and 93% in Level 20) of the total obsidian assemblage at Brak, and was the most commonly knapped on-site (Khalidi 2014: 72). However, a greater number of non-peralkaline obsidian artifacts were selected for analysis to determine the range of less prominent sources represented. This partially explains why Bingöl B makes up the majority of obsidian analyzed from the TW and Majnuna / T2 assemblages (Table 7). Bingöl B obsidian was also knapped to make tools, beads and pendants at Brak and was an important secondary source to Nemrut Dağ (Sıcaksu) across the periods represented, unlike at Khirbat al-Fakhar. 494 495 496 497 Aside from the predominance of eastern Anatolian obsidian, we find obsidian from Sarıkamiş South (northeast Turkey) in LC2 contexts at Brak, and a sample of obsidian from the Armenian source of Arteni 3 in LC3 mass grave deposits at Majnuna. # OBS PER Breakdown of number of artifacts per source/site area SOURCE Nemrut Meydan Bingöl B Sarıkamiş Arteni 3 TOTAL 16 498 SITES EME mid LC3 MTW early LC3 MTW2 early LC3 MTW4 Early LC3 EM early LC3 T2E-A early LC3 T2E-B early LC3 T2E-C early LC3 GPS2 TW-N mid LC2 TW-SW mid LC2 Dağ 1 TOTAL Table 7 Dağ South 1 1 1 2 2 2 1 1 8 10 5 1 10 1 1 1 1 1 1 1 3 3 3 4 1 3 17 7 20 2 1 10 1 8 1 1 46 499 Breakdown of number of artifacts per period at Tell Brak, Tell Majnuna, T2 SOURCE # OBS PER Nemrut Meydan Sarıkamiş Bingöl B Arteni 3 TOTAL Dağ Dağ South PERIOD 1 1 Mid LC3 8 3 14 1 26 Early LC3 7 4 6 1 18 Mid LC2 1 1 Unknown 17 7 20 1 1 TOTAL 46 500 Table 8 501 4.3 Comparison to Halaf and Ubaid Sites in northern Iraq 502 503 504 505 506 Thirteen obsidian finds collected by M.-L Inizan in the late 70s and early 80s in the context of a French program directed by the late Jean-Daniel Forest in northern Iraq, were analyzed by LA-ICP-MS. Save for five samples (2 blades, 1 bladelet and 2 flakes) excavated from Ubaid levels at Kheit Qassim (KQIII-Z12-Carré 3) (Forest 1980; Forest-Foucault 1980; Inizan 1986), the remaining obsidian artifacts are surface finds recovered in Halaf-rich 17 507 508 surface contexts at Khirbet Derek (1 backed circle segment), Umm Dabaghiyah (2 blades and 2 flakes) and Arpachiyah (2 flakes and 1 drill). 509 510 Results of analyses on these samples show that five sources (including „3D‟) are represented across these four Halaf/Ubaid period northern Iraqi sites. 511 Geochemical results from Halaf/Ubaid sites 512 Thirteen obsidian artifacts form 5 groups (Table 9). 513 514 515 The majority of obsidian samples from these four sites match the Sıcaksu outcrop of Nemrut Dağ and the totality of these samples originate from eastern Anatolian sources. Interestingly the source „3D‟ is found at Arpachiyah. Breakdown of number of artifacts per source/site – Halaf/Ubaid North Iraq SOURCE # OBS PER Nemrut Meydan Bingöl Bingöl 3D Dağ Dağ A B SITE Khirbet 1 Derek Umm 2 1 1 Dabaghiyah 2 1 Arpachiyah Kheit 4 1 Qassim III TOTAL 8 1 1 2 1 TOTAL 1 4 3 5 13 516 Table 9 517 4.4 Comparison to Ubaid sites in the Arabian Gulf 518 519 520 521 522 523 524 525 526 527 To date, very few obsidian samples have been recovered in the Arabian Gulf region. While southern Mesopotamian Ubaid ceramics are common on coastal Middle Neolithic sites in the Central Gulf region and can be found as far as Ra‟s al Khaimah in the northern Emirates, obsidian is rare in the area. Our analyses are the first to be carried out on these few obsidian elements from Gulf sites, and results provide an emerging picture of a dynamic and farreaching Ubaid interaction sphere. While obsidian counts are generally extremely low for the region, 16 obsidian samples from three sites were analyzed or compared to our database and provide us with five different eastern Anatolian and Armenian origins for obsidian traded to the Gulf at distances often exceeding 1500 km during the Ubaid 2/3 or Middle Neolithic period. 528 Obsidian from Dosariyah (Saudi Arabia) 529 530 531 532 533 534 535 536 537 Dosariyah is a Middle Neolithic site located approximately 10 km south of Jubail and 800 m inland of the current shore of the Arabian Gulf in the Eastern Province of the Kingdom of Saudi Arabia. More than 20 radiocarbon dates place the site in the late 6th and first half of the 5th millennium BC (Drechsler 2011: 74). As the Ubaid pottery from the site covers the range of elements characteristic of the early Ubaid 3 (Ubaid 2/3) to Ubaid 4, a larger time frame for the occupation of the site can be expected. The distinctively southern Mesopotamian Ubaid elements at the site are clearly imports within an indigenous central Gulf Neolithic repertoire that includes bifacial stone tools including projectiles, bone tools and local ceramic coarse wares. At Dosariyah local ceramic wares make up 20% of ceramics (Kainert and Drechsler 18 538 539 540 541 542 543 544 545 546 2011). The main matrix of the site consists of pearl oyster shell remains and widely lacks visible architecture. There is a complete absence of later settlement making the likelihood of an Ubaid date for surface finds very high. Nine obsidian samples from this site were analyzed using two different laboratories and methods and the results were calibrated and compared. 547 548 549 550 551 The first series consisted of four bladelet segments. Two of these are from surface collections and include a medial and a proximal bladelet segment. The two others are medial bladelet segments recovered from excavations. Three of these are peralkaline obsidian and correspond to Nemrut Dağ (2) and Bingöl A (1), while the fourth belongs to either the Gegham or Sjunik sources in Armenia which are at least 1,500 km as the crow flies from Dosariyah. 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 The second series of obsidian analyzed using LA-ICP-MS by BG included six blade segments. All but one were collected on the surface of Dosariyah. Four of these were peralkaline in composition (2 proximal including that originally analyzed by EP and 2 medial). These all correspond to the source of Nemrut Dağ in eastern Anatolia. The following two blades are medial segments. The first was excavated and corresponds to the source of Pasinler in eastern Anatolia, while the second corresponds to the source of Mount Gegham in Armenia. 572 573 574 575 576 577 578 Excavations at as-Sabiyah (H3) in Kuwait, the most northerly Ubaid related coastal site in the Arabian Gulf, have provided a large gamut of local and southern Mesopotamian material culture that attests to the complexity of local societies and the intensity of maritime interaction in this region in the Ubaid 2/3 period. As well as a local ceramic coarse ware (20% of the assemblage), this site has complex stone architecture and impressed bitumen slabs that confirm the presence of early seafaring technology which facilitated the growth of interaction networks. 579 580 581 582 583 584 In addition to local bifacial tools and local production of shell and pearl jewelry, H3 provided a variety of southern Mesopotamian finds including stone and ceramic flanged discs and nailshaped objects as well as large amounts of Ubaid 2/3 pottery. As well as artifacts with clear Mesopotamian origins, at least seven obsidian bladelet fragments and flakes from Periods 4 and 3 were excavated, of which three (artifacts 1003.04, 3006.20, 3030.F10; Fig. 6.2, Carter and Crawford 2010) were analyzed by LA-ICP-MS in 2000 and 2002 by G. Chazot. At the The first series of four obsidian samples was analyzed using Neutron Activation Analysis (NAA) by EP and the second, comprising five new samples and one previously analyzed by EP, were analyzed using LA-ICP-MS by BG, to which the first series of compositions were compared. Results distinguish five obsidian composition groups among the nine samples, including those of eastern Anatolia, Pasinler in northeast Anatolia, and two Armenian flows. In addition, the obsidian recovered was in the form of finished blade fragments, some of which are likely pressure blades. No obsidian production debris was found, and the pressure technique has yet to be documented in Arabia. These blades comprise one of the many northern materials and objects that shifted hands, ending up at Dosariyah where populations were active participants in an Ubaid interaction sphere. This site exhibits different levels of connectedness to Ubaid period networks and underlines the importance of understanding the way such distinct networks developed, operated and merged in this unique period of intensive social networking among indigenous communities across the region. Obsidian from As-Sabiyah (Kuwait) 19 585 586 587 588 589 590 591 592 593 594 time of their analysis, their peralkaline composition was not comparable to source data available for Turkey and the Caucasus and G. Chazot deduced that their origins might be in Arabia or the Horn of Africa. Previous comparisons to our datasets showed vague similarities between two of these bladelets with the Yemen source of Yafa‟ (Kallweit and Davies 2010: 111). However, recent breakthroughs in distinctions between the two Turkish peralkaline sources of Nemrut Dağ and Bingöl A (Frahm 2012; Khalidi and Gratuze 2013), and recent systematic sampling of the former source (Robin et al. this volume), have provided new source fingerprints that can be compared with G. Chazot‟s data. The comparison reveals these two bladelets are very close in composition to the source of Nemrut Dağ. The third analyzed flake (1003.04) is likely to have come from the Erzincan region. 595 Obsidian from Wadi Debayʿan (Qatar) 596 597 598 599 600 601 A program of survey during the development of the Qatar National Historic Environment Record, led to the discovery of multi-phased chenier beach ridges, a flint scatter and a cemetery all dated to the Neolithic. The sites are located within and around Wadi Ḍebayʿan, lying east of the Ras ʿUshayriq Peninsula, northwest Qatar. The flint scatter (QNHER141) is located on a slightly elevated area (3-4m), 2.5km from the modern shoreline on the northeastern edge of Wadi Ḍebayʿan. 602 603 604 605 606 607 608 609 610 611 An initial surface collection across QNHER141 included an obsidian fragment, Ubaid pottery, and stone tools (Al-Naimi et al 2010). Surface collection and excavation over subsequent seasons at QNHER141 produced >150 pottery sherds (mostly of Ubaid 3/4 period cups, jars, dishes and bowls), bifacial tools, polished stone axes and a further three obsidian fragments. Postholes and hearths excavated at the site are indicative of temporary structures built from mangroves or burasti (Cuttler et al 2011). A total of 22 calibrated radiocarbon dates from QNHER141 place occupation firmly between the mid-6th and the mid-5th millennia BC, when the wadi was probably subject to a sea level high of between +1 and +3m. Analyses by LA-ICP-MS resulted in all four obsidian samples matching the source of Bingöl A in eastern Anatolia. 612 613 Sixteen obsidian artifacts form 6 groups (Table 10). # OBS PER SITE Dosariyah Dosariyah re-analyzed As-Sabiyah Wadi Debayʿan TOTAL Breakdown of number of artifacts per source/site – Ubaid 2/3 Gulf SOURCE Gegham Nemrut Meydan Bingöl Bingöl Pasinler Erzincan Gegham or Dağ Dağ A B 1 Sjunik 3 1 1 2 1 1 2 1 0 5 5 4 3 4 7 TOTAL 4 0 1 1 1 1 614 Table 10 615 5. Implications for early social networks in Ubaid and Chalcolithic Mesopotamia 16 20 616 617 618 The results presented here show that the majority of analyzed obsidian recovered from Halaf to Late Chalcolithic sites in northern Mesopotamia and the Arabian Gulf originates from the Lake Van area in eastern Turkey. 100% 90% 80% Obsidian Percentages 70% Erzincan Gegham/Sjunik 60% Gegham Arteni 2 50% Arteni 3 40% Pasinler Sarıkamiş N 30% Sarıkamiş S 20% 3D Bingöl B 10% Bingöl A 0% Meydan Dağ Nemrut Dağ Sites 619 620 Figure 8 621 622 623 624 625 626 627 628 629 630 The recent distinction made between the peralkaline sources of Nemrut Dağ and Bingöl A, as well as comprehensive geological work along the Nemrut Dağ volcano (Robin et al this volume), has allowed for a reassessment of previous data which has overwhelmingly provided a Nemrut Dağ origin for the majority of artifacts and across periods. These artifacts specifically match the high-quality Sıcaksu outcrop located on the western extent of Nemrut Dağ, providing us with a precise center of extraction of the largest quantities of archaeological obsidian yet analyzed. At Khirbat al-Fakhar, Tell Brak, and Tell Zeidan, obsidian matching this outcrop was preferred for on-site production, while finished products with this same origin can be found at greater distances from the PPNB (Tell Labwe south; Khalidi et al. 2013) until at least the Ubaid period where it appears in the Arabian Gulf. 631 632 633 634 635 636 The Bingöl B source is equally consistent in northern Mesopotamian assemblages and is found in the form of production debris and finished products (including objects) on a few of the sites studied. It appears to have been extremely important as a secondary source from the Neolithic until the Bronze Age. Interestingly it has not yet been identified on Ubaid sites in the Arabian Gulf despite comprehensive analysis of obsidian recovered from the three Gulf sites treated in this paper. 21 637 638 639 640 641 642 643 644 The source of „3D‟, for many years shrouded in mystery, now matches several artifacts from five northern Mesopotamian sites ranging in date from the Halaf to the Bronze Age (Tell Beydar; Milano and Lebeau 2014). It is likely to have a Bingöl localization, even if the precise outcrop has still not been identified. „3D‟ obsidian encountered in the few assemblages where it has appeared is generally of poor quality and workability. It is interesting to note its presence together with high-quality obsidian. Thus far, the periods during which artifacts from „3D‟ are found are variable and there is no discernible pattern to its presence. 645 646 647 648 649 650 651 652 653 Equally significant are results which demonstrate that small quantities of finished obsidian products recovered mainly from Ubaid period sites, and whose minor presence continues thereafter, have northeast Turkish (Pasinler, Erzincan) and Armenian (Gegham, Arteni) geochemical origins. Arteni, for example, is of very good quality and has been shown to have supplied obsidian in high quantities but over a limited territory in the Caucasus (Chataigner and Gratuze 2014a). Its presence at Ubaid levels at Zeidan and in LC3 contexts at Majnuna takes us beyond this limited territory. Likewise, Pasinler obsidian had only been recovered with certainty at the site of Tell Kurdu (1 bladelet, Bressy et al. 2005; Chataigner et al. 2014) prior to matches to Ubaid Zeidan and Dosariyah. 654 655 656 657 658 659 660 These data provide a relatively comprehensive view of the spatial extent of Ubaid networks. The sites of Zeidan, Dosariyah, as-Sabiyah and Wadi Debayʿan together cover the range of sources (10 of 12) represented in this dataset (save for Sarıkamiş), further highlighting the importance and uniqueness of this period for the shaping of intergroup relations and social networks. The far-reaching networks established during this period forged the way for relations with the northeastern source zones which featured in following periods, even while relations with the Arabian Gulf waned after the Ubaid. 100% 90% Erzincan 80% Gegham/ Sjunik Gegham Obsidian percentages 70% Arteni 2 60% Arteni 3 Pasinler 50% Sarıkamiş N Sarıkamiş S 40% 3D 30% Bingöl B 20% Bingöl A Meydan Dağ 10% Nemrut Dağ 0% Halaf Ubaid LC1 LC2 LC3 Unknown Period 661 22 662 Figure 9 663 664 665 666 667 In the LC1-3 periods, we witness a chronological and spatial variability in relations with source zones that are testament to the profound intra- and inter-site changes that occurred in northern Mesopotamia. The variable sizes and content of the obsidian assemblages from the sites of Zeidan, Khirbat al Fakhar and Brak are of particular interest for the Late Chalcolithic periods as they suggest a great variability in site function. 668 669 670 671 672 673 674 675 Study of obsidian assemblages from LC2 levels at Tell Brak further testifies to its role as an important gateway site to which a number of exchange networks were connected, at a time when it was growing substantially (55 ha.) and its population size required greater administration. Small-scale obsidian production was one of many industrial activities that took place there, while it also served as a center of exchange for finished products. At Late Chalcolithic levels (1/2) at Tell Zeidan, we see a similar picture to that of Brak, with a variety of eastern Anatolian sources represented in a context where on-site production was on a small scale and existed alongside exchanged obsidian finished products. 676 677 678 679 680 681 682 683 684 Khirbat al-Fakhar, on the other hand, grew to an unprecedented size (300 ha.) in the LC1-2. The scale of obsidian production on this site is unmatched in this region, and its relations with obsidian sources included direct links with Nemrut Dağ, which likely provided 99% of its obsidian for on-site production and exchange. Interestingly this site was not heavily populated and has a signature that can best be explained as a permanent center that hosted large numbers of itinerant populations during parts of the year. The large quantities of obsidian from the Sıcaksu flow are suggestive of direct links with transhumant populations from the Nemrut Dağ region who descended into the plains seasonally, bringing obsidian with them. 685 686 687 688 689 LC3 levels from Tell Majnuna and T2 provide a picture of an increasingly populated site. The mass graves of this period point to internal conflict spurred by increased aggregation during a period of climatic change. External relations visible through the limited obsidian data we have for this period in the Brak region do not show a great departure from the patterns of previous periods, save for evidence for Armenian obsidian. 690 691 692 The diachronic view provided by this work provides an emerging picture of periods of heightened relations (Ubaid and LC2) between northern Mesopotamia and eastern Anatolia, with obsidian networks consistently centering on the Nemrut Dağ region. 23 50 45 40 35 Nemrut Dağ Meydan Dağ Bingöl A 30 Obsidian Quantities Bingöl B 3D Sarıkamiş S 25 Sarıkamiş N Pasinler 20 Arteni 3 Arteni 2 Gegham 15 Gegham/ Sjunik Erzincan 10 5 0 Halaf Ubaid LC1 LC2 LC3 Period 693 694 Figure 10 695 696 697 698 699 700 701 702 Wilkinson noted that major shifts in ratios of chert to obsidian, and particularly a reduction of obsidian, occurred in northern Mesopotamia during periods of intensified interaction with southern Mesopotamia (Gibson 2014: 190). Confronting this issue requires particular focus on addressing gaps in obsidian research for southern Mesopotamian sites – a consequence of a diminution of research in Iraq in the last twenty-five years – and on building standardized datasets of ratios of chert to obsidian across entire site assemblages. This issue also reinforces the need to further research the extent, if any, of southern Mesopotamian control over its adjacent regions and their resources prior to the establishment of city-states in the 3rd 24 703 704 705 706 707 708 709 710 711 712 713 714 715 millennium BC. Our data, which do not account for entire assemblage counts but for number of sources represented per period and region, demonstrate that a larger variety of sources and source regions supplied sites during periods of most interaction between southern and northern Mesopotamia. Furthermore, detailed study of obsidian assemblages from several northern Mesopotamian sites demonstrates the preponderance of peralkaline obsidian in onsite production debris, while, with the occasional exception of Bingöl B, other types of obsidian were brought or exchanged as final blade products. The overwhelming matches of peralkaline obsidian to the Nemrut Dağ source on the select samples of obsidian analyzed permits us to tentatively assume that most peralkaline obsidian which is predominant in these assemblages comes from this same source. From this perspective, we can make out a constant in the prominence of Nemrut Dağ as a major supplier across all periods, with the integration of far reaching networks during certain periods of increased interaction that would include, but not necessarily be driven by southern Mesopotamia. 716 717 718 719 720 721 722 723 724 The relationships between the forging and development of external relations between groups, partially visible through the mapping of obsidian networks, and the societal changes that occurred are becoming clearer with larger obsidian datasets and more detailed geological information and chemical analyses. Such networks are key to understanding the development of urbanization and potential reasons for its collapse. A preliminary diachronic view of the level and forms of contact and interaction that existed across these periods proves that understanding even one element of interregional contact enables a more profound understanding of the socio-economic changes that occurred during transitions in social complexity. 725 6. Acknowledgements 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 Analyses were made possible as a result of a number of grants, namely a Plan-Pluri formation awarded to the CEPAM and the IRAMAT, and an ANR awarded to the GeObs program. Many people contributed in different ways and at different stages to making this work possible. Among them we would like to thank Joan Oates, Clemens Reichel, Michael Glascock, Harriet Crawford, Jason Ur, Sophie Boucetta, Didier Binder, McGuire Gibson, Christine Chataigner, Jean-Daniel Forest, Jessica Jaynes and Giles Chazot. We would like to thank all the members of the Tell Zeidan Oriental Institute and DGAM Project, the Surezha Oriental Institute Project, the Tell Hamoukar Oriental Institute and DGAM Project, the Tell Brak Cambridge University Project, the Dosariyah Archaeological Research Project (DARP) of the University of Tübingen, the as-Sabiyah (H3) British Archaeological Expedition to Kuwait (BAEK) project of UCL, the QNHER project directed by the Qatar Museums Authority and the University of Birmingham. Excavations at Tell Zeidan and Surezha were funded by the National Science Foundation, the Oriental Institute and the generosity of private donors. Excavations at Tell Brak were funded by the British Academy, the British Institute for the Study of Iraq, and the McDonald Institute for Archaeological Research. Most importantly we would like to thank the Syrian DGAM for supporting our fieldwork and the Syrian colleagues and students who tirelessly worked alongside us in difficult conditions, and without whom this work would not have been possible. 744 7. Bibliography 745 746 747 748 Acquafredda, P., T. Andriani, S. Lorenzoni and E. Zanettin. 1999. Chemical Characterization of Obsidians from Different Mediterranean Sources by Non-destructive SEM-EDS Analytical Method. Journal of Archaeological Science 26: 315–325. 25 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 Adams, R. M. 1981. Heartland of Cities. Chicago: University of Chicago. Algaze, G. 2008. Ancient Mesopotamia at the Dawn of Civilisation: The Evolution of an Urban Landscape. Chicago: University of Chicago Press. Al Kuntar, S. and Abu Jayyab, K. 2014. The Political Economy of the Upper Khabur in the Late Chalcolithic 1-2: Ceramic Mass-production, Standardization and Specialization. In Preludes to Urbanism: The Late Chalcolithic of Mesopotamia. Festschrift in Honour of Joan Oates. A. McMahon and H. Crawford (eds). McDonald Institute Monographs. Cambridge: McDonald Institute for Archaeological Research, pp. 89-108. Al-Naimi, F. A., R. Cuttler, H. Arrock and H. Roberts. 2010. A possible Upper Palaeolithic and Early Holocene flint scatter at Ras 'Ushayriq, western Qatar. Proceedings of the Seminar for Arabian Studies 40: 209-214. Al Quntar, S., L. Khalidi and J. Ur. 2011. Proto-urbanism in the late 5th millennium BC: survey and excavations at Khirbat al-Fakhar (Hamoukar), northeast Syria. Paléorient 37.2: 151-175. Astruc, L., B. Gratuze, J. Pelegrin and P. Akkermans. 2007. From production to use: a parcel of obsidian bladelets at Sabi Abyad II. In Technical Systems and Near Eastern PPN Communities: Proceedings of the 5th International Workshop, Fréjus, 2004. L. Astruc, D. Binder and F. Briois (eds). Antibes: APDCA, pp. 327-342. Balkan-Atli, N., D. Binder, M. C. Cauvin, E. Biçakçi, G. Der Aprahamian and C. Kuzucuoglu. 1999. Obsidian: Sources, Workshops and Trade in Central Anatolia. In Neolithic in Turkey: The Cradle of Civilization. M. Özdogan and N. Basgelen (eds). Istanbul: Arkeoloji Ve Sanat Yayinlari, pp. 133-143. Binder, D., B. Gratuze, D. Mouralis and N. Balkan-Atlı. 2011. New investigations of the Göllüdağ obsidian lava flows system: a multi-disciplinary approach. Journal of Archaeological Science 38(12): 3174-3184. Blackman, J., R. Badaljan, Z. Kikodze and Ph. Kohl. 1998. Chemical characterization of Caucasian obsidian geological sources. In L'obsidienne au Proche et Moyen Orient: du volcan à l'outil. M. C. Cauvin, A. Gourgaud, B. Gratuze, N. Arnaud, G. Poupeau, J.-L. Poidevin and C. Chataigner (eds.). BAR International Series 738 and Maison de l'Orient Méditerranéen, R. Makjanic (general ed.). Oxford: Archaeopress, pp. 205-231. Bressy, C, G. Poupeau, K.A. Yener. 2005. Cultural interactions during the Ubaid and Halaf periods: Tell Kurdu (Amuq Valley, Turkey) obsidian sourcing. Journal of Archaeological Science 32: 1560-1565. Burian, F., Friedman, E. and Mintz, E. 1999. Nahal Lavan 109 - a Pre-Pottery Neolithic site in theWestern Negev, Israel. In Festchrift für Gunter Samulla. Materialien zur von - und frühgeschichte von Hessen 8, pp. 95-120. Campbell, S and Healey, E. 2009. The Challenge of Characterising Large Assemblages of Exotic Materials: a case study of the obsidian from Domuztepe, SE Turkey. Internet Archaeology 26. (doi:10.11141/ia.26.20). 26 799 800 801 802 803 Campbell, S. and Healey, E. 2013. The obsidian at Arpachiyah, Iraq; an integrated study. In F. Borrell, J.-J Ibáñez and M. Molist (eds). Stone Tools in Transition: From HunterGatherers to Farming Societies in the Near East. The Proceedings of the Seventh PPN Conference on Chipped and Ground Stone Artifacts in the Near East. Barcelona, February 2012, pp. 529-542. 804 805 806 807 808 809 810 811 812 813 Cann, J.R. and C. Renfrew. 1964. The Characterization of Obsidian and its Application to the Mediterranean Region. Proceedings of the Prehistoric Society 30: 111-131. 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 Carter, R. 2006. Boat remains and maritime trade in the Persian Gulf during the sixth and fifth millennia BC. Antiquity 80: 52-63. 844 845 846 Cuttler, R., Tetlow, E. and Al-Naimi, A. F. 2011. Assessing the value of palaeoenvironmental data and geomorphological processes for understanding late Quaternary population dynamics in Qatar. Proceedings of the Seminar for Arabian Studies 41:1-14. Carter, R. and H. Crawford (eds). 2010. Maritime interactions in the Arabian Neolithic, Leiden. Carter, R. and Philip, G. 2010. Deconstructing the Ubaid in Beyond the Ubaid: Transformation and Integration in the Late Prehistoric Societies of the Middle East. R. Carter and G. Philip (eds). Studies in Ancient Oriental Civilization, Number 63. Chicago: The Oriental Institute of the University of Chicago, pp. 1-22. Cauvin, M. C. and N. Balkan-Atli. 1996. Rapport sur les recherches sur l'obsidienne en Cappadoce. 1993-1995. Anatolia Antiqua IV: 249-271. Cauvin, M. C., A. Gourgaud, B. Gratuze, N. Arnaud, G. Poupeau, J.-L. Poidevin and C. Chataigner (editors). 1998. L'obsidienne au Proche et Moyen Orient: Du volcan à l'outil. Oxford : Archaeopress. Chataigner, C. 1994. Les propriétés géochimiques des obsidiennes et la distinction des sources de Bingöl et du Nemrut Dag. Paléorient 20 (2): 9-17. Chataigner, C., Gratuze, B. 2014a. New data on the exploitation of obsidian in the Southern Caucasus (Armenia, Georgia) and eastern Turkey. Part 1: Source characterization. Archaeometry 56: 48-69. (http://dx.doi.org/10.1111/arcm.12006). Chataigner, C., Gratuze, B. 2014b. New Data on the Exploitation of Obsidian in the Southern Caucasus (Armenia, Georgia) and Eastern Turkey, Part 2: Obsidian Procurement from the Upper Palaeolithic to the Late Bronze Age. Archaeometry 56: 25-47. (http://dx.doi.org/10.1111/arcm.12007). Chataigner C., Işıklı M., Gratuze B., Çil V. 2014. Obsidian Sources in the Regions of Erzurum and Kars (North-East Turkey): New Data. Archaeometry 56: 351-374. (http://dx.doi.org/10.1111/arcm.12011). Crassard, R. and Khalidi, L. in press. “Dispersals, Inter-Regional Connections and the Development of the Arabian Neolithic”. In Petraglia, M., Boivin, N. and R. Crassard (eds). Proceedings of the Fyssen Foundation Symposium “From Colonisation to Globalisation: Species Movements in Human History”. CUP. 27 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 Drechsler, P. 2011. Places of contact, spheres of interaction. The Ubaid phenomenon in the central Gulf area as seen from a first season of reinvestigations at Dosariyah (Dawsāriyyah), Eastern Province, Saudi Arabia. Proceedings of the Seminar for Arabian Studies 41: 69-82. Forest, J.-D. 1980. Kheit Qassim I: un cimetière du début du troisième millénaire dans la vallée de Hamrin, Iraq. Paléorient 6: 213-220. Forest-Foucault, Ch. 1980. Rapport sur les fouilles de Kheit Qasim III-Hamrin. Paléorient 6: 221-224. Frahm, E. 2012. Distinguishing Nemrut Dağ and Bingöl A Obsidians: Geochemical and Landscape Differences and the Archaeological Implications. Journal of Archaeological Science 39: 1436-1444. Francaviglia, V. M. 1984. Characterization of Mediterranean Obsidian Sources by Classical Petrochemical Methods. Preistoria Alpina 20: 311-332. Francaviglia, V. 1988. Ancient Obsidian Sources on Pantelleria (Italy). Journal of Archaeological Science 15: 109-122. Francaviglia, V. M. 1990. Les gisements d'obsidienne hyperalkaline dans l'Ancien Monde: étude comparative. Revue d'Archéométrie 14: 43-64. Francaviglia, V. M. 1994. L'origine des outils en obsidienne de Tell Magzalia, Tell Sotto, Yarim Tepe et Kül Tepe, Iraq. Paléorient 20(2): 18-31. Gibson, M. 2014. Summary. In Preludes to Urbanism: The Late Chalcolithic of Mesopotamia. A. McMahon and H. Crawford (eds). McDonald Institute Monographs. Cambridge: McDonald Institute for Archaeological Research, pp. 189-192. Gratuze, B. 1998. Les méthodes de caractérisation de l'obsidienne. In L'obsidienne au Proche et Moyen Orient: du volcan à l'outil. M. C. Cauvin, A. Gourgaud, B. Gratuze, N. Arnaud, G. Poupeau, J.-L. Poidevin and C. Chataigner (eds). BAR International Series 738 and Maison de l'Orient Méditerranéen, R. Makjanic (general ed). Oxford: Archaeopress, pp. 31-48. Gratuze, B. 1999. Obsidian characterisation by laser ablation ICP-MS and its application to the prehistoric trade in the Mediterranean and the Near East: The sources and distribution of obsidian within the Aegean and Anatolia. Journal of Archaeological Science 26: 869-881. Healey, E. 2006. “Lithics”. In B. J. Parker, L. Dodd, A. Creekmore, E. Healey and C. Painter The Upper Tigris Archaeological Research Project (UTARP): A Preliminary Report from the 2003 and 2004 Field Seasons at Kenan Tepe. Anatolica XXXII: 23-27. 890 891 892 893 894 895 Healey, E. 2009. The chipped stone assemblage from Ubaid contexts. In B. J. Parker, C. P. Foster, K. Nicoll, J. R. Kennedy, P. Graham, A. Smith, D. E. Hopwood, M. Hopwood, K. Butler, E. Healey, M. B. Uzel and R. Jensen. The Upper Tigris Archaeological Research Project (Utarp): A Preliminary Report from the 2007 and 2008 Field Seasons at Kenan Tepe. Anatolica XXXV: 123-128. 28 896 897 Healey, E. Forthcoming. The Lithic artefacts from the Ubaid Period Contexts at Kenan Tepe. In B. Parker. The Excavations at Kenan Tepe (provisional title). 898 899 900 901 902 903 904 Inizan, M.-L. 1986. Technologie et Préhistoire récente en Mésopotamie : l'exemple du débitage par pression et de l'économie de l'obsidienne. In Coll. intern. du C.N.R.S. Préhistoire de la Mésopotamie. December 1984. Paris: éditions C.N.R.S. pp. 305-315. 905 906 907 Kainert, C. and P. Drechsler. 2014. An interplay of imports and local traditions? The pottery assemblage from Dosariyah, Saudi Arabia. Proceedings of the Seminar for Arabian Studies 44: 213–226. 908 909 910 911 Keller, J., Djerbashian, R., Pernicka, E., Karapetian, S.G., Nasedkin, V. 1996. Armenian and Caucasian obsidian occurrences as sources for the neolithic trade: Volcanological setting and chemical characteristics. In: Ş. Demirci, A. M. Özer and G.D. Summers (eds): Archaeometry 94. Proc. of the 29th Int. Symp. on Archaeometry, 9-14 May 1994, Ankara, pp. 69-86. 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 Khalidi, L. 2014. 5th millennium BC obsidian production in area TW, Tell Brak. In Preludes to Urbanism: The Late Chalcolithic of Mesopotamia. Festschrift in Honour of Joan Oates. A. McMahon and H. Crawford (eds). McDonald Institute Monographs. Cambridge: McDonald Institute for Archaeological Research, pp. 69-88. Jaynes, J. 2009. Provenance investigation of Tell Zeidan obsidian, Poster presented in the Dept. of Anthropology, College of Liberal Arts, California State University, Long Beach. Khalidi, L. and B. Gratuze 2013 Late Chalcolithic lithic assemblage at Tell Hamoukar‟s Southern Extension. Berytus 53-54 (2010-2011): 15-38. Khalidi, L., B. Gratuze, M. Haidar-Boustani, J.J. Ibáñez and L. Teira. 2013. Results of geochemical analyses of obsidian artifacts from the Neolithic site of Tell Labwe South, Lebanon. In Stone Tools in Transition: From Hunter-Gatherers to Farming Societies in the Near East. F. Borrell, J. J. Ibáñez and M. Molist (eds). Barcelona: Universitat Autònoma de Barcelona, pp. 475-494. Khalidi, L., B. Gratuze and S. Boucetta. 2009. Provenance of obsidian excavated from Late Chalcolithic levels at the sites of Tell Hamoukar and Tell Brak, Syria. Archaeometry 51(6): 879-893. Kalllweit, H. and W. Davies. 2010. The Lithic Industries of H3. In R. Carter and H. Crawford eds.: Maritime Interactions in the Arabian Neolithic. Evidence from H3, As-Sabiyah, an Ubaid-Related Site in Kuwait. Leiden: Brill, pp. 105-128. Le Bourdonnec, F.-X., G. Poupeau and C. Lugliè. 2006. SEM–EDS analysis of western Mediterranean obsidians: a new tool for Neolithic provenance studies. C. R. Geoscience 338: 1150-1157. Lugliè, C., F.-X. Le Bourdonnec, G. Poupeau, M. Bohn, S. Meloni, M. Oddone and G. Tandaa. 2006. A map of the Monte Arci (Sardinia Island, Western Mediterranean) obsidian primary to secondary sources: Implications for Neolithic provenance studies. C. R. Palevol 5: 995-1003. 29 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 Masry, A.H. 1997. Prehistory in North Eastern Arabia. The Problem of Interregional Interaction. (Second edition). Kegan Paul, London. McMahon, A. 2014. State Warfare and Pre-state Violent Conflict: Battle‟s Aftermath at Late Chalcolithic Tell Brak. In Preludes to Urbanism: The Late Chalcolithic of Mesopotamia. Festschrift in Honour of Joan Oates. A. McMahon and H. Crawford (eds). McDonald Institute Monographs. Cambridge: McDonald Institute for Archaeological Research, pp. 175-188. McMahon, A. and Stone, A. 2013. Edge of the city: urban growth and burial space in 4th millennium BC Mesopotamia. Origini 35: 83-109. McMahon, A., Sołtysiak, A. and Weber, J. 2011. Late Chalcolithic mass graves at Tell Brak, Syria, and violent conflict during the growth of early city-states. Journal of Field Archaeology 36/3: 201220. McMahon, A. and Oates, J. 2007. Excavations at Tell Brak, 2006 & 2007. Iraq 69: 145-171 Milano, L. and M. Lebeau (eds). 2014. Tell Beydar: Environmental and Technical Studies. Volume II. Brepols Publishers, Turnhout. Montigny, R., J. Tripier, Y. Besnus and M. -C. Cauvin. 1991. Nouvelles analyses d'obsidiennes du Proche-Orient : modèle de géochimie des magmas utilisé pour la recherche archéologique. Paléorient 17(2): 5-20. Pelegrin, J. 2012. New experimental observations for the characterisation of pressure blade production techniques. In The Emergence of Pressure Blade Making. From Origin to Modern Experimentation. P,M. Desrosiers (ed). New York, Dordrecht, Heidelberg, London: Springer, pp. 465-498. Pernicka, E. 1992. Herkunftsbestimmung späturukzeitlicher Obsidianfunde vom Hassek Höyük. In Hassek Höyük – Naturwissenschaftliche Beiträge, Istanbuler Forschungen 38. Behm-Blancke, M. (ed.). pp. 124–131. Poidevin, J.-L. 1998. Les gisements d'obsidienne de Turquie et de Transcaucasie: géologie, géochimie et chronométrie. In L'obsidienne au Proche et Moyen Orient: du volcan à l'outil. M. C. Cauvin, A. Gourgaud, B. Gratuze, N. Arnaud, G. Poupeau, J.-L. Poidevin and C. Chataigner (eds). BAR International Series 738 and Maison de l'Orient Méditerranéen, R. Makjanic (general ed). Oxford: Archaeopress, pp. 105-156. Poupeau, G., G. Bigazzi, L. Bellot-Gurlet and O. Dorighel. 1998. Fission-track dating of obsidians and archaeology. In L'obsidienne au Proche et Moyen Orient: du volcan à l'outil. M. C. Cauvin, A. Gourgaud, B. Gratuze, N. Arnaud, G. Poupeau, J.-L. Poidevin and C. Chataigner (eds). BAR International Series 738 and Maison de l'Orient Méditerranéen, R. Makjanic (general ed). Oxford: Archaeopress, pp. 53-67. Renfrew, C. and J. E. Dixon. 1976. Obsidian in western Asia: a review. In Problems in Economic and Social Archaeology. G. d. G. Sieveking, I. H. Longworth and K. E. Wilson (eds). London: Duckworth, pp. 137-150. 30 990 991 992 993 994 995 996 997 998 999 Renfrew, C., J. E. Dixon and J.R. Cann. 1966. Obsidian and Early Cultural Contact in the Near East. Proceedings of the Prehistoric Society 32: 30-72. 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 Stein, G. 2010-2011. Tell Zeidan. Oriental Institute Annual Report 2010-2011: 121-138. Renfrew, C., J. E. Dixon and J.R. Cann. 1968. Further Analysis of Near Eastern Obsidians. Proceedings of the Prehistoric Society 34(9): 319-331. Stein, G. 2010. Local identities and interaction spheres: modeling regional variation in the Ubaid horizon in Beyond the Ubaid. Transformation and Integration in the Late Prehistoric Societies of the Middle East. Studies in Ancient Oriental Civilisation 63. Chicago: The Oriental Institute of the University of Chicago, pp. 23-44. Stein, G. 2012. The development of indigenous social complexity in Late Chalcolithic Upper Mesopotamia in the 5th-4th millennia BC - an initial assessment. Origini XXXIV: 125-151. Stein, G. and A. Alizadeh. 2013-2014. Surezha, Kurdistan. Oriental Institute Annual Report: 138-151. Tykot, R. H. 1997. Characterization of the Monte Arci (Sardinia) Obsidian Sources. Journal of Archaeological Science 24: 467–479. Ur, Jason. 2002. Settlement and Landscape in Northern Mesopotamia: The Tell Hamoukar Survey 2000 - 2001. Akkadica 123 (2002): 57-88. Ur, J., Karsgaard, P., and Oates, J. 2007. Early urban development in the Near East. Science 317 (5842): 1188. Ur, J. A. 2010. Urbanism and Cultural Landscapes in Northeastern Syria: The Tell Hamoukar Survey 1999–2001. Tell Hamoukar 1. Chicago: Oriental Institute Publications. Wilkinson, T. J. 2002. Physical and cultural landscapes of the Hamoukar area, (Syria). Akkadica 123: 89-106. Wilkinson, T. J. and David J. Tucker. 1995. Settlement Development in the North Jazira, Iraq: A Study of the Archaeological Landscape. Warminster: Aris and Phillips. Wright, Henry T. 2001. Cultural Action in the Uruk World. In Uruk Mesopotamia and its Neighbors: Crosscultural Interactions and Their Consequences in the Era of State Formation. M. S. Rothman (ed). Santa Fe (NM): School of American Research Press, pp. 123-148. 31 1032 8. Figure and Table Captions 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 Table 1: Average compositions and standard deviations for artifact compositional groups; oxides are given in weight percent, elements are given in mg/kg. The number of samples attributed to each group is given in brackets. Empty cells in the table are due to the fact that some elements were not determined in early analyses. Nemrut Dağ (artifacts from Zeidan, Umm Dabaghyah, Arpachiyah, Kheit Qassim, Dosariyah, as-Sabiyah, Surezha, Khirbat-alFakhar, Brak and Majnuna), Bingöl B (artifacts from Zeidan, Khirbet Derek, Umm Dabaghyah, Khirbat-al-Fakhar, Brak and Majnuna), Bingöl A (artifacts from Zeidan, Umm Dabaghyah, Dosariyah and Wadi Debayʿan), Meydan Dağ (artifacts from Zeidan, Kheit Qassim, Surezha, Khirbat-al-Fakhar, Brak and Majnuna), Pasinler (artifacts from Zeidan and Dosariyah), Sarıkamiş North (artifacts from Surezha and Khirbat-al-Fakhar), Sarıkamiş South (artifacts from Surezha and Brak) Table 2: Number of obsidian artifacts analyzed per obsidian source match and per site for totality of sites featured 1047 Table 3: Preliminary analysis of the obsidian assemblage from Tell Zeidan by context 1048 1049 Table 4: Breakdown of number of obsidian artifacts analyzed per source/period at Tell Zeidan 1050 Table 5: Breakdown of number of obsidian artifacts analyzed per source/period at Surezha 1051 1052 Table 6: Breakdown of number of analyzed obsidian artifacts per source/period at Khirbat alFakhar 1053 1054 Table 7: Breakdown of number of analyzed obsidian artifacts per source/site area at Tell Brak (TW), Tell Majnuna (EM, EME, MTW) and T2 1055 1056 Table 8: Breakdown of number of analyzed obsidian artifacts per period at Tell Brak, Tell Majnuna, and T2 1057 1058 1059 Table 9: Breakdown of number of analyzed obsidian artifacts per source/site from Halaf and Ubaid period sites (Khirbet Derek, Umm Dabaghiyah, Arpachiyah, Kheit Qassim III) in North Iraq 1060 1061 Table 10: Breakdown of number of analyzed obsidian artifacts per source/site from Ubaid 2/3 sites (Dosariyah, as-Sabiyah, Wadi Debayʿan) in the Arabian Gulf 1062 1063 Figure 1: Map of obsidian sources and Halaf to Late Chalcolithic sites referenced (D. Mouralis, GeObs). 1064 1065 1066 1067 1068 1069 1070 1071 1072 Figure 2: Binary diagram Fe vs. Zr showing source groups for studied artifacts: Arteni 2 (Zeidan), Arteni 3 (Brak), Bingöl A (Zeidan, Umm Dabaghyah, Dosariyah and Wadi Debayʿan), Bingöl B (Zeidan, Khirbet Derek, Umm Dabaghyah, Khirbat-al-Fakhar, Brak and Majnuna), Erzincan (as-Sabiyah), Gegham and Gegham or Sjunik (Dosariyah), Meydan Dağ (Zeidan, Kheit Qassim, Surezha, Khirbat-al-Fakhar, Brak and Majnuna), Nemrut Dağ (Zeidan, Umm Dabaghyah, Arpachiyah, Kheit Qassim, Dosariyah, as-Sabiyah, Surezha, Khirbat-al-Fakhar, Brak and Majnuna), Pasinler (Zeidan and Dosariyah), Sarıkamiş North (Surezha and Khirbat-al-Fakhar), Sarıkamiş South (Surezha and Brak), 3D (Zeidan, Arpachiyah and Surezha). 32 1073 1074 1075 1076 1077 1078 1079 1080 1081 Figure 3: Binary diagram of Y/Nb vs. Rb/Cs ratios for studied artifacts according to their source groups: Arteni 2 (Zeidan), Arteni 3 (Brak), Bingöl A (Zeidan, Umm Dabaghyah, Dosariyah and Wadi Debayʿan), Bingöl B (Zeidan, Khirbet Derek, Umm Dabaghyah, Khirbat-al-Fakhar, Brak and Majnuna), Erzincan (as-Sabiyah, not represented), Gegham and Gegham or Sjunik (Dosariyah, one not represented), Meydan Dağ (Zeidan, Kheit Qassim, Surezha, Khirbat-al-Fakhar, Brak and Majnuna), Nemrut Dağ (Zeidan, Umm Dabaghyah, Arpachiyah, Kheit Qassim, Dosariyah, as-Sabiyah, Surezha, Khirbat-al-Fakhar, Brak and Majnuna), Pasinler (Zeidan and Dosariyah), Sarıkamiş North (Surezha and Khirbat-alFakhar), Sarıkamiş South (Surezha and Brak), 3D (Zeidan, Arpachiyah and Surezha). 1082 1083 1084 1085 1086 1087 1088 Figure 4: Binary diagram Zr vs. Ba for artifacts originating from calcalkaline obsidian sources: Arteni 2 (Zeidan), Arteni 3 (Brak), Bingöl B (Zeidan, Khirbet Derek, Umm Dabaghyah, Khirbat-al-Fakhar, Brak and Majnuna), Erzincan (as-Sabiyah), Gegham and Gegham or Sjunik (Dosariyah), Meydan Dağ (Zeidan, Kheit Qassim, Surezha, Khirbat-alFakhar, Brak and Majnuna), Pasinler (Zeidan and Dosariyah), Sarıkamiş North (Surezha and Khirbat-al-Fakhar), Sarıkamiş South (Surezha and Brak), 3D (Zeidan, Arpachiyah and Surezha). 1089 Figure 5: Examples of cores, blades and splintered pieces from Tell Zeidan, various contexts. 1090 1091 1092 Figure 6: Obsidian blade core measuring 12.4 cm in length and weighing 81.9 g featuring a crested back and an obliquely set striking platform from a Late Chalcolithic context in Operation 3 on the northeast mound of Tell Zeidan. Tell Zeidan archive. 1093 1094 Figure 7: Three blades, each made from obsidian with different macroscopic characteristics, recovered from excavations at Tell Zeidan. Tell Zeidan archive. 1095 1096 Figure 8: 100% stacked column graph of percentages of analyzed obsidian artifacts per source and per site for totality of sites and sources featured 1097 1098 Figure 9: 100% 3D stacked column graph of percentages of analyzed obsidian artifacts per source and per period for totality of sites and sources featured 1099 1100 Figure 10: Line graph of obsidian quantities (number of obsidian artifacts analyzed) by period/source for totality of sites and sources featured 1101 1102 1103 33