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