The CANEW Project

Upper Mesopotamia (SE Turkey, N Syria and N Iraq)
14C databases: 11th - 6th millennia cal BC

Data compiled by Damien BISCHOFF, with the collaboration of Agathe REINGRUBER and Laurens THISSEN
Last updated: 12/02/2006

Preliminary note
In this document are included data covering the five main regions we take to represent Upper Mesopotamia (see below). In the short comments following each site relevant details are discussed.
The order of sites is alphabetic, while dates are presented, wherever possible, according to the stratigraphic sequence, with earliest levels first. Dates within stratigraphic units ('phases', 'levels' or 'layers') are presented with earliest dates first, based on the calibrated age at 1sigma. The abbreviation 'nd' stands for 'no data available'. The arrangement of sites in the charts is not alphabetic, but proceeds from older to the most recent settlements on the basis of the radiocarbon data. 14C data are calibrated with OxCal v3.10 (Bronk Ramsey 1995, 2001, 2005), using the most recent calibration curve IntCal04 (Reimer et al. 2004).
Many people have helped in providing us, the CANeW team, with 14C dates, often fresh from the lab and still unpublished. A special word of thanks is due to the following scholars: Peter Akkermans (Tell Sabi Abyad I data), Eric Coqueugniot (Dja'de data), Harald Hauptmann (Nevali Çori data), Necmi Karul and Mehmet Özdogan (Mezraa Teleilat data), Mary M. Voigt (Gritille data). Any responsibility for the data presentation and interpretation remains, of course, with the author.
These 14C data are interpreted and presented in the CANeW Upper Mesopotamia chart.

The concept Upper Mesopotamia
In this database Upper Mesopotamia is conceived as including Southeastern Turkey, Northern Syria and Northern Iraq. Covering the territory of these three modern countries we distinguish five subregions, as follows (where radiocarbon dated sites are underlined):

I. East Taurus intermontane basins belt and Upper Tigris basin
Boytepe, Cafer Höyük, Çinaz III, Ikiz Höyük, Tepecik/Makaraz Tepe, Çayönü, Demircitepe/Demirköy, Girikihaciyan, Hakemi Use Tepe, Hallan Çemi, Kötekan, Kortik Tepe, Nevala Denik, Yayvantepe/Til Huzur

II. West Zagros valleys
Deir Hall, Ginnig, Hassuna, Jarmo, Karim Shahir, Magzaliya, M'lefaat, Nemrik, Qermez Dere, Shanidar, Shimshara, Telul et-Thalatat, Yarim Tepe, Zawi Chemi Shanidar, Zarzi

III. Gaziantep-Urfa-Mardin plateau
Girik Tepe, Göbekli Tepe, Gritille, Hayaz Höyük, Kumartepe, Nevali Çori, Sürük Mevkii, Tavuk Çay, Ziyaret Tepe

IV. Harran plain and Balikh valley
Gürcütepe, Küçükhedbe; Haramie, Khirbet esh-Sheref, Tell Assouad, Tell Damishliyya, Tell Eftaim, Tell Hammam et-Turkman, Tell Sabi Abyad I and II, Tulul Breilat

V. Syro-Mesopotamian steppes and Middle Euphrates
Abu Hureyra, Akarçay Tepe, Bouqras, Cheikh Hassan, Dja'de, El-Kowm, Halula, Jerf el Ahmar, Tell Kashkashok, Mezraa/Teleilat, Mureybet, Qdeir, Tell es-Sinn, Umm Dabaghiyah

This subdivision should be understood against the concept of a 'geoarchaeology of landscape' (cf. Wilkinson 1996), which links data on physical geography, annual precipitation, land-use and settlement types to different subsistence strategies. The topographical situation of sites is connected either to hunting strategies depending on seasonal migration of animals, as is clear in the Taurus-Zagros valleys, the Gaziantep-Urfa-Mardin plateau and the Middle-Euphrates area, or else, sites are connected to herding strategies dependent on the location of water resources, as can be demonstrated for settlements in the Taurus-Zagros valleys, the Taurus plateau, and the Syro-Mesopotamian steppes (settlements are established directly near the rivers) (cf. Schmidt 2001). Upper Mesopotamia is part of Braidwood's 'Hilly Flanks' (foothills and intermontane valleys) of the Fertile Crescent (cf. Braidwood and Howe 1960, Braidwood et al. 1983), although Hauptmann (1999) recentered its nuclear area on the Taurus-Zagros axis.

Comment CAFER HÖYÜK
Most complete sequence in the East Area (50 m2), yielding 13 levels with a total depth of 6 m over virgin soil; the occupation is reported to be continuous (J. Cauvin et al. 1999:89). In the West Area only the Late Phase has been uncovered. Sample Ly-3773 from the Middle Phase is reported to lack sufficient carbon (o.c., 94). Samples Ly-2523 and Ly-2522, both from the West Area, might stem from the 'numerous fragments of charcoal' - or 'carbonised wood' as they are described further on - associated with structures 18, 19 and 8 of Level IVc, which, incidentally, is destroyed by fire (o.c.:95, 96). Would these have been the remains of the wooden ceiling beams, as seems suggested? If so, is it warranted to hypothesize similar samples for the earlier phases as well? Most of the wood used at Cafer seems to have been salix and populus growing along the Degirmendere stream (Willcox 1991). The 'old-wood' problem seems to have been not an issue here, for explaining the big discrepancies within the Cafer 14C dates corpus.

Comment ÇAYÖNÜ
GrN-10359 is stated to be possibly contaminated by an intrusive burial (A. Özdogan 1999:41). The phasing and nomenclature of Çayönü follows the recent general overview of the site (A. Özdogan 1999), combined with Biçakçi's 14C date assignments (Biçakçi 1998). However, according to the most recent architectural analysis the Basal Pits would occur after the Grill-Plan subphase (and not before it), whereas the Large-Room subphase would last longer because the existence of a Cell-Building-Large-Room transitional phase is questioned (Biçakçi 2001). For the (early), 'pre-Halaf' pottery stages, situated somewhere between the Large Room Building Subphase and the Chalcolithic, see Özdogan and Özdogan 1993:95ff. They are not covered by 14C dating. The seven dates with tentative phase assignment, and the METU series with unknown phase assignment, have not been used in the chronological chart for SE Anatolia.

Comment GIRIKIHACIYAN
Both samples GrN-6246 and GrN-5882 apparently consist of charcoal combined from different locations, hence stemming from possibly different events.

Comment HALLAN ÇEMI TEPESI
The dates with an *) derived from very small carbon samples (Rosenberg 1994:123). Although the rough sequence of the site is known, assignment of the dates to the three excavated aceramic building levels has not yet been published (deeper levels are reported to exist [Rosenberg et al. 1998:28, note 14]).

Comment JARMO
Sequence and labelling of levels of Operation I and Operation II are independent of each other. Operation I levels 9-4 predate levels 6-1 of Operation II (cf. Kozlowski 1994, 257).

Comment NEMRIK
The lab no. Gd-8180 (8180�120 BP) as given by Kozlowski (1994, 261) is almost certainly erroneous; its correct lab no. Gd- 4222 is provided by Pazdur (1992, 117). Gd-521- (7470�60) in Pazdur (1992, 117) is corrected as Gd-5110 in Kozlowski (1994, 262).

Comment QERMEZ DERE
Unclear is whether samples OxA-3753 and OxA-3752 are coming from the same charred seed sample. If so, a X2-test fails at 5%, when trying to apply R_Combine.

Comment SHANIDAR CAVE
Layer B is dated to the Zarzian period (Solecki 1954, 414, Kozlowski 1994, 261), and consists of two sublevels B2 and B1. Not included here are four dates from the Upper and Middle Palaeolithic layers C and D (W-654: 28,700�700 BP, W-650: 33,000�1000 BP, GRO-2527: 46,000 BP and GRO-1495: 50,000 BP), and one date from the mixed top layer A (W-662: <200 BP) (cf. Solecki 1954, 416; 1971, 253).

Comment SHIMSHARA
Shimshara levels 13-9 date to the Hassuna period, and are overlying Aceramic levels 16-14. Virgin soil was not reached.

Comment SHIMSHARA
The information in Solecki (1981, 67) should be read as "50 cm to native soil" instead of "below 50 cm to native soil". According to Solecki and Rubin (1958) the sample was too small for a normal analysis.

Comment GÖBEKLI TEPE
The Schlangenpfeilergeb�ude pertains to the earlier of the two building phases attested thus far at the site, i.e. Old Phase (level III) and Young Phase (level II) (where level I relates to the topsoil finds). Since coming from the fill of the structure, the two level III samples may postdate the Old Phase. The two dates yielded by pedogenic carbonate coatings of pillars 8 and 11 from structures B and C resp. constitute an terminus ante quem for these buildings. At the time suggested by these two dates, both structures had already been abandoned and buried.

Comment GRITILLE
We are greatly indebted to Mary M. Voigt for permitting us to publish the Gritille data set.

Comment NEVALI ÇORI
We are very grateful to Harald Hauptmann for permitting us to publish the new Nevali �ori dates. The site totals five Early Neolithic building levels (I-V), level I being the oldest. The stratigraphic position of sample Hd-16784-768 (pit 277), which is not stated in Hauptmann 1999:78, is suggested to be area H5, levels III-V, analogous to KIA-14756, details of which are known. As Hd-16784-768 is a clear outlier, it has not been used in establishing the Nevali �ori time range in the chronological chart for Upper Mesopotamia.

Comment TELL DAMISHLIYYA
Tell Damishliyya levels 5-7 date the final occupation of the site, where pottery occurs from level 3 onwards.

Comment TELL SABI ABYAD I
The Sabi Abyad I operations are carried out on different parts of the mound; the levels indicated per operation are valid only for that specific operation (pers. comm. Peter Akkermans, Nov. 11, 2005). For this reason the order used here is by operation, and within operation by level.

Comment ABU HUREYRA 1
The first settlement at Abu Hureyra was excavated only in Trench E. Out of the bulk of Abu Hureyra 1 dates, only one cluster of dates (summing at 11,100�10,440 cal BC is consistent with the material evidence and the stratigraphy of Abu Hureyra 1 phases 1�3, and only these have been entered into the table supra. Dates of two later clusters (summing at 8550�7950 cal BC [OxA�6995, phase 2, BM�1719R and OxA�475, phase 3] and at 7430�7070 cal BC [OxA�6419, OxA�5843, OxA�5842, OxA�7122, OxA�6418, phase 2, OxA�6417, OxA�6336, phase 3], respectively) have been transferred to the overlying Abu Hureyra 2 settlement (see below). For a transition stage between Abu Hureyra 1 and 2 (Moore's 'intermediate period', see Moore 1998, 135, fig. 3; Moore 2000, 492f .), stratigraphic evidence is missing.

Comment ABU HUREYRA 2-3
Dates are given per trench in chronological order. Included in the table are also dates from Abu Hureyra 3 (Ubaid period to modern times: Moore 2000, 507), since no distinction between the dates from AH2 and 3 has been made in the final report. Not included here are two dates (OxA�2044 and OxA�2045) deriving from recent burials (Housley 1994, 62f.) .

Comment AKARÇAY TEPE
Several meters of deposit are present below Phase V.

Comment BOUQRAS
The Bouqras radiocarbon dates are deriving from three different operations: those without level indication are from the 1975 reset section; dates from phases I�IV from the extensive excavation, and dates from levels 1�11 are stemming from the deep sondage. Phase I = Level 1, Phase II = Level 2, Phase III = Levels 3-10 (Eric Lohof, pers. comm., 25 April 2005). Phase IV and Level 11 are not radiocarbon dated.

Comment DJA'DE
We are very grateful to Eric Coqueugniot for permitting us to publish the new Dja'de dates. Ly�12111 is bitumen-contaminated (Coqueugniot, pers. comm.). Ly�2614(Poz) is, according to the lab, erroneous due to a lack of collagene on the bone. UtC�2367 and UtC�2369 are from the same sample, but differently pre-treated; their original BP dates are not known (the BP value indicated in the table is the result of the combination of both dates).

Comment EL KOWM 2-CARACOL
Ly-4440 is erroneously labeled Ly-4400 in Cauvin and Stordeur (1994, 203).

Comment JERF EL AHMAR
The eastern tell yields an uninterrupted sequence running from the basal level VII to levels 0 and -I as the youngest. Levels 0 and -I's numeration is due to the fact that the more recent levels have not been recognized from the beginning (Stordeur and Abbès 2002, 592). The western tell counts three levels I-III, corresponding to levels I-III of Jerf East.

Comment MEZRAA TELEILAT
We would like to thank Mehmet Özdogan and Necmi Karul for permitting us to include these still unpublished dates in the CANeW 14C databases.

Comment MUREYBET
Dates for phase IA (Natufian) are available only from the end of the phase, since basal layers did not yield charcoal (Cauvin 1977, 48).

Comment QDEIR
Problems of charcoal preservation in highly saline sediments are reported (Cauvin and Stordeur 1994, 203).

Comment TELL KASHKASHOK
Except for TK-802 all samples date Tell Kashkashok to the Hassuna Ia period.

Jul	MAR	Nov
	13
2008	2009	2010