The CANEW Project

Lecture

STONES MAKING SENSE: WHAT OBSIDIAN COULD TELL ABOUT THE ORIGINS OF THE CENTRAL ANATOLIAN NEOLITHIC

Didier BINDER
CNRS-UMR 6130, 250 rue Albert Einstein, 06560 Valbonne Sophia-Antipolis, FR.
binder@esep.cnrs.fr

Recent work on the exploitation and diffusion of Cappadocian obsidian shows concrete links between Central Anatolia and the Levant during the Holocene period (Balkan-Atli et al. 1999; Bellot-Gurlet et al. 1999; Cauvin et al. 1998). The fact that these links result from human mobility is clear, but needs to be underlined in a context where some kind of geographic or climatic determinism might come to mind. One has to refrain from an excessively regionalist approach as soon as this mobility is firmly demonstrated by archaeological evidence. Chronological and spatial obsidian patterns over the whole region show that these links are constant from the 11th to the 6th millennia cal BC, but seem to be closer during the first stages of the Neolithic period. These objective data allow us to envisage the existence of interaction processes between the Levantine Neolithic and Central Anatolia beyond obsidian diffusion and exchange themselves.

Analyses of obsidian assemblages from the Middle Euphrates PPNA are very rarely carried out from the perspective of the chaînes opératoires. The study of Cheikh Hassan (Abbès et al., in press) indicates that the first phases of the working could have been realised in situ. Therefore this modality could fit with the introduction of roughing out or partly exploited cores. The corresponding blade production is of good quality, but does not seem to need any special mastery of knapping techniques and methods. Although the skills involved in the obsidian production in PPNA Mureybet seem to be much more sophisticated, the distribution of obsidian in Cheikh Hassan and Mureybet appears rather similar (Cauvin, in press). Nothing could be opposed to the hypothesis of a local distribution organised by the mediation of itinerant hawkers, moving from village to village and knapping for consumers’ demand. In any case, such a mode of acquisition and distribution is probably not structured enough to generate a strong and visible activity close to the obsidian sources, and does not create favourable conditions for local contact.

Within the Early and Middle PPNB period (8600–7500 cal BC), obsidian acquisition and diffusion processes are much better known due to the excavations in the Kömürcü-Kaletepe workshop (sector P and sector East) (Binder and Balkan-Atli 2001). This exploitation is clearly organised by highly skilled craftsmen. The products are rigorously selected in the chaîne opératoire and are diffused over very long distances reaching up to 900km (e.g., Beidha or Nahal Lavan, cf. Cauvin et al. 1998), occasionally using maritime routes (e.g., Shillourokambos – Briois et al. 1997). The intensity of this spread is probably proportional to the value given to these objects. As was true for the previous period, these artefacts were possibly not involved in domestic or utilitarian functions; they may have played a role as status markers in representations of social honour, or they may have constituted prestige goods in competitive exchanges.

Kaletepe sector P yielded two distinct productions: a) prismatic bladelets, and b) points detached from very standardised naviform cores (Binder and Balkan-Atli 2001). A preliminary count of Kaletepe productions indicates for a single ‘production unit’ (Sector P, 3.1–3.2), the fabrication of some 1500 points and maybe 4500–6000 prismatic bladelets, yielding a waste weight approaching ten tons. We can firmly state that this is far from episodic or opportunistic behaviour. The whole production needs high skills for both prismatic and naviform core shaping out, but also for the detachment of bipolar blades, for the conception of debitage algorithms and for the maintenance of the debitage rhythms.

The context of diffusion of the points is very difficult to determine on the basis of very few finds. Apart from the PPNA elements cited supra, these kinds of blades are known from Mureybet 4 (Cauvin, in press; Balkan-Atli and Der Aprahamian 1998). Nevertheless, the patterns of this debitage are very close to those which have been developed on flint in the Levant, for example at Dja’de (Coqueugniot 1994).

The cultural context of diffusion of prismatic bladelets in the Göllü Dag obsidian is homogeneous, viz. the early and middle PPNB in the Levant and Cyprus. Shillourokambos presents the exact complementary to Kaletepe sector P: there were found only central bladelets, the type of which had been selected and taken from the prismatic production of the workshop.

The ambience in which this activity was done should fit with seasonal expeditions. In fact, Kaletepe is 1600m high, located in a continental landscape where access to the sources is often difficult today from fall to spring. This constraint should have been much stronger during the Early Holocene period, with a higher contrast between seasons and annual temperature means probably lower than today.

The total lack of traces of Kaletepe sector P production in a large range in the periphery of the workshops could indicate that the productions are not only organised for, but also by some members of the Levantine PPN koine. No regional archaeological data are available to show today that obsidian diffusion could have fitted with down-the-line patterns or could have employed any relays in order to control the redistribution of goods. On the other hand, Kaletepe P excavations show that the craftsmen were not linked to a single source. They used Kaletepe as a basis for expeditions towards the Kayirli, Eriklidere or Nenezi Dag outcrops (5 to 15km): this is proven by the presence of naviform cores, introduced in Kaletepe P as preforms and knapped exactly in the same ways as Kaletepe obsidian.

A chronological chart has been prepared using the CANeW database for Central Anatolia. I have added new dates for Kaletepe sector P 3.2, (AMS dates by Dr H. Valladas, CNRS–UMR 1572: GifA–100396, 9030±80 BP; GifA–100631: 9000±90 BP.) as well as the principal references for the Middle and Upper Euphrates and Cyprus PPN. (From the BANADORA database in Lyon and from the sources listed in the bibliography.) Data yielding a standard deviation over 100 years, or lacking stratigraphical consistency have been rejected. Large standard deviations, resulting from imprecise measures or heterogeneous material, create a broadening of the chronology. Radiocarbon measurements have been calibrated with OxCal 3.5 (Bronk Ramsey 2000). For a preliminary approach I further considered as significant the intervals with the highest probability distribution, obtained after projection on the calibration curve with one standard deviation. This gives the advantage of a very coherent chronological chart.

The two AMS measurements from Kaletepe date the workshop between 8300 and 8200 cal BC, that is equivalent to the late Early PPNB horizon. At present, these represent the earliest dates for any Central Anatolian Neolithic site. They fit a small set of dates available for the Early and Middle PPNB of the Middle Euphrates (Mureybet 4a: 8560–8200 cal BC; Dja’de: 8450–8270 cal BC), for Cyprus (Shillourokambos: 8650–8400 cal BC) and for the second stage of the PPN of the Upper Euphrates (Çayönü, Channeled Buildings Subphase: 8450–8200 cal BC; Cayönü, Grill Buildings Subphase: 8480–8290 cal BC; Nevali Çori: 8550–8290 cal BC).
The following hypotheses are based upon a chronological background with a very low resolution, around two centuries. To sharpen or reject them, it is necessary to increase the number of AMS dates and to use short-lived material collected in contexts that could be connected to discrete events.

Considering previous lithic and radiometric data, new hypotheses can be proposed for understanding the scenario of the neolithisation of Central Anatolia.

Stage 1, before 8600 cal BC.

No Epipaleolithic or Neolithic traces are as yet available in Central Anatolia. The only traces are virtual ones, considering the earliest diffusion of Cappadocian obsidian in Natufian or PPNA contexts.

Stage 2, 8600–8200 cal BC.

The only Central Anatolian settlements to be taken into account are Pinarbasi A (Watkins 1996) and Kaletepe P.3.2. The radiocarbon measurements for Kaletepe and for the latest layer at Pinarbasi A (ABJ) are nearly identical, the maximum densities of their probability distributions at 1 sigma situated at 8300–8200 cal BC and 8340–8205 cal BC, respectively. As far as we know, the industry of Pinarbasi A with its microliths and narrow bladelets production has to be understood as Epipalaeolithic, linked to the Mediterranean tradition (Oküzini–Belbasi–Beldibi). One can also suppose that it has ramifications with other regions of the Anatolian Plateau on the basis of earlier surveys (e.g., Baradiz, Dervishan, Macunçay – cf. Balkan-Atli 1994). On the other hand, Kaletepe should attest specialised expeditions organised in the Levantine PPNB context, in a cultural sense and not only in the chronological one. This hypothesis strengthens the idea already discussed by Jacques Cauvin (1994), who considered the Early PPNB as a decisive moment for the spread of the Neolithic out of the Levantine Corridor.

At that moment should coexist the actors of adoption processes of the Neolithic way of life: hunters in contact with pioneer craftsmen. The presence of obsidian in Pinarbasi A could indicate that a theatre did exist for such a contact. To go further in the development of this hypothesis, it is necessary to identify the sources of obsidian used next to the flints in Pinarbasi A.

Stage 3, 8200–7500 cal BC.

Two different entities coexist in Central Anatolia: a) an original form of the Neolithic – the Asikli Culture (Esin et al. 1991), and b) the traces of an exploitation of Cappadocian obsidian sources turned towards the Levant and Cyprus. Although its interpretation is not easy, considering the stratigraphical context and the fact that it corresponds to a plateau in the calibration curve, the one date obtained for Kaletepe sector East (8200–7825 cal BC) fits very well with the earliest dates of Asikli Höyük (8200–7950 cal BC).
During stage 3, while the Levant, Cyprus and the Middle Euphrates are still receiving Göllü Dag productions, some of the latter sources are directly exploited by the Asikli people. It is necessary to insist on the fact that Asikli productions, using Nenezi Dag or Kayirli–Eriklidere obsidian, are extremely different from the prismatic or naviform Kaletepe productions (Abbès et al. 1999; Binder and Balkan-Atli 2001). Looking at current data that need of course to be confirmed, it has to be underlined that Asikli does not exploit Kaletepe outcrops: the very rare Asikli-like cores found on the surface at Kaletepe are made of Kayirli–Eriklidere transparent obsidian. Kaletepe could still have kept its Levantine and Cypriot vocations.

On the other hand, at about 8200–7800 cal BC, Kayirli–Eriklidere and Nenezi Dag constitute a place of common resources for people living in Asikli and for the knappers producing for the Middle PPNB settlements in the Levant and on Cyprus. (i.e., Shillourokambos). So we can imagine places for these different groups to meet.

It is obvious that Asikli presents, at least at the beginning, cultural traits directly related to the Epipaleolithic, and especially such geometric microliths as triangles or lunates, retouched after the use of microburin fragmentation techniques. The dates from Oküzini AH 2 (7740–7540 cal BC) and AH 3 (8200–7650 cal BC – Albrecht 1994) are also consistent with the idea of a strong cultural stability of microlithic horizons on the Mediterranean coast during the full development of the Asikli village. These microliths are reasonably associated to hunting techniques and underline well the mixed characters of the Asikli economy and way of life. During the development of Asikli, geometric microliths do not disappear. The assimilation of elements of the Neolithic package by people keeping a part of their Epipaleolithic traditions seems to be very progressive, as the funeral practices may also testify. The practice of a weakly selective breeding, not generating visible morphological modifications in stature and discrete characters on animal bones, could be understood in the same sense (cf. Vigne, Buitenhuis and Davis 1999).

Very specific is also the presence of backed blades, shouldered or tanged points that did not exist before in Epipalaeolithic contexts (Balkan-Atli et al. 2001). I propose to understand these various points as a form of assimilation of PPN models, but produced here with simplified methods (generally an abrupt direct retouch applied to various bi-directional blades). Just as in the Southeast (Astruc, in press) these tools are not to be understood only as weapons: they are multifunctional and sometimes used for plant collecting (P. Anderson, oral comm. (P. Anderson, ‘Functional interpretation of obsidian tools in Asikli (Pre-Pottery Neolithic, Central Anatolia): microscopic and experimental data’. Poster presented at the XI. Arkeometri Sonuçlari Toplantisi, Ankara, May 1995.))

The same simplification seems to characterise the laminar debitage. Just like the naviform patterns, it tends to the production of moderately arched blades using two opposed striking platforms. But on the contrary, Asikli debitage is a broad spectrum one, showing most of the time unidirectional scars: the opposite platform is mainly used for distal corrections of the longitudinal convexity of cores. The skills involved in Asikli productions are without doubt very modest compared to those employed at Kaletepe. Why not consider them as imitations of original Neolithic blade production patterns?

Stage 4, 7500–6700 cal BC.

This period may correspond with an intensification of land control by Central Anatolian Neolithic groups, and with a broadening of Neolithic ways of life. Kaletepe outcrops are now frequented by knappers related to the South and the Konya Plain. An arrowhead decorated with a geometric engraved pictogram has been found in the top levels of Kaletepe (sector P, ensemble 2; Balkan-Atli and Binder 2000). It is close to examples from Canhasan III (Ataman 1987) and Çatalhöyük pre-level XII.D (Carter, in press). Canhasan III and Kaletepe types are drawn on a type of arrowhead very popular in Musular (Özbasaran 2000; Balkan-Atli et al. 2001), perhaps indicating a renewal due to Late PPNB influences from the Levant (Cauvin 2000).

Nevertheless, if this hypothesis is to be accepted, one has to admit that these influences should have been supported by factors other than the obsidian distribution and exchange. Actually, during this period we can observe a decrease of Cappadocian obsidian to the benefit of eastern sources. This trend seems to be clear at Hallula (Molist 1996). Rather than a loss of attraction of these sources linked to changes in fashion, one can see the consequence of an increase of social and political control of Cappadocian sources held by Central Anatolian Neolithic people and the consequent break in exchange or direct procurement roads.

As shown in Çatalhöyük pre-level XII by Carter (op. cit.), it is clear that geometric microliths linked to the Epipaleolithic tradition continue till the second half of the 8th millennium cal BC, associated to bifacial points. These data give much more credence to the long duration of Epipaleolithic industries on the Mediterranean coast during the previous stages. This persistence is also established at Suberde. So what does the absence of microliths in Musular and Canhasan III mean?

What happens during this stage with the naviform blade traditions? We have observed their scarce presence, for example during the episode corresponding to the filling of the Musular channel (Özbasaran 2000; Binder and Kayacan, not published). This tradition will continue in Çatalhöyük PNA with caches of blades from Nenezi Dag, whereas Göllü Dag workshops (and obviously Kaletepe, cf. Balkan-Atli and Binder 2001) provide large bifacial pieces to Çatalhöyük in the double perspective of furnishing flake cores and bifacial tools for domestic activities. (Greetings to T. Carter who let us make some observations of the Çatalhöyük industries in his lab during the field season 1999.)

Lithic technology helps us to identify probable contacts and interaction processes at the origin of the Central Anatolian Neolithic. In Eurasia, we know other prehistoric examples indicating that the tropism towards lithic resources could generate such processes and especially the adoption of Neolithic ways of life by Epipaleolithic or Mesolithic hunters.

In the case of Central Anatolia, it will be necessary to characterise much more precisely the microlithic industries of the 10th and 9th millennia cal BC, analysing the role of obsidian in the hunters’ raw material procurement and strategies, and characterising the specific contribution of obsidian to the technical system.

Understanding the data related to the Neolithic diffusion and to neolithisation processes in Central Anatolia needs to consider a large zone, including the whole space where obvious social links are identified. Analysing the common resources used by interregional communities – as subsistence bases or as supports for maintaining social behaviour – is one way to define Neolithic social spaces or ‘cultures.’

References

Abbès, F., N. Balkan-Atli, D. Binder and M.-C. Cauvin, 1999. Étude technologique préliminaire de l’industrie lithique d’Asikli Höyük. TÜBA-AR 2, 117–137
Abbès, F., M.-C. Cauvin, B. Gratuze, L. Bellot-Gurlet, C. Bressy and G. Poupeau, in press. Nouvelles recherches sur l’obsidienne de Cheikh Hassan (Vallée de l’Euphrate, Syrie) au Néolithique: PPNA et PPNB ancien. Syria
Albrecht, G., 1994. Die frühholozänen Horizonte der Öküzini, Sondage 1989. In H. Gebel and S. Kozlowski [eds.]. Neolithic chipped stone industries in the Fertile Crescent. Proceedings in early Near Eastern production, subsistence and environment 1. Berlin: Ex Oriente, 207–208
Astruc, L., in press. Points and lustred pieces from Tell Sabi Abyad II and Tell Damishliya (Balikh valley, Northern Syria): a view from use-wear analysis. In 4th international workshop on PPN chipped stone industries, Nigde (Turkey), June 2001 (abstracts)
Ataman, C., 1987. A group of projectile points from Can Hasan III. Arastirma Sonuçlari Toplantisi 4, 339–346
Balkan-Atli, N., 1994a. La néolithisation de l’Anatolie. Paris: Institut Français d’Études Anatoliennes–De Boccard
Balkan-Atli, N., 1994b. The typological characteristics of the Asikli Höyük chipped stone industry. In H. Gebel and S. Kozlowski [eds.]. Neolithic chipped stone industries of the Fertile Crescent. Proceedings in early Near Eastern production, subsistence and environment 1 Berlin: Ex Oriente, 209-221
Balkan-Atli, N. and D. Binder (and collaborators), 2000. L’atelier néolithique de Kömürcü-Kaletepe: fouilles de 1999. Anatolia Antiqua 8, 199–214
Balkan-Atli, N. and D. Binder (and collaborators), 2001. Les ateliers de taille d’obsidienne: fouilles de Kömürcü-Kaletepe 2000. Anatolia Antiqua 9, 193–205
Balkan-Atli, N., D. Binder and M.-C. Cauvin, 1999. Obsidian: sources, workshops and trade in Central Anatolia. In M. Özdogan and N. Basgelen [eds.]. Neolithic in Turkey. The cradle of civilization. New discoveries. Istanbul: Arkeoloji ve Sanat Yayinlari, 133–145
Balkan-Atli, N and Der Aprahamian G., 1998. Les nucléus de Kaletepe et deux ateliers de taille en Cappadoce. In M.-C. Cauvin, A. Gourgaud, B. Gratuze, N. Arnaud, G. Poupeau, J.-L. Poidevin and C. Chataigner [eds.]. L’obsidienne au Proche et Moyen-Orient: du volcan à l’outil. Lyon–Oxford: BAR International series, 738 an& Archaeopress, 241–258
Balkan-Atli, N., N. Kayacan, M. Özbasaran and S. Yildirim, 2001. Variability in the Neolithic arrowheads of Central Anatolia (typological, technological and chronological aspects). In I. Caneva, C. Lemorini, D. Zampetti and P. Biagi [eds.]. Beyond tools. Redefining PPN lithic assemblages of the Levant. Berlin: Ex Oriente, 27–43
Bellot-Gurlet, L., G. Poupeau, M.-C. Cauvin, C. Chataigner, E. Coqueugniot and F. Keller, 1999. L’approvisionnement en obsidienne des sites néolithiques de la moyenne vallée de l’Euphrate entre 9000 et 8600 ans av. J.-C. Archéométrie 99 congress, Lyon, GMPCA, 21–24 April 1999 (abstracts)
Binder, D. and N. Balkan-Atli, 2001. Obsidian exploitation and blade technology at Kömürcü-Kaletepe, Central Anatolia. In I. Caneva, C. Lemorini, D. Zampetti and P. Biagi [eds.]. Beyond tools. Redefining PPN lithic assemblages of the Levant. Berlin: Ex Oriente, 1–16
Briois, F., B. Gratuze and J. Guilaine, 1997. Obsidiennes du site néolithique précéramique de Shillourokambos. Paléorient 23, 95–112
Bronk Ramsey, C., 2000. OxCal program v3.5; online: http://rlaha.ox.ac.uk/orau.html
Carter, T., in press. The earliest chipped stone industries from Çatalhöyük: context, form and significance. 4th international workshop on PPN chipped stone industries, Nigde (Turkey), June 2001 (abstracts)
Carter, T., C. Bressy and G. Poupeau, 2001. ‘People and place’: new information on technical change at Çatalhöyük (abstract). American Journal of Archaeology 105, 280
Cauvin, J., 1994. Naissance des divinités. Naissance de l’agriculture. La révolution des symboles au Néolithique. Paris: CNRS Éditions
Cauvin, M.-C., 2000. L’industrie lithique taillée de Ramad. In H. de Contenson [ed.]. Ramad: site néolithique de Damascène (Syrie) aux VIIIe et VIIe millénaires avant l’ère chrétienne. Beyrouth: Institut Français d’Archéologie au Proche-Orient, 57–97
Cauvin, M.-C., in press. L'obsidienne et sa diffusion dans le Proche-Orient néolithique. In J. Guilaine [ed.]. Matériaux, productions, circulations, du Néolithique à l'Age du Bronze. Paris: Errance
Cauvin, M.-C., A. Gourgaud, B. Gratuze, N. Arnaud, G. Poupeau, J.-L. Poidevin and C. Chataigner [eds.], 1998. L’obsidienne au Proche et Moyen-Orient: du volcan à l’outil. Lyon–Oxford: BAR International series, 738 & Archaeopress
Coqueugniot, E., 1994. L’industrie lithique de Dja’de el Mughara et le début du PPNB sur l’Euphrate syrien (sondages 1991 et 1992). In H. Gebel and S. Kozlowski [eds.]. Neolithic chipped stone industries in the Fertile Crescent Proceedings in early Near Eastern production, subsistence and environment 1. Berlin: Ex Oriente, 313–330
Esin, U., E. Biçakçi, M. Özbasaran, N. Balkan-Atli, D. Berker, I. Yagmur and A. Korkut Atli, 1991. Salvage excavations at the Pre-Pottery site of Asikli Höyük in Central Anatolia. Anatolica 17, 123–174
Molist Montana, M. [ed.], 1996. Tell Hallula (Syria): un yacimiento neolitico del valle medio del Eufrates, campanas de 1991 y 1992. Madrid: Instituto del Patrimonio Historico Espanol
Özbasaran, M., 2000. The Neolithic site of Musular, Central Anatolia. Anatolica 26, 129–151
Vigne, J.-D., H. Buitenhuis and S. Davis, 1999. Les premiers pas de la domestication animale à l’ouest de l’Euphrate: Chypre et l’Anatolie Centrale. Paléorient 25, 49–62
Watkins, T., 1996. Excavations at Pinarbasi: the early stages. In I. Hodder [ed.]. On the surface: Çatalhöyük 1993–95. Cambridge and London: McDonald Archaeological Institute and British Institute at Ankara, 47–57

__________________________

Go to the related debate (Panel : Nur Balkan-Atl�, Mehmet �zdoğan, Isabella Caneva, Harald Hauptmann)