Précédent Suivant

Obsidian tool production in the South Caucasus of the 5th and 4th millennium BCE

Technology, typology and sociocultural implications

p. 387-406

Résumés

The ongoing archaeological investigations carried out in Nakhchivan (with the excavations at Ovçular Tepesi and Kültepe I and regional survey) and in eastern Georgia (with the excavations at Sakdrisi and Dzedzvebi) within the joint research project “On salt, copper and gold: the origins of early mining in the Caucasus”, have provided new and important information on the Chalcolithic and Early Bronze Age Kura-Araxes cultural sequences. One major question of this joint research program is how the social structure behind technological transfers and the developments towards specialization in production and raw material exploitation during the 5th to early 3rd millennium BC can be described. Hence, these investigations do not only examine mining and the production of metal or salt, or changes in pottery manufacture (where indeed a striking difference between Late Chalcolithic and Kura-Araxes ceramic productions is to be emphasized), but also the obsidian and flint tool industries in terms of their sociotechnological role.
Thus, the following paper focuses on the lithic collections from the sites under investigation, which date to the Late Chalcolithic and Kura-Araxes periods, between the 5th to early 3rd millennium BCE. Since the regions under study are located near the major obsidian sources of the South Caucasus (Lake Sevan), the lithic collections are self-evidently predominated by a local obsidian industry, for which distinctive artefact categories and technological tool types can be defined. Additionally, the existence of other “traditional obsidian networks”, which started to develop during the PPN, can be demonstrated. Some of these networks have very distinct specificities and seem to have lasted until the Early Bronze Age. This is the case for example with the large blade technology or the bifacial arrowheads that link the South Caucasus with North Mesopotamia.

Les travaux archéologiques en cours au Nakhchivan (fouilles d’Ovçular Tepesi, de Kültepe I et prospection régionale) et en Géorgie orientale (fouilles de Sakdrisi et Dzedzvebi), menés dans le cadre du programme de recherche intitulé « Du sel, du cuivre et de l’or : origines et développement des industries minières au Caucase », ont fourni des données nouvelles et importantes sur les séquences culturelles du Chalcolithique et du Bronze ancien kuro-araxes. Une des questions fondamentales abordées par ces recherches conjointes concerne la structure sociale sous-jacente aux phénomènes de transferts technologiques et de spécialisation, qu’il s’agisse de l’exploitation des matières premières ou de la production, observés entre le Ve et le IIIe millénaires av. J.‑C. De ce fait, ces travaux ne s’intéressent pas seulement aux industries minières et à la production de métal ou de sel, ou aux évolutions dans la fabrication des céramiques (où il faut certainement souligner les différences frappantes entre les céramiques du Chalcolithique récent et kuro-araxes), mais portent aussi sur l’étude des outils en obsidienne et en silex, afin de comprendre leur rôle socio-technologique.
Cet article se concentre donc sur les collections lithiques des sites susmentionnés, lesquelles datent des périodes chalcolithique et kuro-araxe entre le Ve et le IIIe millénaires avant notre ère. Étant donné que ces régions sont situées à proximité des principaux lits d’obsidienne du sud du Caucase (lac Sevan), les collections lithiques sont bien sûr dominées par une industrie en obsidienne locale, dans laquelle des catégories d’artefacts et des types d’outils bien spécifiques peuvent être définis. Par ailleurs, la persistance de « réseaux d’obsidienne traditionnels », en place depuis le PPN, peut aussi être démontrée. Certains de ces réseaux ont des spécificités bien distinctes qui semblent avoir duré jusqu’au Bronze ancien. C’est le cas par exemple de la technologie des grandes lames ou des pointes de flèches bifaces, qui permettent de relier le sud du Caucase à la Mésopotamie du Nord.


Texte intégral

Introduction

1The following study introduces several chipped stone assemblages that were found on three Late Chalcolithic and Early Bronze Age settlements in two distinctive landscapes of the Araxes and Kura regions:1 Ovçular Tepesi and Kültepe 1 (both located in Nakhchivan, Azerbaijan), and the vicinity of the Sakdrisi gold mine in Georgia (fig. 1). All the inventories can be characterized as veritable obsidian industries – apart from a small percentage of flint artefacts – which are obviously not uncommon in the region under discussion. The major obsidian sources of the Lesser Caucasus that were exploited during prehistoric times are located around Lake Sevan in Armenia: Paravani, Siyoni, in the mountains of the Aragats, as well as around Gegamy, Geghasar and Pokr Arteni in southern and western Armenia (Badalyan, Chataigner, Kohl 2004; Blackman 1984; Chataigner, Gratuze 2014; Khademi Nadooshan et al. 2013, fig. 1; Frahm, Campbell, Healey 2016; fig. 2). For Georgia, the deposits of Chikiani (southern Georgia) may have been important in prehistory.2 One should also mention the beds of eastern Anatolia (Bayazıt, Lake Van; Nemrut Dağ; Bingöl), whose obsidian was distributed in the Iranian North-West and the Zagros as far as the Persian Gulf shore (Chataigner 1995; Kushnavera 1997, pp. 176‑178, fig. 71). But the deposits of Cappadocia (Acıgol, Çiftlik und Karakören) in Central Anatolia should also be considered when discussing the Chalcolithic and EBA networks of western Asia and the Caucasus.

Fig. 1 – The regions under study; 1: the Georgian Sakdrisi valley in the north; the Nakhichevani sites; 3: Ovçular Tepesi; 4:  Kültepe I; and other key sites; 2: Aratashen; 5: Ginchi; 6: Djolfa; 7: Geoy; 8: Yanik Tappeh; 9: Sialk; 10‑13: Norşuntepe, Lidar Höyük, Tepecik (J. Thomalsky).

Image

Fig. 2 – Major obsidian sources of the region located in Armenia, around Lake Sevan, eastern Anatolia, around Lake Van, and further west in Central Anatolia; further possible sources are debated in north-western Iran, around Mound Sahand in the north-western Iran and Azerbaijan Province (J. Thomalsky).

Image

The obsidian industry from Ovçular Tepesi, Nakhchivan

2Ovçular Tepesi in Nakhchivan3 is located on a natural hill bordering the Middle-Araxes valley; its archaeological layers are up to 1.5 to 2.0 m thick. The recently completed archaeological excavations at the site identified two major occupation periods, dated to the Late Chalcolithic and the Early Bronze Age, with associated radiocarbon data that gives an occupation time span ranging from the middle of the 5th millennium BC to ca 2500 BC (Marro, Bakhshaliyev, Ashurov 2009, 2011, p. 79). The Late Chalcolithic occupation consists of a first phase with semi-subterranean buildings, followed by rectangular mud-brick houses. Storage features and working devices, such as silos, bonded clay circles (dials), and earth basins in courtyards are attested. Beyond these, one Kura-Araxes pit and one Kura-Araxes grave are embedded or even sandwiched within the LC horizon, but bear no association with any building. One such assemblage produced radiocarbon data that give a terminus ante quem for the Kura-Araxes presence on this site ca 4230-3940 BC (Marro, Bakhshaliyev, Ashurov 2011, p. 66). The uppermost EBA occupation is represented by four half-preserved circular buildings, one rectangular building, several open hearths, two tombs, ashy fills and thin pisé walls; a horizon that was destroyed and bulldozed away for the erection of a Russian radio mast.

3The occupation of the site appears to have been half-permanent, as suggested by certain elements in the stratigraphy, in particular the location of the work areas. It is above all clear that Late Chalcolithic Kura-Araxes groups came to the site when it was seasonally abandoned by “Late Chalcolithic occupants” (Marro, Bakhshaliyev, Berthon 2014, 2015). Similar situations that demonstrate a coexistence of different groups in the same area are attested on sites in the Malatya and Elazığ valleys in eastern Anatolia, e.g. Norşuntepe, Arslantepe VIA. The pottery of these two groups, respectively the “Chaff-Faced Ware” and the classic Kura-Araxes dark-burnished ware are strongly divergent in terms of production procedures and technological attributes. This raises the question whether such distinctive differences are perceptible in other production fields, in the lithic inventories of respectively the Late Chalcolithic and the Early Bronze Age.

4The examination of the lithic inventory of Ovçular Tepesi comprises more than 5,000 pieces that may fuel this discussion. The obsidian assemblage makes up 90% of the total bulk, in which knapping debris such as core preparation and miscellaneous flakes clearly predominate (fig. 3a). Only a small amount is composed of flint, but it should be noted that this material was also brought to the site in larger nodules for further processing (fig. 3b).

Fig. 3a – Ovçular Tepesi, obsidian raw material pieces (J. Thomalsky).

Image

Fig. 3b – Ovçular Tepesi, flint material pieces (J. Thomalsky).

Image

5As concerns the obsidian, a large number of flakes exhibit traces of detached or splintered platforms, which clearly identify these pieces as core remnants (fig. 4). To this category of core remains also belong common core tips and core flanks that still exhibit lines of negative retouch patterns. Such pieces clearly demonstrate that almost every (blade) core was chipped up into a little lump, or broken by mis-chipping events, removed from the cores by failed or accidental impacts. This fact makes it difficult to identify the original number of cores in the assemblage but they obviously exist, in the form of platform flakes, core flanks and tips.

Fig. 4 – Ovçular Tepesi, Late Chalcolithic and EBA obsidian industry; core remnants, splintered pieces, etc. (J. Thomalsky).

Image

6Further to be added to this category of debris are the so-called splintered pieces (fig. 5.9‑13). Splintered pieces are flakes with unifacial or bifacial modification that are the by-products of the bipolar knapping technique. They often still exhibit cortex residues on the surfaces. In a more formal description, splintered pieces bear a certain similarity with bipolar cores. It was first K. Schmidt (Schmidt 1996, pp. 45‑46), who defined this technical group at Norşuntepe (eastern Anatolia), as true tools with invasive bifacial retouches along the edges so that the short edge would become a sort of wedge, modified from both faces. These tools may have been used as stronger chisels for bone- and wood-working.4 As for their dating, Schmidt demonstrated that the variations in their amounts from the LC to the EBA layers of Norşuntepe had no real significance. But on the basis of a more comprehensive study, he claimed that splintered pieces could be regarded as typical of West Asian assemblages from the Palaeolithic to the Late Neolithic and Early Chalcolithic. Their number clearly decreases, however, during the Early Chalcolithic.

Fig. 5 – Ovçular Tepesi, retouched flakes and splintered pieces (J. Thomalsky).

Image

7Nevertheless, the splintered pieces from Ovçular belong to a distinctive obsidian knapping technology that produces flakes with an invasive reduction on one or even on two sides. They may also be regarded as bipolar core remnants. Only a small number of them shows a strong, wedge-shaped edge that may have been used as a tool edge.

8Another very distinctive flake category that was further modified into tools are notched and denticulated pieces, produced in a very haphazard and ad hoc manner (fig. 5.1‑8, 5.14‑16). This tool group should be considered as a non-specific category as concerns its technological attributes – albeit very characteristic of the Neolithic technology of western Asia – but it is still present in small numbers in the Late Chalcolithic ad hoc household production.5 As concerns their function, one can generally place them in the working range of leather, shoemaking and possibly textiles.

Obsidian blade retouch patterns: a blade typology of the South-Caucasian Chalcolithic

9Typical modifications of the blades are nibbled or denticulated edges, which in fact result from heavy usage rather than true retouch operations. Beside these functional modifications, true blade types with distinctive retouch patterns are also attested; they have also been identified in other Chalcolithic inventories of the South Caucasus (Thomalsky 2011). Five types can be described as follows:

  • bilateral steeply retouched blades (fig. 6.1‑5): they are prepared from more irregular blade blanks and even from crested blades (artefact of core preparation);
  • lamellar retouches (fig. 6.6): these are blades with deliberately invasive lamellar modification, which mostly cover the whole face of the tool; this retouch pattern usually occurs on larger obsidian blades and may be regarded as a very specific local tool, typical of the Late Chalcolithic South Caucasus (see below);
  • inverse facial retouched tools (fig. 6.7‑8);
  • facial and bifacial retouched arrowheads with short tangs (fig. 6.14‑15): these are typical shapes of the LC and EBA in the South Caucasus and eastern Anatolia (Tepecik type) with bifacial retouches; together with specimens displaying very delicate facial retouches (mid-row) and more irregular shapes and roughly made pieces, which possibly also mirror different functional aspects, as exemplified by;
  • geometric pieces (fig. 6.9‑11): the latter were probably hafted into the shaft ends of the arrow, not at the tip.6

Fig. 6 – Ovçular Tepesi, obsidian tool types; 1‑5: (type 1) bilateral backed blades; 6: (type 2) lamellar retouched blades; 7‑8: (type 3) facial retouched blades; 14‑15: (type 4) facial/bifacial retouched arrow heads; 9‑11: (type 5) geometric implements (J. Thomalsky).

Image

The obsidian large blades

10Apart from the flake tools and blades described above, a very significant and specialized industry at Ovçular Tepesi is that of the large blade production. In fact, specific blade tool types are usually made from this large blade industry. The aim of this obsidian industry is to obtain a regular blade that attains the dimensions of a large blade, up to 41 cm in length (fig. 7).

Fig. 7 – Obsidian large blade from Ovçular Tepesi, Late Chalcolithic layers (J. Thomalsky).

Image

11Several technological attributes of this large obsidian blade industry compare closely with a specific and meticulously described lithic industry of the 4th and 3rd millennium BC attested in northern Mesopotamia and the Levant, which is called the “Canaanean blade” or “Canaanean blade technology” (hereafter abbreviated as “CBT”).7 The CBT is per definition made from flint material in the shape of large flint nodules or tabular blocks adapted to the large dimensions of the blades, from which large blade cores were prepared with a particular technology: a piece of raw material was first shaped into a rectangular rough-out, then prepared to form a straight core platform; the blades were split-off in a semi-circular or circumferential way from the core with the aid of a specific pressure technique.8 The core preparation was a highly specialized task whose aim was to produce a large number of blades of equal length and width without further rejuvenating the core by putting pressure at an angle between 85­‑90°C. The result was a long blade with straight edges, whose width was generally greater than 22 mm; the butt could be small and straight or larger and facetted. This Canaanean technology soon dominated the lithic production; then it started to be produced only in a few specialised workshops, while the majority of the large blades were used as harvesting implements (sickles and threshing sledge inlays). The Canaanean technology was so far considered to have first developed in south-eastern Anatolia and along the Upper Euphrates during the 4th millennium BC; it was widely distributed and went eastwards as far as north-eastern Iran, and westwards to the southern Levant.9

12But the large obsidian blades from Ovçular Tepesi imply the use of a similar chaîne opératoire or reduction scheme, and may also be analyzed as an example of CBT: these large blades exhibit straight edges and widths greater than 25 mm, the butts are either small, or large and triangular, splintered or concave, indicating the use of a direct punch technique that was not very standardized (yet). The distal end of the blades is fairly curved, sometimes displaying hinge fractures. On the other hand, there is almost no splintering or fractures (which are, by contrast, very characteristic of obsidian flakes) that may indicate a deliberate blow on the core. In fact, a very slow and controlled reduction method was used, leaving only a few traces of stress over the material surface. It is thus very likely that the lever-pressure technique was used as the operating technology, a technique that was similar to that used for the CBT. The large blade industry of Ovçular Tepesi does not seem to be as standardized in its knapping technology as the later “true” CBT, but this may also be inherent in the specific properties of the material, since obsidian is much more brittle than flint.

13Unfortunately, there is no good evidence for demonstrating the existence of a large-blade on-site production at Ovçular Tepesi, since only 1 or 2 large-blade cores were found during the excavations. Moreover, the evidence of the primary industry was destroyed by the opportunistic obsidian knapping that took place on site; this includes the original large blade cores that were knapped up to small pieces.

14As concerns the stratigraphic distribution or occupation area, the evidence is even less clear: large obsidian blades are mostly attested in trenches 5, 6 and 11, 12, 13, which provided an equally large number of chipped stones (fig. 8). The study of the spatial distribution of the large blades throughout the settlement does not produce a more regular picture: large blades occur more or less in isolation throughout the loci (tab. 1); they do not cluster or appear grouped in deposits, they are not related to any special feature either. Curiously enough, the obsidian large blades so far appear to be restricted to Late Chalcolithic contexts, they rarely occur in the upper layers. A corresponding detailed analysis of the stratigraphy is in progress.

Fig. 8 – Ovçular Tepesi, the distribution of large blades per trench.

Image

Tab. 1 – Ovçular Tepesi, large blades per locus (average 1 specimen).

Locus (SU)Total no.Locus (SU)Total no.Locus (SU)Total no.
50071110462130011
50471110551130142
50732110581130311
50761110831130341
50981111061130401
51081111461130421
51111111531130442
51501112281130541
51651112592130611
51822112731131701
52411112901132372
53671113061200231
60721120051200671
61311120111201571
62951120141201661
63331120481210031
63741120721230022
63881120951230111
63931121031230191
64001121061230581
64431121091230601
64711121111240141
64781121151  
101201121161  
101441121281  
102461    
102711    

The flint industry of Ovçular Tepesi

15Next to the bulk of obsidian material, a small quantity of chipped stone artifacts made of flint is also attested; again it is dominated by flakes. Only a few tools are present in our collection: (1) retouched blades, that we may call “sickle blades”, since most of them display glossy edges; and (2) a few geometric pieces (fig. 9). A small quantity of retouched flakes, denticulated and notched pieces are also present. Preliminary analyses indicate that the flint assemblage mostly stems from the uppermost layers of Ovçular, that is from the Early Bronze Age occupation levels. The only exceptions are trenches 5 and 6, which yielded 58 pieces in total – including a refittable core-preparation flake series.

Fig. 9 – Ovçular Tepesi, flint tools (J. Thomalsky).

Image

The technology of the Late Chalcolithic Kura-Araxes lithic industry at Ovçular Tepesi

16First of all, the lithic collection of Ovçular Tepesi is clearly dominated by the processing of obsidian, with a distinctive knapping technology that has been adapted to the specific mechanic properties of this raw material. Flint is underrepresented, but seems to increase in the latest occupation phase, that is in the Early Bronze Age. As concerns the obsidian industry, four major artefact groups may be distinguished for both the Late Chalcolithic and the Kura-Araxes assemblages:

  • a flake-blade production that uses hard percussion, direct or indirect, and produces either large and straight or splintered butts on flakes and blades; a few flake tools also occur as a by-product of this knapping technique; related cores show several irregular flaking faces;
  • a bipolar industry, which produced unifacial or bifacial retouched artefacts; the bulk of this artefact category consists of flaking debris such as splintered core flanks with one or two splintered edges (formerly core platform edges) or tool fragments;
  • a large blade industry, for which almost no evidence of on-site production is attested, since only a small number of large blade cores were found during the excavations; this may be caused by the non-permanent occupation of the site since raw materials and goods were generally carried along with the community; as a result, only a few but very specific tool types can be described, such as the lamellar and the facial/bifacial retouch patterns; the large obsidian blade industry further carries specific attributes that places it technologically near, or even as a forerunner of the (much later!) Canaanean flint blades of North Mesopotamia and the Levant;
  • finally, the existence of a flint industry with a very different chaîne opératoire can be demonstrated, though it produced a quantity of flakes equivalent to that of the obsidian industry; flint was obviously processed and knapped in a different way; its characteristic end product is a harvesting implement that exhibits glossy edges and traces of bitumen; different raw material deposits must have been exploited, possibly in association with changes in the modes of mobility of the Ovçular community (e.g. routes and networks); there seems to be an increase in the proportion of flint in the later levels, but this is without real statistical value since the general number of artefacts in these loci is rather limited.

17To sum up, there are no clear differences between the Late Chalcolithic and the Kura-Araxes technologies at Ovçular Tepesi. Far from it, the lithic inventories of almost all loci are very heterogeneous and rather appear as the remains of a mobile community that comes and goes, which left neither a complete set of blanks, nor a complete tool kit on the site. Only a few complete knapping inventories demonstrate the existence of an uninterrupted working sequence (chaîne opératoire), which started from the reduction of the raw material to the knapping of a series of blades or flakes. Such events can be traced back for both flint and obsidian, since the remains of cores and flakes are attested in almost every context (locus), where they can be easily refitted. In fact, the Kura-Araxes contexts have yielded much less material than the Late Chalcolithic layers because of the smaller number of preserved contexts. The only indication of a rising new technological strategy is the gradual appearance of flint processing, which may be related to the Kura-Araxes contexts. The occurrence of flint may further indicate a change in the raw-material exploitation strategy and thus a “new radius of mobility”. But since the stratigraphic and contextual analyses are still in progress, these suppositions should remain preliminary.

The LC and EBA lithic industry of the Mashavera valley in Georgia

18The second region under study is the Mashavera valley in Georgia, which has become famous because of the prehistoric gold mine of Sakdrisi. The gold mine and its vicinity are still under investigation by a Georgian-German archaeological team (under the direction of I. Gambashidze and T. Stöllner). The lithic assemblage from the Mashavera valley was collected from different find-spots that go under the name of the Balitshi-Dzedzebi occupation: they have yielded some material that may be associated with the Leyla/Sioni sequences of the 4th; but the bulk of the preserved remains includes houses, working places and metallurgical activities, Kura-Araxes pottery and tools, which date to the Early Bronze Age (end of the 4th and 3rd millennium BC).

19Generally speaking, the lithic collections are small in number, since they proceed from surveys and limited soundings. No clusters of artefacts or links with specific contexts or locations are perceptible. Instead, the material appears to come mostly from domestic layers such as floors or dump pits. The examination of the different categories of artefacts supports this impression. Broken tools and chipping debris predominate in the different find-spots; the situation at Dzedzebi is very similar to that perceptible at Ovçular Tepesi: the major chipping categories are flakes and small core residues, splintered pieces, and also modified flakes or core tablets.

20It is evident that the same chipping technology is attested in the Mashavera and in the LC and EBA Ovçular Tepesi collections. Furthermore, we may say that the Mashavera tools exhibit close ties with the Nakhchivan region, in that (1) a significant type is the larger blade with a delicate lamellar retouch pattern along one edge, inverse or covering the whole face (fig. 10a.1‑10); and (2) a specific range of bifacially retouched arrowhead types were produced from two raw-material groups, obsidian and flint (fig. 10a.11‑13). Their typological range appears to be very limited and can be compared with characteristic types of the Late Chalcolithic and Early Bronze Age key sites that are located in eastern Anatolia near the Lake Van, Nemrut Dağ and Bingöl obsidian sources (e.g. Norşuntepe, Tepecik) (Schmidt 1996, pp. 71‑77).

Fig. 10a – Mashavera, EBA tools; 1‑12: obsidian; 13: flint arrowhead (J. Thomalsky).

Image

21Lastly a third, very distinctive, tool group of the Mashavera valley should be noted: it comprises bifacial tools that are made from flint. Almost all pieces can be described as bifacially retouched sickle implements, with deep, denticulated working edges (fig. 10b). These pieces represent a complete new category in the Bronze Age lithic industry, since they differ from the “common” blade industry in that they are crafted according to a bifacial technique, and are closely linked to a specific function just like the harvesting implements. The nearest comparisons are attested in Bronze Age Colchis, e.g. from the EBA/MBA settlement of Pichori (Pkhakadze, Baramidze 2008, fig. 16.1‑7). Other analogies of bifacial sickles are difficult to find; only one similar specimen is published from the Norşuntepe collection (Schmidt 1996, no. 780‑781). Since no more detailed information is available as concerns these tools, their origin and diffusion pattern, judging by the date given for the Mashavera sites, one can simply assume that we are dealing with a Bronze Age type.

Fig. 10b – Mashavera, EBA flint tools (J. Thomalsky).

Image

22However, the Mashavera lithic inventories reflect the establishment of two significantly different technologies: one Chalcolithic tradition and a new one, that have been coexisting for a certain period. These are (1) the “common” obsidian blade industry, and (2) the bifacial flint industry. The latter can possibly be regarded as the local result of an expanded use of the rare flint material, which was originally knapped for the production of sickles. A similar change is attested in the Norşuntepe assemblage, where the production and use of flint (quartzite) is greatly increased from the EBA period onwards in order to produce the large blades needed for the crafting of harvesting tools. It is not clear yet whether the changes (exploitation of a new type of raw material in order to obtain better material properties) that took place in the lithic industry during the transition from the Late Chalcolithic to the Early Bronze Age were deliberate.

Further comparisons with the chipped stone industries of eastern Anatolia

23Hence, the existence of strong ties between the flint tools and the arrowheads of the South Caucasus and their analogies in eastern Anatolia during the EBA demonstrates the existence of “Kura-Araxes lithic types”. But once more, the importance of the large obsidian blades from Nakhchivan should be stressed.

24The obsidian blades of the late 5th millennium BC from Ovçular Tepesi suggest even stronger relationships with the Upper Euphrates region and with its Canaanean blade technology in the background. As presumed, we may consider the manufacturing processes of the large obsidian blades to be similar to the Canaanean blade technology, with only one significant difference, which is the raw material itself, i.e. the obsidian used in the northern inventories. Moreover, these obsidian specimens appear as a possible technological forerunner of the later flint blades (fig. 11). Canaanean blades – produced in flint/chert materials – developed along the Upper Euphrates during the 4th millennium and appeared considerably later during the 3rd millennium in the southern Levant.10 K. Schmidt described several types of Canaanean blade tools from Norşuntepe and other sites in the region, recognizable from specific edge modifications (Schmidt 1996, pp. 60, 63). Large flint blades with a curvilinear back that exhibits an inverse steep lamellar retouch and an opposite straight working edge were thereafter determined as “type Lidar” and “type Norşuntepe” (fig. 12). Both large blade types fulfil the function of a sickle knife that was hafted along the steep back. Both types seem to disappear during the mid 3rd millennium BC at the latest.

Fig. 11 – Distribution of large Canaanean chert blades during the later 5th (red) and 4th (yellow) millennium BCE in south-eastern Anatolia and northern Mesopotamia; black triangles: obsidian sources, red star: Ovçular Tepesi, yellow star: Kültepe 1 (J. Thomalsky).

Image

Fig. 12 – Canaanean blades from Norşuntepe (after Schmidt 1996, fig. 59).

Image

25From a longue durée perspective, large blades exhibiting an incipient pressure-blade technology first occurred in northern Mesopotamia and on the Anatolian Plateau during the Early Pre-Pottery Neolithic, as exemplified by the obsidian prismatic blade production from Cappadocian workshops, as well as from Çayönü.11 The latter stands in contrast to the naviform core technology of the PPNB‑sphere and is further clearly associated with the production of a distinctive tool type, the Çayönü tool. Their geographical and temporal distribution pattern appear to be closely associated with a possible contact route towards the east (the Iranian Zagros, north-western Iran, and the South Caucasus) in use between the 9th and the 6th millennium BC.12 Recently, Chabot and Pelegrin traced a similar route and suggested a possible technological link between the above-mentioned Anatolian Neolithic large obsidian blades and the Neolithic large obsidian blades of the South Caucasus, which were produced by the lever technique (e.g. Aratashen) (Chabot, Pelegrin 2012, p. 187).

26However, a somewhat larger “missing link” between the earliest evidence of pressure and lever techniques (later adapted to obsidian large blades) and the Chalcolithic Canaanean flint blade industry should be discussed. The aim of this paper is not to trace the earliest occurrences of large blades, but to emphasize the typo-chronological link between the large obsidian blades with steep retouches from Ovçular Tepesi and the comparable specimens in flint attested at Norşuntepe, which for the latter are related to the Canaanean technology. Just like the Canaanean technology, which was originally described for the Levantine and northern Mesopotamian industries of the 4th and 3rd millenium BC, the technology employed at Ovçular may be described by its specificities in core preparation and its flaking pattern (Thomalsky 2012, p. 273): the piece of raw material is expertly prepared by removing only a few targeted (cortex-bearing) flakes through direct hammer technique, which produces a straight flaking platform and flaking face. The optimum exploitation of the material should be emphasized here: the cores do not need to be constantly rejuvenated to produce new platforms or flaking faces, as is the case for other blade technologies. The subsequent blade reduction reveals indirect flaking with a blow, where the flake is detached from the core by the pressure method with a lever. The basal ends of the blades exhibit so-called flaking platform residues, which demonstrate a technological standardization from broader straight or facetted residues to punctiform residues in the course of the 4th to 3rd millennium BC. It appears that the Canaanean technology later followed a similar pattern of development, and may be regarded as a technological innovation of the Chalcolithic period, inherited from the 5th millennium BC.

27In short, our point is not to discuss the emergence of large blades industries and pressure/lever technique as a whole, but the development of a more distinctive type: the large obsidian blade with steep lamellar retouches and an opposite straight working edge. At Ovçular, several pieces of this type are attested (fig. 13). In point of fact, we may venture to say that the obsidian large blade industry of Nakhchivan is the earliest “Canaanean blade technology” known to date; this includes the very first specific tool types that first appeared in the region. Interestingly enough, an early preliminary analysis shows that large blades with this specific retouch pattern do not occur before Phase II (Late Chalcolithic occupation with some Kura-Araxes elements) at Ovçular Tepesi. This suggests that this particular tool type may be linked to or correlated with the Kura-Araxes occupation of the site and the wider region. One may assume that the Canaanean technology was first locally developed on obsidian, but later adapted to other stone material (flint), since obsidian deposits are geographically limited to certain areas. The evidence from Nakhchivan suggests that Kura-Araxes groups may be regarded as the bearers and carriers of this innovation in lithic technology that further underwent a change of raw material usage. One step further in presumption: their further diffusion and the adoption of the Canaanean blade technology outside the highlands strikingly parallel the evidence of the Kura-Araxes expansion (fig. 14).

Fig. 13 – Large obsidian blades in Canaanean technology from Ovçular Tepesi (MBA, J. Thomalsky).

Image

Fig. 14 – Expansion of the Kura-Araxes phenomenon (after Rothman) correlated with the Canaanean large blade technology during the 4th and 3rd millennium BC.

Image

28If we compare the development of the large blade technology with the diffusion of the Kura-Araxes culture towards the south (Batiuk, Rothman 2007; Kohl 2009), it might be interesting to look for further technological similarities in lithic technologies between the Caucasus and eastern Anatolia.

29Unfortunately, there is only one well-known LC/EBA lithic inventory available for close comparisons with our South Caucasian assemblages: the Chalcolithic and Early Bronze Age sequence of Norşuntepe on the Turkish Euphrates in the Altınova plain, where an early Kura-Araxes component is also attested (Schmidt 1996; Gülçur 2000). The lithic material of the Chalcolithic levels at Norşuntepe is also clearly dominated by an obsidian industry (90% – 80% of the total bulk). Considering the locally available obsidian sources in the region of Bingöl, this does not come as a surprise. We may note, however, a striking change in the procurement strategy of the raw materials, as concerns the different varieties and sources of obsidian. In the lower Chalcolithic levels, a brown translucent variety predominates in the obsidian industry, which may correspond to sources located either in Central Cappadocia or around Erevan in Armenia.13 During Phase III (or LC 2), a greenish translucent variety attributed to the local eastern Anatolian sources of Bingöl and the Nemrut Dağ, constitutes the majority of the finds and thus demonstrates a shift towards the intensification of a local exploitation and production on site. Nevertheless, at the very end of the LC towards the transition to EBA, a second shift in the exploitation of the obsidian deposits is perceptible: the greenish obsidian was again replaced by the brown, grey and black varieties. In the meantime, a general but moderate decline of the obsidian industry correlates with a slowly growing flint industry. As concerns the knapping technology, the similarity in flake attributes and categories of the obsidian industry with the Ovçular obsidian collection has already been emphasized above. Beyond that, Norşuntepe yields a large inventory of bifacial arrowheads. As concerns the more irregular geometrical pieces from Ovçular, analogies with the Urfa region may also be found (Schmidt 1996, fig. 46). However, the striking difference is that the Norşuntepe large blades (of CBT) are made of flint/quartzite. The occurrence of the Canaanean blade technology first appeared during the Late Chalcolithic, but seems to have developed as a true innovation only during the Early Bronze Age. Not only here but throughout North Mesopotamia down to the southern Levant, these large blades were almost exclusively used as sickle implements and can be regarded as the first phenomenon of economic mass production in lithic industries.

Hypothesis: a Kura-Araxes lithic industry?

30There are striking links in the lithic (obsidian) technology between the South Caucasus and the Upper Euphrates (Norşuntepe), as demonstrated by the overall predominance of core residues and splintered pieces. Beyond this, several sophisticated tool types or retouch patterns indicate more local and/or chronological developments, which are illustrated by the facial retouch patterns from the Late Chalcolithic Middle-Araxes basin or the bifacial sickles from the Mashavera region during the Early Bronze Age.

31Moreover, from the late 5th millennium onwards, we can trace back the establishment of a new blade technology, made of obsidian, that was distributed from a core area, defined by the South Caucasus and eastern Anatolia, where it seems to be closely linked with the local obsidian deposits, towards North Mesopotamia, where it was adapted several centuries later to the flint technology. This diffusion lasted until the early 3rd millennium, when the Canaanean blade technique appears in the southern Levant (e.g. Byblos, Arad). It appears more and more likely that this new technology was carried on by Kura-Araxes groups via the Upper Euphrates region as far as the southern Levant, where a specific red-black burnished pottery (the Khirbet Kerak ware) appears during the earlier 3rd millennium BC, obviously concurrent with the appearance of the large flint blade industry (CBT).

Bibliographie

Des DOI sont automatiquement ajoutés aux références bibliographiques par Bilbo, l’outil d’annotation bibliographique d’OpenEdition. Ces références bibliographiques peuvent être téléchargées dans les formats APA, Chicago et MLA.

Abe, Azizi Kharanaghi 2015: M. Abe, A. Azizi Kharanaghi, “A study on the Early Pottery Neolithic chipped stone assemblage from Rahamatabad”, in H.A. Kharanaghi, M. Khanipour, R. Naseri (dir.), Proceedings of the International Congress of Young Archaeologist, Tehran, University of Tehran Press, 2015, pp. 27‑40.

Altinbilek-Algul et al. 2012: Ç. Altinbilek-Algul, L. Astruc, D. Binder, J. Pelegrin, “Pressure blade production with a lever in the Early and Late Neolithic of the Near-East”, in P. Desrosiers (dir.), The emergence of pressure blade making from origin to modern experimentation, Westmount, Springer, 2012, pp. 157‑179.

Badalyan 2010: R. Badalyan, “Obsidian of the South Caucasus: the use of raw materials in the Neolithic to Early Iron Age”, in S. Hansen, A. Hauptmann, I. Motzenbäcker, E. Pernicka (dir.), Von Majkop bis Trialeti. Gewinnung und Verbreitung von Metallen und Obsidian in Kaukasien im 4.‑2. Jt. v. Chr., Bonn, Rudolf Habelt, 2010, pp. 27‑38.

Badalyan, Chataigner, Kohl 2004: R. Badalyan, C. Chataigner, P. Kohl, “Trans-Caucasian obsidian: the exploitation of the sources and their distribution”, in A. Sagona (dir.), A view from the highlands. Archaeological studies in honour of Charles Burney, Leuven, Peeters, 2004, pp. 437‑465.

Balkan-Atli 2003: N. Balkan-Atli, “Use of obsidian at Değirmentepe: an Ubaid settlement in eastern Anatolia”, in M. Ozdogan, H. Hauptmann, N. Başgelen (dir.), From villages to towns. Studies presented to Ufuk Esin, Istanbul, Arkeoloji ve Sanat Publications, 2003, pp. 373‑384.

Batiuk, Rothman 2007: S. Batiuk, M. Rothman, “Early Transcaucasian cultures and their neighbors: unraveling migration, trade, and assimilation”, Expedition 49/1, 2007, pp. 7‑17.

Behm-Blancke, Boese 2001: M.R. Behm-Blancke, J. Boese, “Zu spätchalkolithischen Erntegeräten in Nordsyrien und Südostanatolien”, in R.M. Boehmer, J. Maran (dir.), Lux Orientis. Archäologie zwischen Asien und Europa. Festschrift für Harald Hauptmann zum 65. Geburtstag, Rahden, Leidorf, 2001, pp. 27‑37.

Behm-Blancke et al. 1992: M.R. Behm-Blancke, F. Collin, J.M. Leotard, M. Otte, J. Pelegrin, J. Weiner, “Die lithische Industrie (Hornstein und Obsidian)”, in M.R. Behm-Blancke (dir.), Hassek Höyük. Naturwissenschaftliche Untersuchungen und lithische Industrie, Istanbuler Forschungen 38, Berlin, Wasmut, 1992, pp. 165‑260.

Bernbeck, Pollock, Coursey 1999: R. Bernbeck, S. Pollock, C. Coursey, “The Halaf settlement at Kazane Hoyük”, Anatolica 25, 1999, pp. 109‑147.

Blackman 1984: M.J. Blackman, “Provenience studies of Middle Eastern obsidian from sites in highland Iran”, in J. Lambert (dir.), Archaeological chemistry III, Advances in chemistry series 252, Washington DC, American Chemical Society, 1984, pp. 19‑50.

Blackman et al. 1998: J. Blackman, R.S. Badalijan, Z. Kikodze, P. Kohl, “Chemical characterization of Caucasian obsidian: geological sources”, in M.‑C. Cauvin, A. Gourgaud, B. Gratuze, N. Arnaud, G. Poupeau, J.‑L. Poidevin, C. Chataigner (dir.), L’obsidienne au Proche et Moyen-Orient: du volcan à l’outil, BAR International Series 738, Oxford, Archaeopress, 1998, pp. 205‑231.

Cauvin 1981: M.‑C. Cauvin, “À propos de lames-faucilles en silex”, Bulletin de la Société préhistorique française 78, 1981, pp. 168‑169.

Cauvin 1994: M.‑C. Cauvin, “La circulation de l’obsidienne au Proche-Orient néolithique”, in H.G. Gebel, S. Kozlowski (dir.), Neolithic chipped stone industries of the Fertile Crescent. Proceedings of the first workshop on PPN chipped lithic industries. Seminar Für Vorderasiatische Altertumskunde (Free University of Berlin, 29 March‑2 April 1993), Studies in early Near Eastern production, subsistence, and environment 1, Berlin, Ex oriente, 1994, pp. 15‑22.

Chabot, Pelegrin 2012: J. Chabot, J. Pelegrin, “Two examples of pressure blade production with a lever: recent research from the southern Caucasus (Armenia) and northern Mesopotamia (Syria, Iraq)”, in P. Desrosiers (dir.), The emergence of pressure blade making from origin to modern experimentation, Westmount, Springer, 2012, pp. 181‑198.

Chataigner 1995: C. Chataigner, La Transcaucasie au Néolithique et au Chalcolithique, BAR International Series, Oxford, 1995.

10.30861/9780860548058 :

Chataigner, Gratuze 2014: C. Chataigner, B. Gratuze, “New data on the exploitation of obsidian in the southern Caucasus (Armenia, Georgia) and eastern Turkey. Part 1: source characterization”, Archaeometry 56, 2014, pp. 25‑47.

Edens 1999: E. Edens, “The chipped stone industry at Hacinebi: technological styles and social identity”, Paléorient 25/1, 1999, pp. 23‑33.

10.3406/paleo.1999.984 :

Frahm, Campbell, Healey 2016: E. Frahm, S. Campbell, E. Healey, “Caucasus connections? New data and interpretations for Armenian obsidian in northern Mesopotamia”, Journal of Archaeological Science: reports 9, pp. 543-564, https://doi.org/10.1016/j.jasrep.2016.08.023 (accessed 01/10/2018).

10.1016/j.jasrep.2016.08.023 :

Gülçur 2000: S. Gülçur, “Norşuntepe, die chalkolithische Keramik”, in C. Marro, H. Hauptmann (dir.), From the Euphrates to the Caucasus. Chronologies for the 4th-3rd millennium BC. Actes du colloque d’Istanbul (16‑19 décembre 1998), Varia Anatolica XI, Paris, De Boccard, 2000, pp. 375‑418.

Gülçur, Marro 2012: S. Gülçur, C. Marro, “The view from the North: comparative analysis of the Chalcolithic pottery assemblages from Norşuntepe and Ovçular Tepesi”, in C. Marro (dir.), After the Ubaid. Interpreting change from the Caucasus to Mesopotamia at the dawn of urban civilization (4500-3500 BC). Papers from The Post-Ubaid horizon in the Fertile Crescent and beyond International Workshop (Fosseuse, 29 June‑1 July 2009), Varia Anatolica XXVII, Paris, De Boccard, 2012, pp. 306‑352.

Herling 2007: L. Herling, “Die frühbronzezeitliche Lithik”, in J. Becker (dir.), Nevali Čori. Keramik und Kleinfunde der Halaf- und Frühbronzezeit, Archaeologica Euphratica Band 4, Mainz, Philipp von Zabern, 2007, pp. 177‑188.

Khademi Nadooshan et al. 2013: F. Khademi Nadooshan, A. Abedi, M.D. Glascock, N. Eskandari, M. Khazaee, “Provenance of prehistoric obsidian artifacts from Kul Tepe, north-western Iran, using X-ray fluorescence (XRF) analysis”, Journal of Archaeological Science 40/4, 2013, pp. 1956‑1965.

Kohl 2009: P.L. Kohl, “Origins, homelands and migrations: situating the Kura-Araxes Early Transcaucasian ‘culture’ within the history of Bronze Age Eurasia”, Tel Aviv 36, 2009, pp. 241‑265.

Kushnavera 1997: K.K. Kushnavera, The southern Caucasus in Prehistory, Philadelphia, University of Pennsylvania Museum, 1997.

Marquet, Verjux 2012: J.‑C. Marquet, C. Verjux (dir.), L’Europe, déjà, à la fin des temps préhistoriques. Des grandes lames en silex dans toute l’Europe. Actes de la table-ronde internationale (Tours, Indre-et-Loire, France, vendredi 7 septembre 2007), published in Revue archéologique du Centre de la France 38, 2012.

Marro 2004: C. Marro, “Itinéraires et voies de circulation du Caucase à l’Euphrate: le rôle des nomades dans le systéme d’échanges et l’économie protohistorique des IVe‑IIIe millénaires avant notre ère”, in C. Nicolle (dir.), Nomades et sédentaires dans le Proche-Orient ancien. Compte rendu de la XLVIe rencontre assyriologique internationale (Paris, 10-13 juillet 2000), Amurru 3, Paris, ERC, 2004, pp. 53‑64.

Marro 2005: C. Marro, “Cultural duality in Eastern Anatolia and Transcaucasia in Late Prehistory (c. 4200‑2800 BC)”, Archäologische Mitteilungen aus Iran und Turan 37, 2005, pp. 27‑34.

Marro, Bakhshaliyev, Ashurov 2009: C. Marro, V. Bakhshaliyev, S. Ashurov, “Excavations at Ovçular Tepesi (Nakhichevan, Azerbaijan). First preliminary report: the 2006‑2008 seasons”, Anatolia Antiqua 17, 2009, pp. 31‑87.

Marro, Bakhshaliyev, Ashurov 2011: C. Marro, V. Bakhshaliyev, S. Ashurov, “Excavations at Ovçular Tepesi (Nakhichevan, Azerbaijan). First preliminary report: the 2009‑2010 seasons”, Anatolia Antiqua 19, 2011, pp. 53‑100.

Marro, Bakhshaliyev, Berthon 2014: C. Marro, V. Bakhshaliyev, R. Berthon, “On the genesis of the Kura-Araxes phenomenon: new evidence from Nakhchivan (Azerbaijan)”, Paléorient 40/2, 2014, pp. 131‑154.

Marro, Bakhshaliyev, Berthon 2015: C. Marro, V. Bakhshaliyev, R. Berthon, “A reply to G. Palumbi and C. Chataigner”, Paléorient 41/2, 2015, pp. 157‑162.

Pelegrin 1988: J. Pelegrin, “Débitage expérimental par pression: du plus petit au plus grand”, in J. Tixier (dir.), Technologie préhistorique, published in Notes et monographies techniques du CRA 25, 1988, pp. 37‑53.

Pelegrin 2006: J. Pelegrin, “Long blade technology in the old world: an experimental approach and some archaeological results”, in J. Apel, K. Knutsson (dir.), Skilled production and social reproduction. Aspects on traditional stone-tool technology, Uppsala, Societas Archeologica Upsaliensis, 2006, pp. 37‑68.

Pkhakadze, Baramidze 2008: G.P. Pkhakadze, M. Baramidze, “The settlement at Pichori and Colchian Bronze Age chronology”, in A. Sagona, M. Abramishvili (dir.), Archaeology in southern Caucasus: perspectives from Georgia, Ancient Near Eastern Studies 19, Leuven/Paris/Dudley, Peeters, 2008, pp. 249‑267.

Rokitta 2005: D. Rokitta, Obsidiangeräte im Vorderasiatischen Neolithikum. Die Çayönü Tools, MA thesis, Berlin, 2005 (unpublished).

Rosen 1997: S.A. Rosen, Lithics after the Stone Age. A handbook of stone tools from the Levant, Walnut Creek/London/New Delhi, AltaMira Press, 1997.

Rothman 2015: M.S. Rothman, “Early Bronze age migrants and ethnicity in the Middle Eastern mountain zone”, PNAS 112, 2015, pp. 9190‑9195, https://doi.org/10.1073/pnas.1502220112 (accessed 01/10/2018).

10.1073/pnas.1502220112 :

Schmidt 1996: K. Schmidt, Norşuntepe, Kleinfunde 1, Die lithische Industrie, Archaeologica Euphratica 1, Mainz, Philipp von Zabern, 1996.

Stöllner et al. 2010: T. Stöllner, I. Gambashidze, A. Hauptmann, G. Mindiasvili, G. Gogocuri, G. Steffens, “Goldbergbau in Südostgeorgien. Neue Forschungen zum frühbronzezeitlichen Bergbau in Georgien”, in S. Hansen, A. Hauptmann, I. Motzenbäcker, E. Pernicka (dir.), Von Majkop bis Trialeti. Gewinnung und Verbreitung von Metallen und Obsidian in Kaukasien im 4.‑2. Jt. v. Chr., Bonn, Rudolf Habelt, 2010, pp. 103‑138.

Thomalsky 2011: J. Thomalsky, “The lithic industry from Ovçular Tepesi: a preliminary report”, Anatolia Antiqua 19, 2011, pp. 81‑83.

Thomalsky 2012: J. Thomalsky, Lithische Industrien im Vorderasiatischen und Ägyptischen Raum. Untersuchungen zur Organisation lithischer Produktion vom späten 6. bis zum ausgehenden 4. Jahrtausend v. Chr., Tübingen, Publikationen uni-tuebingen.de, 2012, http://hdl.handle.net/10900/47038 (accessed 01/10/2018).

Thomalsky 2017: J. Thomalsky, “Large blade technologies in the southern Caucasus and in northern Mesopotamia in the 6th‑4th millennium BC”, in E. Rova, M. Tonussi (dir.), At the northern frontier of Near Eastern archaeology. Recent research on Caucasia and Anatolia in the Bronze Age. Proceedings of the international Humboldt-Kolleg (Venice, January 9‑January 12 2013), published in Subartu XXXVIII, 2017, pp. 79‑90.

Thomalsky forthcoming: J. Thomalsky, “Lithic networks in Iran and Caucasus”, in S. Hansen et al. (dir.), New research on the Neolithic in the Circumcaspian regions. Proceeding of a conference held in Tbilisi, Georgia (27‑30 September 2011), to be published in Archäologie aus Iran und Turan, forthcoming.

Notes de bas de page

1 The joint-archaeological DFG/ANR project “On salt, copper, and gold: the origins of early mining in the Caucasus” is directed by T. Stöllner on the German side and C. Marro on the French side.

2 Blackman et al. 1998; the analyses of the obsidians from the Mashavera valley excavations (see below) are in progress.

3 The investigations are directed by Dr Veli Bakhshaliyev (Azerbaijan National Academy of Sciences, Nakhchivan Branch) and Dr Catherine Marro (Maison de l’Orient et de la Méditerranée, Lyon, France); the author is responsible for the study of the chipped stone industry.

4 Use-wear analysis is planned be carried out on several of these pieces from Ovçular Tepesi in the near future.

5 For an overview of the distribution of notched pieces, see for example Edens 1999 (Late Chalcolithic Hacınebi); Bernbeck, Pollock, Coursey 1999 (Neolithic Kazane Höyük); Balkan-Atlı 2003 (Chalcolithic Değirmentepe).

6 For a techno-functional reconstruction, see Abe, Azizi Kharanaghi 2015 (on specimens from late Neolithic/early Chalcolithic Fars, Iran).

7 For a definition and technological details, see Herling 2007.

8 Rosen 1997; the methods of flaking techniques have been thoroughly discussed and tested by J. Pelegrin (1988, 2006).

9 For a summary, see Thomalsky 2012, pp. 288‑292.

10 Thomalsky 2012, pp. 275‑279; for further references see Chabot, Pelegrin 2012.

11 Altinbilek-Algul et al. 2012, p. 159. Other large blades industries appear later in south-eastern Europe, among which one may emphasize the “superblades” from the Varna cemetery, Bulgaria, of the late 5th millennium BC.

12 Thomalsky forthcoming; for a detailed analysis of the Çayönü tool pattern, see Rokitta 2005.

13 Schmidt 1996, p. 19, fig. 8a. No provenance analyses have been carried out for the Norşuntepe obsidian so far, but it is clear that the “green translucent obsidian” belongs to the peralkaline group of the Bingöl/Nemrut Dağ deposits.

Précédent Suivant

Le texte seul est utilisable sous licence Creative Commons - Attribution - Pas d'Utilisation Commerciale - Pas de Modification 4.0 International - CC BY-NC-ND 4.0. Les autres éléments (illustrations, fichiers annexes importés) sont « Tous droits réservés », sauf mention contraire.