Paléorient Contacts between the Southern and Northern Levant in the first half of the 5th millennium BC: A pottery perspective from Jordan Eva Gabrieli Citer ce document / Cite this document : Gabrieli Eva. Contacts between the Southern and Northern Levant in the first half of the 5th millennium BC: A pottery perspective from Jordan. In: Paléorient, 2016, vol. 42, n°2. Connections and Disconnections between the Northern and Southern Levant in the Late Prehistory and Protohistory (12^th – mid-2^nd mill, BC) pp. 151-184; https://www.persee.fr/doc/paleo_0153-9345_2016_num_42_2_5725 Fichier pdf généré le 18/01/2019 Résumé Dans la première moitié du 5e millénaire av. J.-C., les territoires qui correspondent à la Jordanie actuelle ne semblent pas être très impliqués dans les réseaux commerciaux à longue distance, mais les similitudes entre leurs céramiques et celles du Levant septentrional et central attestent l’existence de «sphères d’interaction » ou «réseaux socio-matériels » qui impliquent une sorte de connexion avec ces zones. Bien que, très probablement, la plupart des contacts n’aient pas dépassé de courtes ou moyennes distances, ces interactions ont permis l’émergence et l’adoption d’importantes innovations à l’échelle pan-régionale, en contribuant à stimuler de profonds changements au niveau local. En effet, elles semblent avoir contribué au développement différentiel de zones telles que la partie nord de la vallée du Jourdain au Chalcolithique ancien. Abstract In the first half of the 5th millennium BC, the territories which roughly correspond to present-day Jordan do not seem to be significantly involved in long-distance trade networks, but the similarities to the pottery of the Northern and Central Levant are evidence of the existence of ‘interaction spheres’ or ‘socio-material networks’ which imply some sort of connection with those areas. Although most probably contacts did not usually exceed short or middle-range distances, such interactions allowed the emergence and adoption of important innovations at a pan-regional scale, contributing to stimulate deep changes at a local level, which can account for the differential development of areas like the northern part of the Jordan Valley in the earlier Chalcolithic. Contacts between the Southern and Northern Levant in the first half of the 5th millennium BC: A pottery perspective from Jordan E. Gabrieli Abstract: In the first half of the 5th millennium BC, the territories which roughly correspond to present-day Jordan do not seem to be significantly involved in long-distance trade networks, but the similarities to the pottery of the Northern and Central Levant are evidence of the existence of ‘interaction spheres’ or ‘socio-material networks’ which imply some sort of connection with those areas. Although most probably contacts did not usually exceed short or middle-range distances, such interactions allowed the emergence and adoption of important innovations at a pan-regional scale, contributing to stimulate deep changes at a local level, which can account for the differential development of areas like the northern part of the Jordan Valley in the earlier Chalcolithic. Résumé : Dans la première moitié du 5e millénaire av. J.-C., les territoires qui correspondent à la Jordanie actuelle ne semblent pas être très impliqués dans les réseaux commerciaux à longue distance, mais les similitudes entre leurs céramiques et celles du Levant septentrional et central attestent l’existence de « sphères d’interaction » ou « réseaux socio-matériels » qui impliquent une sorte de connexion avec ces zones. Bien que, très probablement, la plupart des contacts n’aient pas dépassé de courtes ou moyennes distances, ces interactions ont permis l’émergence et l’adoption d’importantes innovations à l’échelle pan-régionale, en contribuant à stimuler de profonds changements au niveau local. En effet, elles semblent avoir contribué au développement différentiel de zones telles que la partie nord de la vallée du Jourdain au Chalcolithique ancien. Keywords: Jordan; Chalcolithic; Ubaid; Supra-regional networks; Pottery production. Mots-clés : Jordanie ; Chalcolithique ; Obeid ; Réseaux suprarégionaux ; Production céramique. INTRODuCTION In the Near East, middle- to long-distance circulation of materials began at least as early as the Epipaleolithic (Richter et al. 2011). Similarities between the “Neolithic”/ “Chalcolithic” assemblages of the Levant were already noted starting from the 1920s (e.g., Frankfort 1924; Wright 1937 and 1951; Kaplan 1960; Kirkbride 1971; Garfinkel 1999), but until recently such evidence was interpreted within the cultural-historical framework, being basically explained in diffusionistic terms. Chronological uncertainties, gaps in knowledge and heavy reliance on ‘legacy data’, however, still prevent a real under- Paléorient, vol. 42.2, p. 151-184 © CNRS ÉDITIONS 2016 standing of the phenomenon. This paper, focusing on ceramic evidence and reviewing the existing literature, outlines some of the basic issues and attempts to provide a first re-evaluation of the question from a Jordanian point of view, in order to call attention to new and older data which can contribute to better document the existence of important albeit neglected interregional connections. Manuscrit reçu le 9 décembre 2015, accepté le 14 juin 2016 152 ‘EARLY CHALCOLITHIC’ JORDAN: CHRONOLOGY AND LONG-DISTANCE CONTACTS E. Gabrieli Hagoshrim N Mediterranean Sea Horvat ‘Uza 1. As to Jordan, “(l)ittle is known about the earlier phases of the Chalcolithic (i.e. ca 5000-4500 cal. BC), […] and comments are generally confined to vague observations on ceramic parallels with the Halafian north (Kaplan 1960; Mellaart 1975; Gophna and Sadeh 1988/1989) or the Badarian south (Kaplan 1959; Kantor 1992; Friedman 1999). Such statements rarely document actual imports, far less method and motivation” (Bourke 2008: 143). 2. This is due to the fact that in the Southern Levant the Chalcolithic was originally defined based on assemblages which contained some copper(/ metal) objects. 3. But see Braun 2008 for a revision of Garfinkel’s ‘Middle Chalcolithic’ concept, in particular with respect to ‘Beth Shean Ware’. 4. J.W. Hanbury-Tenison mentioned also the painted pottery from Tel Tsaf and Kataret es-Samra, which was assumed to date to an early phase of the Chalcolithic. 5. For a review of the various positions and, above all, for a phasing of the first half of the 5th millennium different from that adopted here, see in Sea of Galilee Tel ‘Ali Nahal Zehora II Tel Tsaf Wadi Rabah Jo r dan River In the first half of the 5th millennium BC, unlike the vast majority of the Near East, the territories that roughly correspond to present-day Jordan do not seem to have been significantly involved in long-distance exchange/interaction networks (Bourke 2008).1 At present, however, only a few Jordanian sites are dated to this time. According to the local traditional chrono-typologies, most scholars refer to their archaeological assemblages as a ‘pre-classic Ghassulian’ entity, usually termed ‘Early Chalcolithic’ (despite the lack of metal finds or evidence of metal-working)2 and corresponding to Yosef Garfinkel’s ‘Middle Chalcolithic’ (1999)3 (Banning 2007a and 2012; Bourke 2008). Still in the 1980s, the Early Chalcolithic was basically defined by the presumed absence of the characteristic hallmarks of the Ghassulian (‘cornets, chalices, metal objects’), as attested in the “lower” levels of Teleilat Ghassul (Hanbury-Tenison 1986: 106),4 but even today it cannot be characterized satisfactorily (Rowan and Golden 2009). Although in the last fifteen years a growing corpus of good radiometric dates linked to sound stratigraphic sequences, has allowed us to progressively refine the absolute chronology of the south Levantine Pottery Neolithic and Chalcolithic periods (Philip 2011: 197), the transition between the two remains scarcely understood. Debate still centres to a large extent around partially published excavations and small-scale investigations with limited material culture, while considerable disagreement exists among scholars as to how conceive the chronological status and ‘cultural’ affiliations of the pre‘classic-Ghassulian’ assemblages (Banning 2007a and b; Rowan and Golden 2009; Lovell and Rowan 2011; Philip 2011; Gopher 2012).5 Based on recent radiometric evidence, how- Tell esh-Shuna Tabaqat el-Buma Pella Tubna Tell Abu Habil Abu Hamid Kataret es-Samra Tell el-Mafjar Teleilat Ghassul Dead Sea 0 50 km Fig. 1 – Map of the Southern Levant showing the relevant sites. ever, a number of Jordanian sites—Tell esh-Shuna (Bronk Ramsey et al. 2002), Tubna (Banning 2007a), Abu Hamid (Lovell et al. 2007), Pella (Bourke 2007), Teleilat Ghassul (Bourke 2007)—can be securely dated to the first half of the 5th millennium cal. BC. All of them are located in the Jordan Valley and its surroundings (see fig. 1), as well as the other few sites (Kataret es-Samra, Tell Abu Habil…) known through old surveys and soundings that are traditionally attributed to a ‘pre-classic Ghassulian’ phase on typological grounds (de Contenson 1960; Mellaart 1962; Leonard 1992).6 The Jordan Valley and the areas immediately adjacent to it might have well constituted a major settlement focus, but several reasons lead us to consider the ‘Early Chalcolithic’ phase as underrepresented (Bourke 2008; Gabrieli 2009): its still inadequate characterisation, the paucity of radiocarbon dating, the lack of highly diagnostic elements, the large numbers of particular the different contributions contained in the thematic issue of Paléorient edited by E. Banning (33.1, 2007) and in the volume edited by J. Lovell and Y. Rowan (2011). 6. Older radiometric dates, however, would indicate that limited Early Chalcolithic occupation occurred in more southerly areas too (the important site of Tell Wadi Feinan, ephemeral camps in Wadi Hisma; Bourke 2008). Paléorient, vol. 42.2, p. 151-184 © CNRS ÉDITIONS 2016 Contacts between the Southern and Northern Levant in the first half of the 5th millennium BC 153 Table 1 – Schematic chronological frame of the occupational phases recorded at the southern Levantine sites mentioned in the text based on 14 C dates calibrated at 2σ (Tel Tsaf: Streit and Garfinkel 2015; Tell el-Mafjar: Anfinset et al. 2011; Tel ‘Ali: Garfinkel 1999; Tell esh-Shuna: Bronk Ramsey et al. 2002; Wadi Ziqlab: Banning 2007a; Pella: Bourke 2007; Abu Hamid: Lovell et al. 2007; Teleilat Ghassul: Bourke 2007). Dashed lines are used to indicate that the sample quality and contextual data are not considered (the table is aimed at guiding the reader along the text). sites generically dated to the Chalcolithic, the high degree of regionalisation which distinguishes the archaeological assemblages, the low visibility of sites and the complex geomorphological history of the Jordanian territory.7 Yet, despite the underrepresentation of sites in the archaeological record, and even if the evidence relating to exchanged items is rather meagre, similarities to the coeval material culture of the Northern and Central Levant point to the existence of ‘interaction spheres’ or ‘socio-material networks’ which imply some sort of connection with those areas (see also Gómez et al. and Milevski et al., this volume) (table 1). INTERACTION SPHERES AND SOCIO-MATERIAL NETWORKS The concept of “interaction sphere” was introduced by Joseph Caldwell (1964) in his analysis of the Hopewell phenomenon to explain the distribution over large areas of specific elements of material culture and behaviour that encompassed different environmental and ethnic areas. Since then, such a 7. As to low site visibility and geomorphological history, see Bar-Yosef and Goren 1980; Banning 1996 and 2014; Field and Banning 1998; Hitchings et al. 2013. Paléorient, vol. 42.2, p. 151-184 © CNRS ÉDITIONS 2016 concept has been used to refer to different archaeological phenomena, and relevant applications in Near Eastern prehistory include the Levantine Pre-Pottery Neolithic B, the Halaf and the ubaid (Carter and Philip 2010: 6) (see also Milevski et al., this volume). Although Caldwell’s model implies that interaction increased the status of local elites, thus laying the foundation from which more stratified societies could emerge (Yoffee 2005: 205), Gil Stein (2010: 36) has stressed that its original formulation explicitly proposed that an interaction sphere can be originated by a variety of interactive processes, which do not need to be restricted to exchange of prestige goods among elites or religion. To this regard, Stein cites the works of authors such as Rita Wright (2002), who has suggested that the 3rd-millennium interaction sphere that linked the Indo-Iranian borderlands with Oman was based on connections most clearly expressed in technological style, and Christopher Carr (2005), who has proposed different interactive processes, such as vision questing, pilgrimage, and trade in ritually powerful objects, to explain the material similarities that characterize the Hopewellian horizon. Stein (2010: 37-38) has further argued that the utility of the interaction sphere model resides precisely in the fact that it is a non-hierarchical construct. Since it does not presuppose any political or cultural unity within its boundaries, nor specific social, economic or power E. Gabrieli 154 relations, it acknowledges the importance of the local specificities alongside the overarching similarities. Local variation in resources, in fact, would be what generates the interaction sphere in the first place. Yet, as pointed out for instance by Philip Karsgaard (2010: 60), the mechanism and reality of regional and pan-regional identities and their material correlate remain scarcely understood. Further insights may come from network theory and related concepts. ‘Socio-material networks’ are in fact inherently relational in nature, and therefore transcend space and the need for immediate contiguity in interaction. Network thinking seems thus to provide an even more flexible framework to interpret inter-regional interaction (Knappett 2011). Small-world theory, for instance, has shown that transmission over considerable distances does not necessarily entail a higher number of steps because small numbers of individuals with wider contacts (‘weak ties’) can act as links in a broader chain. This can “help explain large areas of broad similarities even for prehistoric societies in which travel may have been infrequent, difficult and time consuming” (Karsgaard 2010: 60). But the concept of weak ties appears to be meaningful also in relations to other aspects, like specialized technologies. Specialization is an acknowledged major source of weak ties in modern societies, but, as stated by Fiona Coward (2013: 257), starting from the Neolithic, specialization seems to have constituted an increasingly relevant part of communities’ life, and, since then, “the presence of specialists in manufacture and trade seems to have been a significant mechanism by which more geographically distinct connections were forged between individuals and groups”. More in general, following a relational approach, social groups can be conceived as ‘communities of practice’ (Lave and Wenger 1991), established by ways of doing and approaching things that are shared to some significant extent among members. According to the practices in question, multiple, overlapping communities that cross-cut different types of boundaries may exist, which, again, can more easily account for a patterning of material culture that eludes spatial proximity or entails a differential distribution of elements (Knappett 2011 and 2013). In sum, even if we do not use formalized procedures related to network analysis, this set of concepts seems to provide a more suitable framework of reference in comparison with the spatially discrete entities presupposed by the historical-cultural paradigm in order to model supra-regional interaction. THE ‘BuRNISHED TRADITION’ AND THE ‘RED WASH WARE’ HORIzON Among the lines of evidence that seem to support the existence of relations among the Levantine communities the occurrence of burnished wares and red-slipped/washed wares can be cited in the first place. Their presence characterizes in various amounts the Neolithic and Chalcolithic ceramic assemblages throughout most of the Levant, but their relatedness has been documented more firmly for the periods preceding the 5th millennium BC. The available evidence is therefore reviewed, following the canonical geographical tripartition of the Levant, in order to reconsider their occurrence in the assemblages of the first half of the 5th millennium. THE NORTHERN LEVANT Burnished ceramics are the earliest pottery produced in the Northern Levant (fig. 2). The so-called Dark Faced Burnished Ware (DFBW) was manufactured starting from the beginning of the 7th millennium BC (Le Mière et Picon 1998). At present, it is best conceived as an extended ‘family’ of pottery groups, characterised by mineral-tempered wares, a prevalence of dark colours and a function connected to the preparation and serving of food. Burnishing would no longer be intended as a distinctive feature despite the relevance of this surface treatment to such ceramic tradition (Balossi Restelli 2006), even though its presence and quality (ranging from a few strokes to a lacquer-like polish) can provide useful criteria to differentiate types and subtypes (Diebold 2000: 65). DFBW was manufactured for over a millennium, as shown in the first place by the lengthy sequence of the Amuq (A to E phases) on which was based its typology (Braidwood and Braidwood 1960)8 (fig. 3). During this considerable time span DFBW underwent significant changes, particularly concurrent with the introduction of painted ceramics (Balossi Restelli 2006). In the Halaf-related Amuq C phase (ca 6000-5800 cal. BC; Özbal 2010), new forms were introduced (in some cases borrowed from the painted pottery repertoire, like ‘cream-bowls’), the thickness of body walls increased, the use of a slip became generalized, surfaces were either highly 8. The Amuq Valley, the classical ‘plain of Antioch’, is located in the Turkish Hatay province. Here, in the 1930s, a team of the Oriental Institute of the university of Chicago led by Robert Braidwood carried out a survey and excavations, which provided a long chronological sequence still considered a point of reference for dating sites in most of the Levant. Paléorient, vol. 42.2, p. 151-184 © CNRS ÉDITIONS 2016 Contacts between the Southern and Northern Levant in the first half of the 5th millennium BC N Tell Kurdu O ronte s River ALEPPO Ras Shamra Mediterranean Sea Tell Sukas Hama Arjoune Ard Tlaili Byblos BEIRUT DAMASCUS Jisr Tell al-Baharia Tell Kazzami Jorda n River Tell Ramad Tell Qarassa North 0 100 km Fig. 2 – Map of the Northern and Central Levant showing the relevant sites. lustrous or dull (differing from the intermediate type that predominated previously), red vessels (“covered with an ochreous slip”) became much commoner and decoration rarer (Braidwood and Braidwood 1960: 138-139). ‘Red Wash Ware’ is a label originally created by Robert and Linda Braidwood (1960) in relation to some Amuq materials. At Tell Kurdu, red-washed ceramics occur in phase D in a burnished and an unburnished version, respectively called ‘Wiped-Burnish Ware’ and ‘Red-Washed Ware’ (fig. 4). The latter has a slightly coarser fabric, and only larger jars are represented in its repertoire, but the forms are analogous to those of the former. Both feature a characteristic type of jar exhibiting a bowed neck (‘bow-rim jars’). Red-washed pottery was considered as a probable variant of later, coarse DFBW types by the Braidwoods (1960: 160), but it was regarded as part of a distinct ceramic tradition by Paléorient, vol. 42.2, p. 151-184 © CNRS ÉDITIONS 2016 155 Roger Leenders (1989: 89), who defined it as a heterogeneous ware group, “the red wash” from which it derives its name being “the only characteristic that all the variants have in common”. Admittedly, it was not possible to provide a proper description, given the meagre and sparse data available at the time, but Leenders seems to use the label ‘Red Wash Ware’ indistinctively to indicate both washed and slipped wares,9 and includes among them even specimens whose rim alone was covered by red wash (/paint). Even today, however, a satisfactory, comprehensive study is lacking, and clearly such pottery still awaits an appropriate characterisation. Leaving aside the surface colour,10 fabric, finishing and other technological aspects should be taken into account in order to discriminate between possible variants and to properly distinguish Red Wash from coarse DFBW or other similar wares. ‘Red Wash Ware’, according to Leenders, spread across Western Syria, Lebanon and the Khabur Valley, with derivatives reported from Palestinian sites attributed to the ‘Wadi Rabah Culture’. In the Northern Levant, it is believed to make its appearance considerably later than DFBW, first occurring at Ras Shamra among common wares in the phase IV C (‘poterie à couverture rouge mat’), along with the earliest Halaf-related ceramics and a finer, burnished red/brown/ black-slipped ware reminiscent of DFBW of the Amuq phase C (Leenders 1989; de Contenson 1992). At other northern Levantine sites, however, red-washed pottery would have appeared even later. In the Amuq, at Tell Kurdu, its advent has been linked with the so-called Halaf-ubaid Transition (Leenders 1989; Özbal 2011). The Halaf-ubaid Transition probably dates to ca 5400-5200 cal. BC, although its actual duration could be as long as 5500-5000 cal. BC, at least in Northern Mesopotamia (Campbell and Fletcher 2010).11 Anyhow, by ca 4900 cal. BC, ubaid pottery had spread across much of the Near East (Carter and Philip 2010). At both Ras Shamra and Tell Kurdu, while the DFBW declines markedly before eventually disappearing after the 9. According to Prudence Rice, ‘wash’ should be reserved to post-fire coatings (1987). With regard to the Amuq assemblages, however, the Braidwoods (1960: 33-34) defined a ‘slip’ as “a separate application of a fine clay suspended in water” and a ‘wash’ as “a more or less all over application of any relatively pure pigment in suspension (probably aqueous) and containing little or no clay” without any reference to the time of firing. 10. Surface colour is of course indicative of technological choices (iron-rich clays, oxidation at the end of firing, etc.), but it cannot be considered in isolation to characterise wares. 11. For a discussion of the concept of ‘Halaf-ubaid transition’, see Campbell 2007 and Campbell and Fletcher 2010. E. Gabrieli 156 Fig. 3 – Dark Faced Burnished Ware from the Amuq (after Braidwood and Braidwood 1960: Pl. 13). introduction of ubaid painted pottery,12 red-washed pottery dominates the assemblages of the post-Halaf levels (de Contenson 1992; Braidwood and Braidwood 1960). Still, at Tell Kurdu, by the ubaid-related phase E (starting earlier than 4800 cal. BC: Özbal 2010), red-washed pottery has dropped to a very few percentage points. On the contrary, at Ras Shamra, in the sounding SH, it peaks during the very phase III C,13 where it is reported to constitute 85% of the assemblages of the 12. Actually, at Tell Kurdu, based on data produced by the Braidwoods, DFBW seems to have “staged a last comeback” in phase E (attaining the 5-9% of the ‘total selected bulk’) before disappearing completely. In phase E, forms are mostly those seen in phase C, but some sherds attest to the presence of quite unusual forms, like a sinuous-sided, pedestalled bowl. Intentional surface blackening drops from to 67% of the previous phase D to 46% (1960: 177). At Tell Kurdu, as well as at Ras Shamra, prehistoric levels were excavated by arbitrary levels (unless in the presence of discernible floor surfaces or architectural remains), and phases were distinguished based on pottery contained in such levels (Braidwood and Braidwood 1960; de Contenson 1992). The percentages provided in the final publications are therefore to be considered as indicative. 13. Phase III C is radiometrically dated to 5470-4700 2 σ cal. BC or 52004900 1 σ cal. BC (P-389: 6134 ± 173; de Contenson 1998). Fig. 4 – ‘Red Washed Ware’ from Tell Kurdu (The Bryn Mawr Study Collection). earliest ubaid-related levels, before rarefying in the following phase III B. At Ras Shamra, in phase III C, red-washed ware is characterized by a buff, mineral-tempered paste. The form repertoire, like in the previous phases IV C-A, is quite restricted, including hemispherical or carinated bowls, plates with in-turned rims or straight-sided plates, bow-rim jars, jars with a pyramidal, bulging neck, hole-mouthed jars, and pedestalled vessels. Paléorient, vol. 42.2, p. 151-184 © CNRS ÉDITIONS 2016 Contacts between the Southern and Northern Levant in the first half of the 5th millennium BC Bow-rim jars are considered to be the most distinctive form. Large flat, vertical handles with splayed attachments are the commonest moyens de préhension. Bottoms are flat and bear circular or spiral mat impressions. Decoration is rare (de Contenson 1992: 176). The situation of Ras Shamra in phase III C is apparently mirrored by the inland, quite substantial site of Hama. Here, even though it is not possible to estimate the relative quantities of the various wares because only decorated pottery had been recorded, in phase L, DFBW is totally absent, and a substantial amount of ubaid painted pottery is associated with undecorated ceramics, which were at times covered by a reddish slip or wash and whose paste, in the case of larger vessels, was heavily tempered with chaff (Thuesen 1988). Nevertheless, along the coast immediately to the south of Ras Shamra, at Tell Sukas, the earliest ubaid-related layers (phase M2) are reported to contain a certain amount of (intrusive?) DFBW and very rare ‘red-washed’ fragments along with more numerous ‘dark-’ and ‘light-washed’ sherds (Oldenburg 1991).14 THE CENTRAL LEVANT Further south, in Lebanon and Southern Syria, 6th/5th-millennium archaeological evidence is rather patchy (Bartl 2012). Apart from a few sites, like Byblos or Arjoune, vast areas have never been investigated or are known only through surface finds and inadequately published materials from older excavations, so that assemblages are difficult to correlate even on typological grounds. Despite this, both burnished and red-slipped/washed15 ceramics conspicuously feature among assemblages traditionally labelled as ‘Late Neolithic’ or ‘Early Chalcolithic’. Their production seems to relate to the northern Levantine ceramic tradition. Earliest (Neolithic) central Levantine burnished vessels recall the less fine types of northern Levantine DFBW, and it has often proven difficult to distinguish between burnished and unburnished specimens, so that Francesca Balossi Restelli (2006) has argued that initially central Levantine communities would have imitated coarse burnished wares because they constituted the only pottery available to them for cooking. More 14. The deep sounding excavated by the Danish Carlsberg Expedition at Sukas had a very limited extent (ca 4 by 4 m at the top) and yielded only a small amount of pottery. 15. Based on the description, it is not so easy distinguish between slipped and washed ceramics: scholars do not usually define the terms they use, and a close correspondence between the English and French terminology cannot be always established. Paléorient, vol. 42.2, p. 151-184 © CNRS ÉDITIONS 2016 157 generally, although some regional trends are clearly evident, clear-cut distinctions based on oppositions like ‘burnished/ unburnished’ or ‘fine/coarse’ tend to blur to an even greater extent, indicating that central Levantine communities recombined the traits that they had selectively adopted. This is further confirmed by the fact that in the Central Levant DFBW is decorated with incisions much more frequently than in the northern areas (a trait considered to be a southern influence) (Balossi Restelli 2006). At different central Levantine sites, the evolution of DFBW and red-washed/slipped wares and their respective percentages seem to vary to a considerable degree, though this could partly be due to the uncertainties of chronology and the patchiness of evidence. On the whole, however, DFBW would seem to remain in use for longer than in the Northern Levant—where it had eventually been displaced by ubaid/ubaid-like painted pottery—, showing the persistence of older traditions. At Arjoune, for instance, in the latest prehistoric settlement phase (VI, ca 4700-4450 cal. BC), the ceramic assemblage was still characterised by the presence of a large amount (>50%) of grit-tempered, red or black burnished vessels (virtually all the medium- and small-sized ones were burnished!). The repertory of shapes is limited to very simple bowls and jars with gently curving profiles, as befits a small settlement occupied by farmerpastoralists. Open bowls account for the majority of the identifiable rim sherds, and a type with a conical or basin-shaped profile is characteristic, whose distinctive flattening of the sides is perhaps due to the use of a tournette (Mathias 2003: 36). Slip was never used; red-washed vessels are present,16 but no bow-rim jars were found at the site. Alongside the negligible presence of ubaid sherds, the high percentage of burnished pottery has been interpreted as a sign of marked provincialism in an epoch when burnished ceramics “had ceased to be current in [...] the northern Levant” (Mathias 2003: 37), consistent with the alleged seasonal character of the site. But the very pottery tradition evinced by the Arjoune VI assemblage is thought to have influenced the resumption of pottery manufacturing on Cyprus following a 1000-year hiatus (Clarke 2010), suggesting that in the second quarter of the 5th millennium BC this part of the Central Levant could have partaken in a rather extended contact network. At Byblos, where unslipped pottery very similar to DFBW was manufactured starting from the early Neolithic, red(/black/ dark grey)-slipped, (soft-)burnished vessels made their appear16. “Although only about 6% of the total sherds [dating to phase VI] had a red wash, 10% of the rims had this type of treatment, indicating that this is more likely the true proportion of vessels of this class. Presumably the wash did not always extend as far as the base” (Campbell and Phillips 2003: 37). 158 ance in the middle phase (Néolithique moyen), when bow-rim jars also occur. In the later Neolithic (Néolithique récent), pottery exhibits a more utilitarian character: fabrics are generally coarser, and, although slip prevails as a surface treatment, vessels are more frequently unburnished, while the lustrous burnishing is used only in combination with pattern-burnish decorations. Slip is predominantly black, and a red-ochre ‘wash’ (barbouillage) is still coarsely applied to holemouths and larger vessels. In some cases, there seems to be a clear association between form and finishing, e.g. for shallow bowls with everted rim, covered with a coarse, black slip, or V-shaped bowls, frequently covered with a dark-brown slip and burnished. Bow-rim jars nearly disappear (fig. 5). In the succeeding Early Chalcolithic phase (Énéolithique ancien), slip, now red ochre in colour, is still very frequent and normally covers the exterior of the vessels extending to the inner of the rim; the use of red ‘wash’ persists, while burnish is confined to rare examples with a well-polished exterior (Dunand 1973). Even though it has been proven difficult to precisely anchor Byblos’ chronology to those of the rest of the Levant, Néolithique moyen is generally equated to the southern Levantine Wadi Rabah phase on typological grounds (ca 5800-5200 cal. BC); Néolithique récent is considered to follow, but its chronological status cannot be assessed more precisely than this, nor can that of the short-lived Énéolithique ancien occupation.17 More inland, in the Beqa’a Valley, ‘red-washed’ wares and bow-rim jars appear in the upper levels of the Halaf-related ‘Ard Tlaili’ phase at the eponymous site, along with DFBWrelated pottery, here produced starting from the earliest ceramic Neolithic ‘Labweh’ phase (Kirkbride 1969). In the (probably) later ‘Jisr’ phase,18 documented only through surface finds in the southern part of the Valley mainly at tell sites, burnished, red/black-slipped pottery is common, and indeed burnishing is distinguished by its glossiness; bow-rim jars still occur, ‘in a drab ware’, but they are said not to be as arched as the ‘classic’ specimens (Copeland 1969: 99; Marfoe 1995). In the Damascus basin, dark, burnished ceramics were produced from the beginning of the Pottery Neolithic phase but, in later periods, red-slipped pottery seems to constitute an important or even exclusive component of the ceramic assemblage. At Tell Ramad, for instance, among the surface or 17. In the Atlas des Sites du Proche Orient (Hours et al. 1994), the Néolithique récent and Énéolithique ancien periods are respectively equated to periods 8 (5400-5000 cal. BC) and 9 (5000-4500 cal. BC) (Bartl 2012). 18. According to Leon Marfoe (1995), the Jisr phase, equated to Byblos Néolithique récent on typological grounds, would be the only phase in the Beqa’a that can be placed with some confidence between the ‘Ard Tlaili phase and the beginning of the Early Bronze Age. E. Gabrieli Fig. 5 – A bow-rim jar from Byblos, ‘Néolithique récent’ (after Dunand 1973: Pl. LXXXIV; fig. 64). residual finds were some bowl sherds made of a light-coloured, mineral-tempered paste, covered with a red-slip(/wash?) on the external surface (céramique claire à engobe rouge) (de Contenson et van Liere 1964; de Contenson 2000: 219). Similar artefacts were also abundantly retrieved during a rescue excavation conducted at Tell Khazzami, a single-phase site probably contemporary to Byblos Néolithique récent. Here, pottery was fairly homogeneous in character, being manufactured in a light paste with fine inclusions and red-slipped(/washed?) on the exterior (avec une couverte rouge externe). The form repertoire comprised a restricted range of little varied types (deep bowls, basins, holemouths and necked-jars usually equipped with strap handles with splayed attachments, rare big spouted bowls) (de Contenson 1968: 61). The Chalcolithic material recently retrieved at a site located at the borders of the Leja plateau, Tell Qarassa North, seems to bear some similarities to the ceramic assemblage of Khazzami (in particular, the substantial quantity of ‘red-slipped’ pottery). Here, however, linkages with the areas lying to the south Paléorient, vol. 42.2, p. 151-184 © CNRS ÉDITIONS 2016 Contacts between the Southern and Northern Levant in the first half of the 5th millennium BC appear to be more in evidence. Since the Pottery Neolithic phase, many parallels can be drawn between the site’s assemblage and those of the Southern Levant, even though a more localized character seems to emerge over time. Red-slipped, burnished ceramics amount to about 15% of the assemblage of the Wadi Rabah-related, ‘Early Chalcolithic’ phase (58005200 BC), but they are still present in the following ‘Middle’ (4800-4600 BC) and ‘Late’ (4600-3800 BC) phases, although they progressively decrease to eventually disappear at the end of the Chalcolithic period (Godon et al. 2015). Qarassa has provided the first stratified evidence of a Neolithic-Chalcolithic occupation in southern Syria, albeit still published in a preliminary form. Similarly, a stratified sequence, which, according to the brief summary report of the excavation, spanned the 6th to 4th millennium BC, has been obtained at Tell al-Baharia in the Damascus basin (Sulaiman 2012). Very recent investigation is thus beginning to fill the gap about our knowledge of the central Levantine Neolithic and Chalcolithic periods, but, unhappily, the available evidence still prevents a clearer understanding of the role possibly played by the Central Levant in mediating contacts between the northern and southern areas of the Levant. To this regard, particularly critical is of course the lacking of data concerning the Beqa’a, considering the role the Rift has always played in the region as a route of communication and cultural transmission. THE SOUTHERN LEVANT In the Southern Levant (fig. 1), although burnished, red/ black-slipped and red-slipped pottery19 is found starting from the Pottery Neolithic A (ca 6500-5800 cal. BC), they are traditionally associated with the succeeding (Pottery Neolithic B) Wadi Rabah ‘culture’, whose ceramic repertoire is in fact distinguished by the presence of significant amounts of burnished, red/black-slipped and red-slipped vessels20 (fig. 6). According to several scholars, the production of the ‘Wadi Rabah’ bur19. In the Southern Levant, pottery is generally described as red-slipped by English-speaking authors (but see for instance Gophna and Sadeh [19881989] for the use of both “slip” and “wash”), while French-speaking authors, in their much rarer and usually older contributions, use the terms engobe or couverte. 20. The various Wadi Rabah assemblages contain different proportions of slipped and burnished pottery. At Munhata (phase 2a), “about 5.4% of the pottery has a red or black slip, or both, and in most cases heavily burnished. Decorated pottery, in general, makes up 6.2% of the assemblage. At Abu zureiq, similarly, red slip constitutes some 32% of the decoration, and is often burnished, while burnished black slip occurs on 7% of decorated sherds. In strata X-VIII at Tel Te’o, some 50% of the assemblage is reportedly slipped, and some 10% burnished” (Banning Paléorient, vol. 42.2, p. 151-184 © CNRS ÉDITIONS 2016 159 nished pottery must be related to that of the DFBW,21 even though the relatedness of the Wadi Rabah pottery to the northern assemblages has more commonly been seen in terms of Halafian affinities (Kaplan 1960; Garfinkel 1999). Halaf painted decoration was never adopted in the Southern Levant, but Halafian influence is thought to be evidenced by the presence of ‘Halafian’ forms (carinated or flaring-rim bowls; bowrim jars…) closest to those of the north Syrian Halaf sites (Garfinkel 1999: 150-151). Yet, the coeval northern Levantine assemblages also feature typical (or so alleged) Halafian forms among burnished pottery (Campbell and Phillips 2003), providing a more direct source of inspiration. ‘Normative’ Wadi Rabah assemblages,22 retrieved throughout Israel north of the Dead Sea and as far east as Wadi ziqlab in Jordan, have been radiometrically dated between ca 5800 and 5200 cal. BC (Garfinkel 1999; Banning 2007a). In the first half of the 5th millennium BC, however, burnished, red-/black slipped ceramics and/or red-slipped ceramics still feature in most southern Levantine assemblages (Garfinkel 1999; Banning 2007a; Lovell et al. 2007; Gopher 2012), suggesting a continuation of contacts with the territories to the north. To the west of the Jordan River In Northern Israel and Palestine, alongside sites like Tel ‘Ali (Garfinkel 1999), Tel Tsaf (Streit and Garfinkel 2015) or Tell el-Mafjar (Anfinset et al. 2011), whose Chalcolithic levels can be partly dated radiometrically to the first half of the 5th millennium (table 1), a number of sites have been recently assigned to this time, based on stratigraphic and/or typological grounds (Horvat ‘uza, level 16: Getzov et al. 2009; Nazur 4: Gilead 2011;23 Nahal zehora II, ‘PoWR-PG’ levels: Gopher 2012; Abu Ghosh: Milevski et al. 2015).24 Most of the ceramic 2007a: 81). For a detailed presentation of the Wadi Rabah pottery, see Garfinkel 1999: 104-152. 21. See for instance Lovell et al. 2007: 53 and references therein. 22. Avi Gopher and Ram Gophna (1993: 334-336) distinguished between a ‘normative’ Wadi Rabah entity, which “does fulfill most of the criteria for the definition of an archaeological culture”, and a number of variants that seemed “to be part of the same entity and differ from each other mostly in details of their ceramics”, although at the time available data did not suffice to characterize them and their dating was unclear. 23. See also E. Yannai, The Natzur Culture – a Phase in the Development of the Chalcolithic Culture or an Independent Culture? http://www.antiquities.org.il/article_eng.aspx?sec_id=17&sub_subj_id=458&id=1149, accessed on 20/05/2016. 24. Other few sites are traditionally referred to a pre-classic Ghassulian phase (see for instance Garfinkel 1999: 155-158, “Middle Chalcolithic pottery”). Such attributions, however, are now regarded with more caution, although these sites could date, at least in part, to the first half of the 5th millennium (Gopher 2012). E. Gabrieli 160 Fig. 6 – Wadi Rabah pottery from Hazorea and ‘Ein el-Jarba (after Garfinkel 1999: Fig. 1, pl. II; photo A. Venezian; courtesy of Y. Garfinkel). assemblages from these sites are characterized by the presence of a more or less substantial amount of red-slipped pottery usually associated to a much more limited amount of red- or blackslipped burnished vessels (Garfinkel 1999; Gopher 2012).25 Slipped vessels can account for very high percentages of the assemblage, like at Tel Tsaf (around 60% of the sherds, Gophna and Sadeh 1988-1989: 24), but can also represent a lesser component, like at Tel ‘Ali in phase 1b (66.1% of the decorated vessels, or nearly 15% of the assemblage; Garfinkel 1999: 184), although such variability most probably entails a difference in date. Slip is present in various shades of red, but at times can be 25. In Southern Israel, other sites dating to a pre-classic Ghassulian phase were discovered, but their material culture seems to show major differences compared to those lying to the north and they are generally attributed to the Besorian and Timnian ‘cultures’ (Gilead 2011; Rosen 2011). black, brown or grey. It is typically applied to bowls. Bowls usually represent the major component of the pre-classic Ghassulian assemblages and this can account for its high frequency. Different shapes are known, but particularly frequent are deep straight-sided, flaring-sided or rounded bowls (with simple, outturned or in-turned rims). Slip, however, can cover also holemouths and jars, the other two most commons types in the assemblages (Gophna and Sadeh 1988-1989; Garfinkel 1999; Getzov et al. 2009). A limited presence of bow-rim jars is reported at different sites (e.g., 1.6% at Tell ‘Ali 1b, 2.8% at Horvat ‘uza), but the bow is less arched than the ‘classic’ Wadi Rabah specimens (Garfinkel 1999; Gophna and Sadeh 1988-1989). Holemouths and jars are frequently equipped with strap handles with splayed attachments. The amounts of red/black-slipped and burnished vessels are significantly much lower. At Tel Tsaf, for instance, Paléorient, vol. 42.2, p. 151-184 © CNRS ÉDITIONS 2016 Contacts between the Southern and Northern Levant in the first half of the 5th millennium BC burnished sherds make up only about 1% of the assemblage, and can be exclusively referred to bowls and amphoriskoi (Gophna and Sadeh 1988-1989: 26); at Tel ‘Ali, they account for the same percentage, being 3.4% of the decorated sherds, and can be referred to small round bowls, deep medium-sized bowls and holemouth jars (Garfinkel 1999: 185-186). Further south, at Tell el-Mafjar, in the Jericho oasis, burnishing seems to be absent, at least based on the preliminary reports published so far (Anfinset et al. 2011). Slipping and burnishing, like other features (bow-rim jars, albeit characterised by a less prominent bow…),26 relate preclassic Ghassulian Chalcolithic assemblages to the ‘normative’ Wadi Rabah ones, although their frequency is significantly different in comparison (Garfinkel 1999; Banning 2007a). Slip still features quite conspicuously in the pre-classic Ghassulian assemblages, though it seems to decrease progressively, while burnishing is much rarer and lower in quality than in Wadi Rabah times (Garfinkel 1999; Gopher 2012). As demonstrated most clearly by the (fully published) sequences provided by the excavation carried out at sites like Horvat ‘uza and Nahal zehora, change seems to occur rather smoothly, implying a progressive decrease/increase in frequency of the various elements (Getzov et al. 2009; Gopher 2012), and thus evidencing continuity in the manufacturing of pottery between the 6th and 5th millennia BC. In fact, as already put forward by Ram Gophna and Avi Gopher (1993), the so-called ‘Wadi Rabah variants’ most probably differ not only spatially (suggesting a quite strong degree of regionalization), but also chronologically. To the east of the Jordan River In Jordan, as already pointed out, quite abundant radiometric evidence has become available in recent years, which has allowed us to date pre-classic Ghassulian assemblages more securely, and, in some cases, has led to the correction of attributions based exclusively on typological grounds. Yet, apart from Teleilat Ghassul, whose materials have been the subject of several publications (even though an updated, comprehensive study is still lacking), excavation results have been published only in preliminary or partial form. Even so, the presence of redslipped and red/black-slipped, burnished ceramics can be identified as a distinctive feature of most assemblages dating to the first half of the 5th millennium cal. BC. 26. Due to reasons of space, the issue of decoration is not discussed in this paper, but incised and impressed decorations constitute a further element that indicates connections with the Wadi Rabah assemblages (see for instance Garfinkel 1999). Paléorient, vol. 42.2, p. 151-184 © CNRS ÉDITIONS 2016 161 In the Northern Jordan Valley, at Tell esh-Shuna, both older and more recent excavation indicate that red-slipped vessels make up a significant proportion of the Early Chalcolithic (ca 5100-4750; Bronk Ramsey 2002) ceramic assemblage (de Contenson 1960; Gustavson-Gaube 1986; Gibson 1994).27 Based on the catalogue of types listed by Carrie Gustavson Gaube (1986: Appendix A, 88-113) and the relative descriptions,28 slip is applied to both coarse and medium-coarse (mineral-tempered) wares, and slipped vessels include flaringsided shallow bowls, flaring-sided or straight-sided deep bowls, holemouths and jars with everted rims. Among the pottery recovered from the Chalcolithic levels during the last, 1993 excavation campaign, between 35% and 45% of the fragments show indication of an applied slip, usually in the form of a fairly thin application of a creamy clay. Only six sherds could instead be attributed to what Gustavson-Gaube had termed ‘Dark-face slipped and burnished’ ware,29 all related to the earliest levels. Such ware has a very fine, well smoothed fabric, with small, sub-round, calcite inclusions; its surfaces treatment, which distinguishes it from the rest the Early Chalcolithic assemblage, involves the application of a thick slip, heavy burnishing and a firing in reducing atmosphere. Most of these sherds came from small rounded bowls (ca14 cm in diameter) (Gibson 1994: 122-123). At Tubna, in Wadi ziqlab, the earliest levels (ca 5000-4800 cal. BC; Banning 2007a), containing the remains of possible pit-structures, have yielded a limited amount of ceramic finds. Pastes are generally coarse and grit-tempered. The repertoire of forms is rather varied, but it is apparently dominated by jars of various types, many of which are holemouths with an everted rim; splayed loop handles are present. Slip occurs on a low but significant number of sherds (jars, flaring-sided bowls, pedestalled vessels), and can cover all the vessel, or only its exterior (sometimes extending to the inner of the rim) or interior. (Black) burnished fragments are said to be rare (Banning 1998; 2007a: 96). Pottery from earlier, Neolithic sites in the valley, like Tabaqat al-Buma or al-Basatin, shows some 27. De Contenson (1960: 15) claimed that in several cases sherds’ surfaces were slightly burnished. 28. Gustavson-Gaube listed vessel forms according to types without any regard to their different dating (“late Chalcolithic/PNB related”, Early Bronze Age), even though their precise context of finding is always provided. In fact, notwithstanding the chronological gap intervening between the two occupational phases, she envisaged a strong continuity of development in pottery production at the site, which is due to the acknowledged ‘odd’ similarities occurring between the Early Chalcolithic and Early Bronze Age materials (see for instance Braun 2004: 42). 29. In her descriptions, however, Gustavson-Gaube also mentioned reddish hues and coarser fabrics (1986: 111-112, fig. 21). 162 affinities to the Wadi Rabah ceramics.30 At Tabaqat al-Buma, slipping is the most common pottery surface finish. Slip is mostly red, but brown, white, yellow, or black colours are known; slipped sherds can also be burnished. Through phases 3 to 5, the proportion of slipped vessels is high and fairly constant (77-87%), whereas the amount of slipped and burnished sherds is lower and declines from 25% in phase 3 to 17% in phase 4 and 12% in phase 5. Based on the data available for the site and the Wadi ziqlab radiometric sequence, the estimated date for the occupation of phase 3 is placed between ca 5700-5550 and 5400-5300 cal. BC, while that of the occupation of phase 5 ca 5100-5000 cal. BC (68% confidence) (Gibbs 2008; Banning et al. 2011). The high proportion of red-slipped pottery along with the decreasing presence of burnished vessels clearly parallel the trend observed to the west of the Jordan. At the multiphase site of Pella, in the eastern foothills of the Northern Jordan Valley, ‘Neolithic’ occupation resumed after a long gap towards 4850-4750 cal. BC, while early Chalcolithic levels are dated ca 4700-4500 cal. BC (Bourke 2007).31 The Chalcolithic horizon seems to follow the Neolithic one more or less directly, probably indicating continuity in occupation across the two. According to Stephen Bourke (2008), the Neolithic assemblage shows some parallels with those of Wadi ziqlab, while the unedited Chalcolithic materials seem most closely to resemble the “Northern Chalcolithic” assemblages excavated at Tell el-Farah North and at Tell esh-Shuna.32 In the central part of the Jordan Valley, at Abu Hamid, the 30. The recognition of the Wadi Rabah tradition to the east of the Jordan is due to the very investigation conducted in the Wadi ziqlab starting from the mid-1990s by a team of the university of Toronto led by Ted Banning (Rollefson 2008). 31. At both Pella and Teleilat Ghassul, the earliest levels radiometrically dated to the 5th millennium are labelled as ‘Neolithic’ by Stephen Bourke and other authors. 32. In a chapter dedicated to Neolithic and Chalcolithic Palestine contained in the Cambridge Ancient History (1966), Roland de Vaux observed that the Ghassulian (‘Ghassul-Beersheva’) culture was not spread in the northern part of the region, where Neolithic ‘pre-metallurgical’ cultures (Jericho PNB, ‘middle Chalcolithic’ levels of Tell el-Farah North, ‘Beth Shean XVIII and pits’, Sheikh ‘Ali II, Tell esh-Shuna I…) could instead be found. In the same years, an assemblage partly different from those of the Ghassul-Beersheva culture was discovered by Jean Perrot (Perrot et al. 1967) at Neve ur. Its peculiar characteristics are the lack of churns and cornets, and a higher amount of red-slipped vessel. According to Perrot, comparable materials had already been found in the lower levels of Tell esh-Shuna. These assemblages, in some way related to the earlier ‘Munhata’ (i.e. Wadi Rabah) culture, allowed the definition of a ‘northern’ Chalcolithic facies in the Jordan Valley. Despite the fact that some correlations established between the different assemblages by these authors are now considered chronologically wrong, their identification of such a facies still holds true. E. Gabrieli earlier Chalcolithic phase (II, ca 4900-4400 cal. BC; Lovell et al. 2007), follows, after an occupational break, the (less substantial) Late Neolithic phase I (dated to the end of the 6th millennium cal. BC), whose ceramic assemblage includes a quite limited amount of painted vessels, usually decorated with redorange linear designs. In phase II, pottery shows a significantly higher variability compared to that of the lower levels, although it is not clear if this relates to the different nature of occupation. It is in this phase that fabrics (still exclusively mineraltempered) begin to stabilize into easily definable groups, and new forms, like churns, pedestalled or spouted vessels, make their appearance. The assemblage is still dominated by bowls, but necked-jars and holemouths increase in frequency. Bowls, either shallow or deep, are variable in diameter, ranging between less than 10 to more than 20 cm, and include flaringsided or straight-sided types. Jars can have either short or high necks; some have a slightly arched neck, reminiscent of the ‘classic’ Wadi Rabah bow-rims. Slipped sherds make up nearly 10% of the assemblage (over 5% are red-slipped), while burnished sherds represent 2.7% (1.3% are red-slipped and burnished). Slip is applied to both open and closed forms, including churns, and can cover the whole vessel or only either its exterior/interior extending sometimes to the inner/outer rim (Lovell et al. 1997; 2004 and 2007; Ali 2005) (fig. 7). In the Southern Jordan Valley, at Teleilat Ghassul, which gave the Ghassulian ‘culture’ its name, the earliest levels (phases I/H), commonly labelled as ‘Neolithic’, are roughly dated to 4800-4600 cal. BC. Throughout the early phases, the form repertoire is rather restricted, consisting mostly of small and medium flat-based bowls with simple rims, medium-sized holemouths and squat narrow-necked jars; finer variants increase in quantity over the course of phase H. In phase I, decoration is very rare, but in phase H the lip of small bowls is occasionally painted with a red-brown band, and a small number of vessels is self-slipped in pale off-white to buff. The ‘Neolithic’ ceramic materials differ markedly from both those collected at the coeval northern sites and those contained in the later, ‘Early Chalcolithic’ levels (phase G begins ca 4500 cal. BC). Phase G/F assemblages, characterised by the new use of a ferruginous clay and growing elaborateness of shapes, include a limited amount (around 5%) of red-slipped vessels (bowls, small jars), and the earliest examples of churns, cornets and fenestrated vessels. At Ghassul, however, discontinuity is more dramatically mirrored by the replacement of the ‘Neolithic’ semi-subterranean roundhouse architecture with the free-standing rectilinear stone and mudbrick dwelling units characteristic of the all subsequent Chalcolithic (G-A) levels (Bourke 2007). Paléorient, vol. 42.2, p. 151-184 © CNRS ÉDITIONS 2016 Contacts between the Southern and Northern Levant in the first half of the 5th millennium BC 163 Fig. 7 – Red-slipped, burnished pottery (above); red-slipped pottery and a bow-rim jar from Abu Hamid ‘middle levels’ (below) (after Lovell et al. 2007: Figs. 7.1, 7.7, p. 65; 11.4, p. 73; 7.17, p. 65; 10.4, p. 71; 9.8, p. 69; 11.5, p. 73; 10.2-3, p. 71; 11.10, p. 73; 10.1, p. 71). The ceramic assemblages recovered at the Jordanian sites lying in the Northern/Central Jordan Valley and its immediate surroundings include amounts of slipped and burnished pottery clearly comparable to those observed for the assemblages excavated to the west of Jordan. Similarly, in some places, a certain degree of continuity with the preceding Neolithic Wadi Rabah-related pottery can be discerned. But the very same features point to the existence of connections with the territories of the rest of the Levant. Although these connections cannot be documented in detail due to the partial edition of the Jordanian materials and the dearth of data concerning Lebanon and Southern Syria, the presence of burnished and slipped ceramics indicates that North-Western Jordan, like Northern Israel and Palestine, partook to some extent in the tradition33 of DFBW. Such tradition, which is distinct from the painted tradition of both upper Mesopotamia (culminating in the Halaf phenomenon) and the southern Levantine Neolithic Yarmukian and Lodian ‘cultures’ (Le Mière et Picon 1998), in the 7th millennium included the Central Levant and reached as far as the Hulah Valley in Northern Israel,34 but in the 6th and 5th millennia also spread to more southernly territories. This ‘burnished’ (Campbell and Phillips 2003) or ‘burnishing tradition’ (Lovell et al. 2007) is of course in the first place related to the manufacture of DFBW, but, by the 6th millennium, a wider spectrum of mineral-tempered, dark- or redslipped/washed ceramics, either burnished or unburnished, appears to be spread throughout the vast majority of the Levant. The very same morphological types can be both burnished or unburnished (like the distinctive bow-rim jars), and the use of burnishing and slipping or washing is not mutually exclusive and characterizes functionally equivalent forms (bowls, holemouths, jars) in various combinations. Red-washed pottery, as already said, made its appearance in the Halaf-related phase IV C at Ras Shamra, concurrently with a slipped variant of DFBW that frequently displayed a red 33. According to Olivier Gosselain (2000: 190), “(p)otting traditions are what one could call “sociotechnical aggregates,” an intricate mix of inventions, borrowed elements, and manipulations that display an amazing propensity to redefinition by individuals and local groups.” 34. At the recently excavated site of Tell Ro’im West, unlike the rest of the Southern Levant, the PNA ceramic assemblage comprises a significant percentage of dark, burnished sherds, which is considered to represent a northerly-oriented tradition (Nativ et al. 2014). A PAN-LEVANTINE MONOCHROME POTTERY HORIZON Paléorient, vol. 42.2, p. 151-184 © CNRS ÉDITIONS 2016 E. Gabrieli 164 Fig. 8 – Phase C and D pottery from Tell Kurdu (after Braidwood and Braidwood 1960: Figs. 105, p. 140; 107, p. 141; 121, p. 159; 126, p. 162; 127, p. 163). surface. However, at Tell Kurdu, where these changes would be more easily detectable, phase D red-washed pottery seems to have evolved out from the slipped DFBW version of the previous Halaf-related phase C. As the Braidwoods suggested long ago, in fact, most forms of phase C DFBW can be found among those of the Wiped-Burnish and Red-Washed Wares of phase D (fig. 8). DFBW is a variable category of pottery, and its evolution after the appearance of the Halaf-related pottery has not been studied in detail, but it may have undergone a process of quite marked diversification. Apart from their aesthetic potential (both in terms of visual and tactile properties), burnishing and slipping have the same basic function, allowing for a reduced porosity, and thus improving the performance of vessels.35 The 35. Actually, slip is an ideal media for polishing (e.g., Santancreu 2014). generalized application of slipping and washing could indicate that in the 6th millennium more care was taken in the manufacture of the vessels commonly used for the preparation, serving and consumption of food, even though further variability would have been dictated by the intended purpose and context of use. In particular, treatments like heavy burnishing or firing in a reducing atmosphere, which require a higher investment in time and resources and presuppose higher skills, would have been applied only to some vessels, and, actually, they seem to be used more and more rarely over time. Of course, in absence of macroscopic re-examination and, above all, of chemico-physical analyses, it is not possible to say whether or not the DFBW family can be extended to include red-slipped/washed ceramics. Indeed, the validity of such a wider grouping could be called into question, since, despite some overlap, differences probably entailed rather different Paléorient, vol. 42.2, p. 151-184 © CNRS ÉDITIONS 2016 Contacts between the Southern and Northern Levant in the first half of the 5th millennium BC manufacturing techniques. Nonetheless, in the 6th and early 5th millennia, the presence of burnished and red-/dark-slipped or washed ceramics does characterize most Levantine assemblages, and burnishing and slipping/washing seem to contribute to define a tradition of monochrome pottery extending to nearly all the Levantine region. THE SPREAD OF SPECIALIzED TECHNOLOGIES Such pan-Levantine (albeit heterogeneous) phenomenon involved a sharing of technological knowledge over a quite large area. Nevertheless, different levels of expertise are attested, hinting at the possible co-existence of distinct communities of practice. The technology involved in the production of highly lustrous, slipped pottery, for instance, differs from that required by the manufacture of most contemporary pottery. According to Yuval Goren (1992: 339; Goren and Cohen-Weinberger 2002: 435-436), it was a rather sophisticated technique which entailed a double firing process: “[...] the vessel was fired at very low temperature, slipped with a suspension of iron-rich clay and then fired again in either oxidizing or reducing atmosphere in order to obtain the desired colour. In the [latter] case the dark colour was achieved by converting the haematite (Fe2O3) into magnetite (FeO). After this process was completed, the vessel was polished.36 This technology is known from Halaf, Samarra and ubaid contexts where it was used on vessels produced from calcareous raw material.” Goren claimed that such technique implied a control of the firing atmosphere that can only be achieved by advanced kilns. So far, no potter’s workshops or manufacturing facilities of any kind from the first half of the 5th millennium have been identified in the Southern Levant. Ethnoarchaeology and experimental archaeology have shown that it is very difficult to identify with certainty firing procedures, and that, independently of firing structures, potter’s skills are the critical factor in the control of firing (Gosselain 1992; Livingstone Smith 2001 and 2007). Two-chamber kilns, however, were discovered at contemporary sites like Tell Kurdu or Tell al-‘Abr, and, recently, a kiln was reported to be found at Tell al-Baharia (Sulaiman 2012). Be that as it may, by the first half of the 36. Actually, burnishing is generally considered as a pre-firing treatment (e.g., Rice 1987). Paléorient, vol. 42.2, p. 151-184 © CNRS ÉDITIONS 2016 165 5th millennium, southern Levantine potters undoubtedly possessed great expertise, being able to produce highly burnished, slipped ceramics. Evidently, there existed extended communities of practice within which potters could transmit such specialized technique. In this view, the relatively restricted diffusion of the highly burnished and slipped Wadi Rabah pottery—mostly confined to the north-west, apart from smaller concentrations in the southern coastal and eastern areas—raises questions about the mechanisms through which the specialised technology used in its manufacture was introduced and spread in the Southern Levant (Lovell et al. 2007: 53). Similar concerns are also raised by the occurrence at Abu Hamid (phase II) of vessels whose rim and/or upper part have been smoothed with the help of a rotary movement. They differ from the rest of the pottery not only for the finishing, denoting the adoption of a new technique, but also, as the other vessels of ‘group B’, for the narrow range of morphological types, the specific stylistic traits which made them unique pieces, the accurate manufacture and their very low numbers. All these features suggest that these were particular productions, whose function was rather specific and maybe symbolically laden, and that such vessels were probably manufactured by groups of producers distinct from other potters (Roux et al. 2011). Anyhow, their presence shows that in the Southern Levant a rotary device of some sort was already in use for the finishing of coiled-made pottery in the first half of the 5th millennium,37 although its use appears to be quite restricted (Roux 2010). Clearly, transfers of technology differ from the mere circulation of objects, and imply the acquisition of new skills. Nevertheless, if most production does not benefit from innovation, this means that we are dealing with a ‘closed’ system, within which techniques are seldom transmitted or borrowed. The end result is, among other things, an inherently fragile system, which can hardly resist major socio-economic changes (Roux 2010). Tournettes are said to be an innovation brought from the north. Indeed, Hans Nissen (1988 and 2001) has linked the vast diffusion of ubaid pottery with their spread. In the Southern Levant, however, the ‘wheel-coiling technique’, introduced in the second half of the 5th millennium, took more than three millennia to become widely adopted. It disappeared twice, once after the end of the Chalcolithic and once after that of the Early Bronze III, which indicates that it was handed 37. In the very same period are seen other advances in pottery technology, like the creation of highly specialised (possibly skeuomorphic) ceramic forms (churns, cornets). 166 E. Gabrieli down by a few (itinerant?) specialised craftsmen, attached to an elite group or serving particular social needs. Thus, the spread of the use of ‘rotary kinetic energy’ in the Southern Levant would seem to relate to a restricted transmission network, and the diffusion of the lost-wax technique is another case in point (Roux 2010). This would indicate that in the Southern Levant, despite the partaking in an extended knowledge transmission network, the dispersal of specialized technologies was subject to strict constraints, or, alternatively, that only a few communities or individuals were able or disposed to adopt such innovations. TSAF WARE AND THE ‘uBAID CONNECTION’ The existence of a different kind of socio-material network seems to be documented by the occurrence in the Southern Levant of the so-called Tsaf Ware. First identified in the late 1970s at the eponymous site, located in the western part of the Jordan Valley and dating between 5200 and 4750 cal. BC (Streit and Garfinkel 2015), Tsaf Ware is characterised, as Gophna and Sadeh (1988-1989: 10) have written, by “painted geometric bichrome designs. These designs appear on the upper part of vessels – mainly bowls and amphoriskoi of fine ware, but also on cruder ware and larger vessels. The designs usually consist of horizontal bands of geometric motifs painted in black against a smoothed, white-slipped surface. Below the design the vessel is slipped in red. Above the design a red band is painted along the rim, outside and inside. Most of the designs are notable for their fine, delicate execution” (fig. 9). Tsaf Ware has a distinctive trichromatic appearance, and it represents the earliest example of polychrome pottery in the Southern Levant. Polychrome pottery made its appearance in upper Mesopotamia during the intermediate Halaf phase, but both Halafian and ubaidian ceramic assemblages include small amounts of polychromatic vessels. Basically, as known, three paint colours were used in decorating Halaf and ubaid polychromatic pottery: black/brown, red and white, occurring in different shades according to the type of pigments, firing conditions and surface treatments. Based on the use of two or three colours for decorations, pottery is usually labelled respectively as bichrome or polychrome. The vast majority of ceramics are decorated with two colours, normally black or brown and red (Cruells 2013), but bichromepainted decorations can be applied on a white or light slip, which plays a part in determining the chromatic aspect of pottery. Fig. 9 – Tsaf Ware (after Bar-Yosef and Garfinkel 2008: Fig. 224; photo D. Harris). NORTHERN LEVANTINE BICHROME WARE As already pointed out, the duration of the Halaf-ubaid transition is still questioned, but by ca 4900 cal. BC the ubaid horizon spread across much of the Near East.38 To the west, the diffusion of ubaid/ubaid-like pottery is considered to reach as far as the Northern Levant: south of a line roughly stretching from Hama up to Ras Shamra, ubaid pottery has indeed been retrieved only very sporadically (Bartl 2012)39 (fig. 10). In the Northern Levant, bichrome-painted sherds were retrieved in significant amounts only at the large sites of Tell Kurdu (Braidwood and Braidwood 1960; Yener et al. 2000), Ras Shamra (Schaeffer 1962; de Contenson 1992) and Hama (Thuesen 1988) (figs. 11-13). Bichrome-painted pottery made its first appearance at Ras Shamra in phase IV C among the Halaf-related materials, but it was much more common in the following phases IV B and IV A, just like Red Wash Ware. In 38. The ubaid horizon has been very recently defined as “the aggregate of sites, site phases, or assemblages that contain ubaid-like black-on-buff pottery. […] Pottery is the defining characteristic simply because it is the single factor found in all assemblages that have historically been described as ubaid. […] The degree of variability is very high”, and in some sites “the ubaid element may be very small, particularly along the geographical and chronological limits of the black-on-buff distribution” (Carter and Philip 2010: 3). 39. So far, very few ubaid/ubaid-like sherds have been collected at Arjoune (Campbell and Phillips 2003) and perhaps at Byblos (Dunand 1973: 141, 150). Nevertheless, in the Homs region, where Chalcolithic evidence has generally proven to be elusive during reconnaissance survey, significant quantities of ubaid/ubaid-like pottery were observed at the SHR 094 site (Philip et al. 2002: 18). Still, as already pointed out, in the Central Levant, the Late Neolithic and Chalcolithic periods remain poorly known. Paléorient, vol. 42.2, p. 151-184 © CNRS ÉDITIONS 2016 Contacts between the Southern and Northern Levant in the first half of the 5th millennium BC 167 N Değirmentepe Arslantepe Domuztepe Kenan Tepe Chagar Bazar Fistikli Höyük Kosak Shamali ‘al-Abr Tell al-‘Abr Tell Kurdu Ras Shamra Tell Hamoukar Tell Brak Hajji Firuz Tepe Gawra Hammam et-Turkman Tell Sheikh Telul Tell Ziyadeh eth-Thalathat Hassan Tell Arpachiya sR gr i Ti Mediterranean Sea Tell Beydar Hama iver Tell es-Sawwan p Eu BEIRUT hr at DAMASCUS es Tell Madhur Ganj Dareh Tell Abada Choga Mami Tell Asmar Rive r Tell Agrab BAGHDAD BAGHDAD Choga Sefid Tell Uqair Choga Bonut JERUSALEM Susa Uruk AMMAN Hajji Mohammad Choga Mish ‘Oueili Ur al-‘Ubaid Eridu as-Sabiyah Persian Gulf KUWAIT CITY 0 300 km Fig. 10 – Map showing the Ubaid pottery distribution with key sites (after Carter and Philip 2010). phase IV B, based on drawings and photos, the decorative repertoire of the bichrome ware is quite varied and shows some similarities to that of the Halaf matt-painted pottery of the previous phase. Finely executed linear or geometric motifs (lozenges) are arranged in horizontal bands to form complex patterns and, most probably, bichromy characterizes the most complex painted designs. Bichrome motifs decorate a variety of forms, and, at least in some cases, there seems to be a correlation between ornamentation and form. Phase IV A, according to de Contenson, is marked by the impoverishment and decline of the Halafian pottery. Bichrome-painted pottery, Paléorient, vol. 42.2, p. 151-184 © CNRS ÉDITIONS 2016 however, would have peaked during this very phase, but, unhappily, only a few painted sherds have been illustrated and none is bichrome. The chronological status of phases IV B and IV A is uncertain, but they are traditionally referred to the ‘transitional Halaf-ubaid’ phase (e.g., Schwarz and Weiss 1992). The ubaid-related pottery of the following phase III C, however, seems to be rather different, and is said to be exclusively monochrome-painted. Such a situation is partly paralleled by Tell Kurdu, where the ‘transitional’ phase D levels contained a limited amount of painted pottery that seems to resemble the ear- 168 E. Gabrieli Fig. 11 – Ras Shamra Bichrome Ware: Schaeffer 1962 (table IV; fig. 25V, p. 358); de Contenson 1992 (figs. 190.4-5, p. 199; 191.3-4, p. 200; 207.1-2, 4, 6, p. 216; 208.3, 209.1, 3-4, p. 218; p. 217; 211.4-8, p. 220; 212.1, 5, p. 221; 228, p. 237) (not to scale). Paléorient, vol. 42.2, p. 151-184 © CNRS ÉDITIONS 2016 Contacts between the Southern and Northern Levant in the first half of the 5th millennium BC Fig. 12 – Tell Kurdu Bichrome Ware: Braidwood and Braidwood 1960 (figs. 128.6, p. 164; 129.9-10, p. 165; 130, p. 166; 131.11-18, p. 167; 158, p. 202); Yener et al. 2000 (fig. 14, p. 107) (not to scale). Paléorient, vol. 42.2, p. 151-184 © CNRS ÉDITIONS 2016 169 170 E. Gabrieli Fig. 13 – Hama Bichrome Ware: Thuesen 1988 (pls. IV.9, p. 216; V.3, p. 217; VIII.8, 11, p. 220; XII.7, p. 224; XIII.8, p. 225; XV.4, 13, p. 227; XX.3-4, 7, p. 232; XXII.2, p. 234; XXIII.5, p. 235; XXIV.5, p. 236; XXV.9, p. 237; XXVI.4-5, 7, p. 238; XXVII.3, 6, 9, p. 239; XXXIX.9-10, p. 251; XLIII.8-12, p. 255) (not to scale). Paléorient, vol. 42.2, p. 151-184 © CNRS ÉDITIONS 2016 Contacts between the Southern and Northern Levant in the first half of the 5th millennium BC 171 lier Halafian type more closely than the much more abundant later ubaid-related ceramics. Phase D bichrome pottery was only found during the Braidwoods’ 1930s investigations, while the recent excavation has yielded none. The presence of ‘transitional’ bichrome pottery seems to be instead documented at Hama, in phase M. At Tell Kurdu, in phase E, painted, ubaid-like pottery makes up a conspicuous part of the assemblage (75% of the ‘selected bulk’ of the Braidwoods; 45% of all rim sherds collected during all the recent excavation campaigns, Özbal 2010), but, as reported by the Braidwoods, bichrome pottery seems to represent only a minor fraction (1-5%). Moreover, the Braidwoods had noted that bichrome pottery was concentrated at the bottom and at the top of the phase E levels, and sherds coming from different depths featured different characteristics: about half of those from the lower range have a chalky white slip (like in phase D), while the most recent ones are unslipped with painted decoration covering a much wider area. The occurrence of different qualities of bichrome ware has been confirmed by the new excavations, even though the finest type (comprising sinuous-sided bowls and cups, often white-slipped, and decorated with motifs carefully outlined with thin, black lines) would seem be later in date than the lesser quality one (decorated with broad strokes of paint on untreated surfaces, and with motifs barely constrained or not by rough black outlines), possibly indicating the existence of two distinct bichrome traditions (Diebold 2000: 60-61)40 (fig. 14). Early ubaid-related materials similar to those of Tell Kurdu have also been found at Ras Shamra in phase III C, but, as said above, according to de Contenson, bichrome pottery was absent. In the following phase III B, whose abundant painted ceramics are considered to resemble more closely late northern Mesopotamian ubaid types, bichrome-painted sherds are rare (amounting only to 1% of the materials recovered in the sounding SH), and are decorated with rather simple combinations of linear and geometric motifs. A similar, simple bichrome ornamentation is found also at Hama, where bichrome-painted pottery was never characterized by the complex patterns found at Tell Kurdu and Ras Shamra. In sum, it can be said that at northern Levantine sites, despite a certain degree of variability, ‘transitional’ and ubaid- The earliest examples seem to recall the Halafian decorative tradition more closely, generally being more finely executed.41 Later, unslipped specimens would have become increasingly common, as well as a less elaborated and coarser decoration including new motifs such as wavy lines. These, at least, are the changes evidenced by the available data, though they cannot be firmly documented: Tell Kurdu is the only site where recent excavations have been carried out, which, of course, has allowed us to partially refine the results obtained in the past, but the sequences established through the older excavations conducted at Ras Shamra and Hama still present serious problems in terms of reliability, and it is often not even possible to properly assess whether the retrieved materials are indigenous, intrusive or residual. 40. Actually, at Tell Kurdu, the presence of bichrome pottery varies also according to places: for instance, nearly all the bichrome sherds collected during the 1999 excavation season were found in trenches 11-15, where a pottery workshop has been discovered (Yener et al. 2000). 41. In the Syrian ubaid, however, the heritage of Halaf seems to be rather marked. Paléorient, vol. 42.2, p. 151-184 © CNRS ÉDITIONS 2016 Fig. 14 – ‘Transitional’ (1) and ‘Ubaid-like’ (2) Bichrome Ware from Kurdu (after Braidwood and Braidwood 1960: Pl. 82-83). like bichrome-painted vessels are distinguished by the same basic traits: – (mostly or exclusively) fine or medium wares; – the presence of mineral inclusions and, in the case of bigger vessels, organic temper (at Hama); – small to medium shapes (mostly bowls and jars), suitable for displaying, serving and consuming food; – a red- and black-painted decoration frequently confined to the upper part of the vessel; – a whitish/light-coloured slip or untreated background; – a relatively restricted set of mostly geometric and linear motifs usually arranged in horizontal bands, but at times combined to form more complex patterns; – the nearly exclusive use of black paint for outlines; – much lower numbers compared to black-on-buff pottery. 172 SOUTHERN LEVANTINE BICHROME WARE Tel Tsaf, as already mentioned, lies in the Central Jordan Valley, far beyond the western limit of the so-called ubaid oikoumene (Braidwood and Braidwood 1960) or interaction sphere. Nonetheless, here, four ubaid sherds were collected during the 2005 and 2006 excavation seasons. They find comparisons with types of Ras Shamra’s level III B and the excavators consider them as imports from the Northern Levant (Garfinkel et al. 2007). Even if phase III B is almost certainly to relate to a time later than the initial occupation at Tel Tsaf, their presence could hint to the existence of contacts linking the Jordan Valley and the Northern Levant. This is surely a fact worthy of note since Tsaf Ware bears some similarities to the northern Levantine bichrome-painted pottery. Such similarities were noted from the very early days of its discovery, when Tsaf Ware was compared to Halafian wares (Gophna and Sadeh 1988/89; Leonard 1989).42 Radiometric evidence, however, has now shown that Tsaf Ware is in fact chronologically related to the northern Levantine ubaid/ubaid-like assemblages or, at the earliest, to those contained in the Halaf-ubaid ‘transitional’ levels, as Leonard suggested on a stylistic basis some decades ago, even though very close parallels cannot be found.43 Tsaf Ware has turned out to have a rather restricted distribution, being found only at a handful of sites. Indeed, so far the only other site, apart from Tel Tsaf, to yield an appreciable quantity is Kataret es-Samra (Garfinkel et al. 2007), which lies on the eastern bank of the Jordan River. Limited soundings carried out by Albert Leonard at this site in the 1980s revealed the existence of “five major archaeological phases”, whose earliest two represented two successive occupational episodes clearly predating the Early Bronze I levels (Leonard 1989). Tsaf Ware was found on both the living/ working surfaces and in the relative destruction strata. Leonard (1989: 4-6) describes it as a “fine ware” characterised by “poorly or only moderately well-levigated clays, heavily tempered with white and grey grit (flint?) and straw/ chaff temper”, and covered by a slip. “On the interior, the slip is thin enough to allow both the mineral and organic temper 42. “Elaborately painted pottery […] is so rare in Chalcolithic Palestine and in the Southern Levant that one should at least look at the roughly contemporary cultures to the north, where painted pottery has a rich and lengthy heritage.” (Leonard 1989: 10). 43. “[This kind of] fine ware does not (nor probably should it be expected to) fit neatly with material from either Amuq phase C (Halaf) or Phase E (“ubaid”) horizon; yet it does seem to be part of the same cultural milieu (…), demonstrating an artistic tradition that seems more at home in Syria during Amuq Phase D” (Leonard 1989: 11-12). E. Gabrieli to break through the surface. On the exterior, however, the pottery is covered by a thick slip to which the painted decoration is applied.” Two colour variants could be distinguished: vessels with red and black paint on a white slip and vessels with reddishpurple paint on a pinkish slip.44 The colour contrast is of course sharper in the former case, but, based on Leonard’s descriptions and drawings, the overall visual effect seems to be not too different: “In both groups, the designs are applied with a very thin brush (some of the brush strokes are less than [1 mm] wide). […] Three basic patterns are utilized in the decoration of the vessels: framed, cross-hatched (latticed) band; solid, lozenge chain, and a composite motif combining both elements. All decoration is applied horizontally around the upper portion of the exterior of the vessel, at or slightly below the rim. A thin band of similar color usually complements the interior of the rim. The decoration on all of the pieces appears to have been confined to the upper portion of the vessels since neither body sherd nor base with similar or related decoration has been found.”45 (fig. 15). The rest of the ceramic assemblage was said to find the best parallels in the “common wares of Chalcolithic Ghassul (Mallon et al. 1943; Köppel et al. 1940; North 1961; Hennessy 1969)” (Leonard 1989: 6-8).46 Tsaf Ware sherds recovered at Kataret es-Samra seem to differ from those collected at Tel Tsaf in some ways. They are made of a light red or red-orange (2.5YR 6/6-6/8; 5YR 7/4-7/6; 10R 6/6-6/8) instead of a light-coloured clay (“white to pink”, Gophna and Sadeh 1989: 9), which was tempered with both mineral and organic matter. Moreover, forms and the decorative repertoire appear to be less varied, and the number of sherds retrieved is smaller. The difference could be explained in chronological terms, but, as both the ceramic assemblages seem to relate to successive occupational phases or subphases (suggesting a relatively prolonged production), this also raises the question of whether Tsaf pottery was manufactured at different places. A few other finds known from excavations conducted at other Jordanian sites—Tell esh-Shuna (where the ‘lattice work’ was among the most common painted motifs in the later 44. Gophna and Sadeh (1988-1989: 12) noted that “(i)n some cases, there is no white slip and the background is beige or pink, or the design is painted in a purplish red colour.” 45. This prevents us from recognizing sherds pertaining to the lower part of the vessels’ body. 46. At the time, the Jordanian Chalcolithic was not commonly clearly subdivided into an early and a late phase. Paléorient, vol. 42.2, p. 151-184 © CNRS ÉDITIONS 2016 Contacts between the Southern and Northern Levant in the first half of the 5th millennium BC 0 0 10 cm 10 cm Chalcolithic contexts; Gustavson-Gaube 1986; Gibson 1994), Abu Hamid (Lovell et al. 2007), Tubna (Banning et al. 1998, Banning 2007), Abu Habil (de Contenson 1960; Leonard 1992)—, seem to attest further variability, but their often inadequate characterisation and, above all, the paucity of data do not allow us to reach a clear picture (fig. 16). More recently, a further site located along the western bank of the Jordan River in the Jericho oasis area, Tell el-Mafjar, is said to have yielded Tsafian pottery; some rim sherds there “have painted geometric decoration, particularly triangles with a thick line above and below, in addition to a net pattern bounded above and below by a thick line” (Anfinset et al. 2011: 103);47 the finds, however, remain unpublished. The results of the petrographic analysis carried out on a representative sample of sherds of all pottery groups from Tel Tsaf’s Stratum I indicated, as written by Gophna and Sadeh (1988-1989: 31, note 3) “that most of the pottery […], including all the painted ware and the Dark-faced Burnished Ware, was produced from raw material easily accessible to the inhabitants of the site, whether in the immediate vicinity or on the other side of the Jordan River in the Wadi Yabis [now called Rayyan] area”. Even if at present it is not possible to identify any precise manufacturing centre or place, it seems quite reasonable to assume that Tsaf Ware was produced somewhere in the (Central?) Jordan Valley, as its rather 47. Among the comparanda for the ceramic assemblage is cited Gophna and Sadeh 1988-1989 (figs. 6-8, illustrating Tsaf Ware). Paléorient, vol. 42.2, p. 151-184 © CNRS ÉDITIONS 2016 173 Fig. 15 – Tsaf Ware from Kataret es-Samra: vessels with red and black paint on a white slip (above); vessels with reddish-purple paint on a pinkish slip (below) (after Leonard 1989: Figs. 6.3-5, p. 8; 5.1, 4-6, p. 7). circumscribed diffusion and uneven distribution would confirm. As already highlighted, Tsaf pottery seems to fall within the (Halaf/)ubaid ‘Bichrome Ware’ tradition, and the presence of the ubaid(/ubaid-like?) sherds at Tel Tsaf would indicate that Jordan Valley people were aware of this kind of polychromatic decorations. However, the striking fact is that it was this kind of pottery that inspired a local production in the Southern Levant instead of the less elaborate and more common ubaid/ubaid-like black-on-buff ceramics. The Jordan Valley communities chose to imitate a very distinctive type of pottery, working out their own version, based on a rather limited set of decorative patterns and motifs, and slipped on the vessel bottom. Actually, we do not know if they looked at the Bichrome Ware as a rarer or more exotic (and therefore more valuable) item, but, evidently, they considered it as more suitable to their taste and needs. Far from being a mechanical replication, this selective adoption implies that those people played an active role in the reproduction of material culture—a phenomenon very well known for both the Halaf and ubaid interaction with regard to the appropriation of foreign/external traits, which is indicative of the complex and diverse mechanisms that such interactions entailed. Later finds would seem to attest to the persistence of the Bichrome Ware ceramic tradition in the Southern Levant beyond the mid-5th millennium. At Teleilat Ghassul, both the middle (F-D) and upper phases’ (C-A) levels contained sherds bearing the characteristic Tsafian decoration (Mallon et al. E. Gabrieli 174 Trichromy is quite rare (Mallon et al. 1934: 124) but, even if a third paint colour was not applied, the use of a whitish or cream slip still usually distinguished the finest fragments. Actually, there would seem to be a close association between the Tsaf-like geometric decoration and the presence of the slip, which characterizes what are maybe some of the finest Ghassulian vessels. The co-occurrence of some specific traits points to the existence of a definite, recognizable template, one that would have been replicated for a very long time. Far from being a mere decorative style, the ‘Tsafian’ should be considered as a true ‘ware group’, as acknowledged by Jaimie Lovell (2001).49 The less elaborated vessels manufactured at Ghassul (?) would seem to parallel similar later productions at the northern Levantine sites.50 In fact, ubaid pottery decorative styles gradually evolved toward plainness, which might correspond to the inception of a more standardized vessel production (Akkermans 1988 and 1989), according to a process described in broader terms as the ‘evolution of simplicity’ by David Wengrow (2001).51 Given the lack of more precise dating, sites containing ubaid/ubaid-like pottery can be correlated only in broad terms52 but, even if a chronological discrepancy might be implied by the modes and tempo of the exchange/transmission, the sequences in the Northern Levant seem to attest to a certain parallelism in the evolution of the Bichrome Ware tradition, which was paralleled, at least in a loose way, by the developments of the southern Levantine sites. This would hint at a continuity of contacts between these areas, even though the mechanisms that determined the uneven distribution of the Tell esh-Shuna Tubna Abu Hamid Abu Habil 0 5 cm Fig. 16 – Tsaf-like sherds from various Jordanian sites (Tell eshShuna: Gustavson-Gaube 1986: Fig. 21.107-109, p. 112; Baird and Philip 1994: Fig. 9.2, p. 121; Tubna: Banning 2007: Fig. 5.7, p. 95; Abu Hamid: Lovell et al. 2007: Fig. 10.10, p. 71; Abu Habil: Leonard 1989: Figs. 5.2, p. 7, 6.1-2, p. 8; de Contenson 1960: Fig. 23.2). 1934; Köppel et al. 1940; Blackham 1999; Nigro e Panciroli 2001),48 although usually executed in a less elaborated and coarser manner (fig. 17 a-b). 48. Even if phase F cannot be precisely dated, the preceding phase G begins around the mid-5th millennium cal. BC (Bourke 2007: 25). Phases F-D are thought to correspond to PBI’s phases II-III, while phases C-B and A are equated respectively to PBI’s phases IV A and IV B (Bourke 2007: 17-19). Köppel observed that geometric painted decoration was nearly absent in phase IV B (i.e. A), and the lozenge motif appeared “starting” from the previous (underlying) phase IV A (i.e. C-B) (1940: 73-75). Tsaflike painted decoration, however, is included among “the ‘pre-classic’ Ghassulian practices” (referred to D-F horizons) by Stephen Bourke and colleagues (2007: 58, figs. 29-30), so that they consider the ‘Tsaf-like’ fragments retrieved in the latest levels during the 1999 campaign as altogether residual. 49. Lovell, however, defines it in somewhat different terms, based on “the presence of finely painted geometric decoration (…) on particularly fine ceramics” (Lovell et al. 2004: 268, note 2). 50. At Tell Kurdu, actually, this kind of pottery might have become an even more luxurious item in the course of time (Diebold 2000). Here, however, the latest levels of the ubaid-related phase E are dated to the beginning of the 5th millennium (ca 4800 cal. BC). 51. According to Wengrow (2001: 181), after an unprecedented phase of elaboration and experimental design which took place in the (6th millennium) Halaf period, “the simplification of pottery designs began throughout Mesopotamia during the fifth millennium […], and reached its peak with the onset of urbanization during the fourth millennium. Throughout the vast network of Mesopotamian villages, the form of painted vessels became markedly less diverse and ornamental designs were reduced to concentric bands filled with simple rotary patterns”. 52. As for the Northern Levant, dependable radiocarbon dates are available only for Tell Kurdu. Paléorient, vol. 42.2, p. 151-184 © CNRS ÉDITIONS 2016 Contacts between the Southern and Northern Levant in the first half of the 5th millennium BC Fig. 17a – Bichrome ware from Teleilat Ghassul (after Mallon et al. 1934: Pl. 50, 54, 65; Blackham 1999: Fig. 17.16, p. 55). Fig. 17b – Bichrome ware from Teleilat Ghassul (after Köppel et al. 1940: Pl. 56, 80 and 90; Blackham 1999: Fig. 10.6, 9, p. 41; Garfinkel 1999: Fig. 3, pl. IV, p. XVIII; Nigro e Panciroli 2001: Fig. 11.1, 3, p. 41). Paléorient, vol. 42.2, p. 151-184 © CNRS ÉDITIONS 2016 175 176 ubaid/ubaid-like pottery in the Levant, both in spatial and quantitative terms, remain obscure. The manufacture of bichrome-painted pottery entails considerable expertise, since bichromy is obtained in one single firing by using aqueous slurries or iron-rich clays as ‘paint slips’ and alternating reducing with oxidising firing conditions (Noll et al. 1975: 602). Its very limited diffusion could offer a further example of the restricted spread of an innovation in the Southern Levant, while its apparent absence in the Central Levant could instead imply the existence of contact networks ‘between centres’. The different roles played by the various communities/sites and the uneven settlement distribution are probably meaningful in this regard. Interaction and the possibility of reception are certainly affected by population numbers and density as much as by the socio-economic context and ideological beliefs. Tel Tsaf, where the highest and most significant concentration of bichrome painted pottery was found, seems to have been a quite substantial site, and, as will be apparent in the following paragraph, is the only centre that has so far yielded evidence of both long-distance exchange and a certain complexity between the end of the 6th and the beginning of the 5th millennium. DETECTING COMPLEXITY ‘Techno-petrographic’ analyses53 carried out at Abu Hamid have revealed that the ceramic assemblage of the first half of the 5th millennium BC is characterised by a great heterogeneity both in terms of clay provenance and technological practices. In contrast with what is often assumed, the local production constitutes only a minor part of the assemblage, and the observed variability points to the coexistence of different communities of practice (Roux et al. 2011), some of them originating in or, at least, collecting raw materials from the Jordanian plateau. At present, it is not possible to assess if that phenomenon can be referred to the particular role played by the site of Abu Hamid as a regional cultic centre (or ‘transaction centre’: Struever and Houart 1972), or if it is related to the high degree of mobility that might have characterised the communities at the time (Roux et al. 2011). Anyway, the available data seem to indicate that people from Abu Hamid 53. “The techno-petrographic approach consists of classifying ceramic assemblages according to a hierarchical order that distinguishes technological, petrographic and morpho-stylistic groups, in that order, in relationship to one another” (Roux et al. 2011: 115). E. Gabrieli interacted more or less regularly with the communities of the plateau, or, at least, that they frequented or exploited the plateau. Although it is held that the plateau was probably not fully settled until the Early Bronze Age (Bourke 2008), its widespread exploitation has been pushed back to the second half of the 5th millennium, being connected to olive cultivation (Lovell 2002), but evidence from both Wadi ziqlab (Banning et al. 1998) and Wadi Rayyan (Lovell et al. 2010) indicates that both occupation of the plateau and olive cultivation or harvesting began earlier. It is not possible to say whether olive oil represented a trading commodity at these early dates, even though the presence at Gilat of torpedo jars—namely, containers (most probably) specifically designed for the transport of oil—imported from Judea and the upper Shephelah might indicate this very occurrence54 (Commenge et al. 2006). Nevertheless, comparable evidence is so far lacking for the Jordanian territory.55 The availability of surplus commodities is obviously among the factors, which initiate or at least favour exchange, and complementarity stemming from the differential distribution of resources across the environment has been identified as a primary cause for interaction. At the time, perishables could not circulate over long distances, and the Jordanian territory does not seem to offer so much in terms of rare and/or distinctive raw materials. In fact, the distribution of artefacts made of basalt, semi-precious stones and the fine brown flint quarried in the eastern arid areas, seems largely, if not exclusively, confined to the southern Levantine region (Rowan and Golden 2009). Of course, the potential that goods like textiles or other crafts had cannot be underestimated, and, above all, we should not forget that interaction does not necessarily entail the exchange of goods (Carr 2005), but current evidence seems to be more in line with a situation where the bulkexchange of raw materials and products occurred at a local or regional level. If at Teleilat Ghassul, even in the Late Chalcolithic indications of long-range exchange/interaction are discouragingly scarce (Bourke 2002), in the first half of the 5th millennium in more northerly areas of the Jordan Valley contacts with the territories of the Central/Northern Levant are more in evi- 54. Actually, it is not certain whether oil transported to Gilat can be considered as ‘traded’: it probably served ritual needs and could have been purposely produced, offered, etc. Its presence at Gilat, however, implies that at the time at least some products were moved over some distance. 55. At least one of the torpedo vessels found at Gilat is designated an import from Ghassul by Goren (2006: 378); ‘Lower Cretaceous’ clays, however, are very widespread in the Southern Levant, with outcrops to both the east and west of the Jordan River (see also Bourke 2002: 157-158). Paléorient, vol. 42.2, p. 151-184 © CNRS ÉDITIONS 2016 Contacts between the Southern and Northern Levant in the first half of the 5th millennium BC F. 60 a wall 48 (platform) wall 60 F. 97 177 b N F. 100 (steps) F. 93 (pit) F. 99 wall 56 wall 58 wall 57 F. 98 0 2m Fig. 18 – a) Pella granary (after Bourke et al. 2003: Fig. 6, p. 338); b) a reconstruction of the Beth Yerah granary (after Greenberg and Paz 2010). dence. Actually, the northern part of the valley would seem to attest to precocious developments, pointing to a differential growth, conspicuously evidenced in the earlier appearance of those traits which will later characterise the so-called classic Ghassulian ceramic assemblages (i.e. cornets and churns),56 but more significantly expressed by aspects like the establishment of a different economic regime, aimed at intensification of both plant and animal exploitation (Bourke 2007), with all the possible implications for internal social stratification and external relationships. The contacts with the Central/Northern Levant would have contributed not only to create a distinct ‘northern’ facies, as first envisaged by R. de Vaux in the 1960s (Bourke 2008; Gabrieli 2009), but could have also contributed to making these communities more dynamic and capable of being more receptive to external stimuli. While at Teleilat Ghassul the subsistence system underwent relevant changes only around the mid-5th millennium, at Pella the faunal assemblage attests, along with an increased amount of cattle, to a more advanced age at death for both cattle and ovicaprines already starting from the first centuries of the same millennium. Moreover, high-yield cereals, pulses 56. Churns, however, are supposed to have been used for processing milk, so their presence should be related to dairy production. The earliest specimens seem to appear in the Wadi Rabah phase (Gopher 2012). Paléorient, vol. 42.2, p. 151-184 © CNRS ÉDITIONS 2016 and olives were consumed from this very early date (Bourke et al. 2003). Comparable figures have also been obtained for Tell esh-Shuna I and Abu Hamid II, despite slight variation associated with differences in the local environment (Baird and Philip 1994; Dollfus and Kafafi 1988). Even more significantly, recent excavations at Pella have brought to light what seems to be a massive grain storage complex dating to the second quarter of the 5th millennium (4600-4500 cal. BC), which finds a rather close parallel in the EBA III granary unearthed at Tel Beth Yerah (Bourke et al. 2003; Bourke 2007)57 (fig. 18). unfortunately, notwithstanding this unprecedented find for such an early date, at present further evidence hinting at a corresponding degree of complexity is lacking for Pella, as well as for other sites in the Jordanian sector of the Northern Jordan Valley. A dramatic increase in storage capacity preceding periods, however, is already indicated at Tel Tsaf by the size and number of silos (Garfinkel et al. 2009; Graham 2014), “suggesting a jump in the scale of surplus production and accumulation of wealth by at least some households” (Banning 2012: 411). Furthermore, this site yielded additional indicators of complexity: the excavators recovered an unusual quantity of rare 57. A series of large storage jars set in a row were found adjacent to a ‘wheat storage facility’ (Lovell 2010: 106). 178 exotica, like some obsidian tools, over 2500 ostrich eggshell beads (1668 in a single grave!), and over 30 seals and tokens (Garfinkel et al. 2007). Seals and tokens are rarely retrieved at Chalcolithic southern Levantine sites. Apart from Tel Tsaf, they have been discovered in appreciable quantities only at the earlier site of Hagoshrim (Getzov 2011).58 As known, these ‘artefacts of complexity’ originated and had a much wider use in Syria-Mesopotamia. Their restricted occurrence in the Southern Levant might imply a more limited concern for property marking and/or record-keeping, perhaps stimulated by the northern practices (Bourke 2008), though at present the exact function performed by such objects in the Southern Levant remains unclear. Moreover, at Tel Tsaf, a copper awl/pin was retrieved “while sieving the sediment collected during the exposure of the burial (i.e., the sediments that covered the burial)” where the large concentration of ostrich eggshell beads was found, apparently deposited as a grave good (Garfinkel et al. 2014: 2). If the association with the phase C3 deposits was correct, the awl/pin would be the earliest copper artefact so far recovered in the Southern Levant, and, even more remarkably, the earliest one made from a natural tin-copper.59 ‘Natural alloys’ or ‘complex metals’ contain copper with arsenic, antimony or other trace metals, whose presence seems not to be the product of deliberate admixing. The earliest artefacts made of complex metals recovered in the Southern Levant are currently dated to the Late Chalcolithic, but it is not known when complex metals began to be used in this area, nor, more importantly, is it known where they were sourced exactly. A southern Levantine origin, however, can almost certainly be excluded, and the closest sources have been identified in Anatolia, Transcaucasia and, more recently, in Syria (Golden 2014). The awl/pin retrieved at Tel Tsaf seems to be made of a metal sourced in the Caucasus region, but in the Southern 58. Hagoshrim is located in Northern Israel close to the Lebanese border and is dated to the second half of the 6th millennium cal. BC. This site yielded further evidence for long-range contacts (obsidian, chlorite vessels, figurines) (Rosenberg et al. 2010; Getzov 2011)). Comparable evidence, however, is now also offered by the more southernly site of Ein zippori, where coeval levels contained hundreds of obsidian items, a significant number of darkfaced burnished sherds, a few seals and incised bone figurines (Milevski and Getzov 2014; Yaroshevich 2016; Milevski et al. this volume). 59. The pit excavated for the burial was dug through the floor of a silo, whose abandonment is possibly dated to 5034-4931 1 σ cal. BC or 5187-4761 2 σ cal. BC (Streit and Garfinkel 2015). Based on the description (Garfinkel et al. 2014: 2, 5), however, it is not possible to completely rule out that the pit was excavated from a higher level later removed by agricultural activities or other post-depositional events (see Rosenberg et al. 2014: 152-153, 175), nor that the awl/ pin is not intrusive. E. Gabrieli Levant the earliest known objects made of a tin-copper alloy are dated to the Middle Bronze Age (i.e., the 2nd millennium BC). Recent discoveries made in Georgia and Serbia, however, indicate that tin-copper could have been used starting from the 6th millennium, while it was cast at least since the mid-5th millennium (Garfinkel et al. 2014). The find from Tel Tsaf would hence be a very early attestation of such use. Instead, its provenance from a funerary context is more in line with the earliest evidence of metal objects discovered in the southern Levantine (cave) tombs (Golden 2014). Though the copper ore deposit at Wadi Feinan (located in Southern Jordan) had been exploited at least since the PrePottery Neolithic B, as testified by the use of the mineral as a pigment for the ‘Ain Ghazal statues, and for the manufacture of beads (Levy et al. 2004: 78; Hauptmann 2007), the establishment of extractive metallurgy in the Southern Levant does not seem to predate the second half of the 5th millennium and it would not have been preceded by any preliminary phase of native copper working (Golden 2014).60 Given the frequent recycling that early metals underwent, however, the beginnings of metallurgy are difficult to investigate: the Nahal Mishmar hoard, first and foremost, shows how elusive the archaeological record can be in this regard. Heavy corrosion prevents us from understanding whether the awl/pin recovered at Tel Tsaf had been re-smelted and re-cast, but if its dating was correct, its presence could document the circulation of copper alloys or objects made of copper alloys over great distances at this very early date, and would offer a further indication that metallurgical technology diffused from the north, as finds such as those made at Mersin or Değirmentepe seem to suggest. In the Southern Levant, copper artefacts would have first been imported through networks like those discussed so far and only later would have been produced locally (Garfinkel et al. 2014).61 In short, thanks to research carried out at Tel Tsaf (Garfinkel et al. 2007 and 2009; Rosenberg et al. 2014), it is now possible to better document what level of complexity southern Levantine communities could have attained between the end of the 6th millennium BC and the beginning of the 5th millennium. Tel Tsaf stands out as an unusual site, and it could have been played a particular role in contact networks, but, given the rather limited data available for the first half of the 5th millen- 60. Although copper ore beads were manufactured starting at least from the PPNB, it is far from clear whether this “greenstone tradition” represents “the necessary step of applying pyrotechnology to” copper or not (Golden 2014: 561). 61. To this regard, see now Golden 2014. Paléorient, vol. 42.2, p. 151-184 © CNRS ÉDITIONS 2016 Contacts between the Southern and Northern Levant in the first half of the 5th millennium BC nium (both for Jordan and Israel), we cannot exclude that comparable discoveries will not be made in the future.62 CONCLuDING REMARKS The Southern Levant is most commonly considered to lie beyond both the Halaf and ubaid oikoumenai, and, more generally, inter-regional exchange is reputed not to have greatly contributed to the variability of Jordanian material culture during the Early Chalcolithic (Bourke 2008: 143). Actual imports are in fact very few, and at present the intensity of exchange observed at Tel Tsaf cannot be posited for any other site. Archaeological evidence, however, indicates that contacts of some significance did actually take place between the Chalcolithic Levantine populations, even though their meaning and modes are still scarcely understood. Most probably, they did not usually exceed short- or middle-range distances, apparently perpetuating older, well-established practices (Richter et al. 2011), but they involved the transfer of technologies, accounting for the spread of important innovations all over the Levant. In the case of pottery, a prevalent directionality from north to south would seem to be more clearly attested. Clearly, distinct contact networks were in existence: at different scales, different interaction spheres or socio-material interaction networks can be identified, reflecting the multiple ties which can simultaneously bind individuals, groups of people and communities. Accordingly, socio-material networks can intersect or merge, and vary to a more or less considerable degree over time, contributing to shape practices and therefore identities at both individual and collective level, even if they 179 should not to be confused, given their multi-scalar and fluctuating character, with ‘cultural’ or ‘ethnic’ markers of some sort (Asouti 2006). Objects were an effective and a necessary medium to establish and maintain relationships, providing a sort of common code embodied in them and conveyed by the shared practices that might relate to their manufacture and use. In fact, starting from the Pottery Neolithic, the greatly increasing numbers and variety of objects seem to inextricably link people in networks of variable size (Hodder 2012). Of course, conflicts could arise and disconnections could occur, but the re-examination of the available evidence, despite differences occurring at a local level, allows us to recognise some broad parallel lines of development or commonalities, highlighting the greatest importance of these contacts in entangling and structuring Levantine communities over time, even in the case of areas which seem to be less intensively connected like the Transjordanian territories. ACKNOWLEDGEMENTS I wish to thank Sandro Salvatori, Yorke Rowan and Jeremy Hayne for their insights on an earlier version of this text, as well as the three anonymous reviewers whose critical comments helped to improve it considerably. Any omissions or errors are of course my own responsibility. I also thank I. Milevski for his thoughtful remarks and providing me indispensable material, J. Hayne for revising the English text and Martino Gasparini for helping me with the images. Eva GABRIELI Milan – ITALY evagabrieli@gmail.com BIBLIOGRAPHY Akkermans P.M.M.G. 1988 An updated Chronology for the Northern ubaid and the Late Chalcolithic Periods in Syria: New Evidence from Tell Hammam et-Turkman. Iraq 50: 109-145. 1989 Tradition and Social Change in Northern Mesopotamia during the later Fifth and Fourth Millennia B.C. In: Henrickson E.F. and T huesen I. (eds.), Upon this Foundation: The ‘Ubaid 62. At the recently excavated site of Ein zippori, for instance, Pre-Ghassulian levels, still being investigated, were found above Wadi Rabah and later levels that, as already mentioned, have yielded evidence of complexity (Milevski and Getzov 2014). Paléorient, vol. 42.2, p. 151-184 © CNRS ÉDITIONS 2016 Reconsidered, Proceedings from the ‘Ubaid Symposium, Elsinore, May 30th-June 1st 1988: 339-367. Copenhagen: Museum Tusculanum Press (Carsten Niebuhr Institute Publications 10). Ali N. 2005 The Development of Pottery Technology from the Late Sixth to the Fifth Millennium B.C. in Northern Jordan. Ethno- and archaeological Studies: Abu Hamid as a key site. Oxford (BAR Int. Ser. 1422). Anfinset N., Taha H., al-zawahra M. and Yasine J. 2011 Societies in Transition: Contextualizing Tell el-Mafjar, Jericho. In: Lovell J.L. and Rowan Y.M. (eds.): 97-113. E. Gabrieli 180 Asouti E. 2006 Beyond the Pre-Pottery Neolithic B interaction sphere. Journal of World Prehistory 20: 87-126. Baird D. and Philip G. 1994 Preliminary Report on the Third (1993) Season of Excavations at Tell esh-Shuneh North. Levant 26: 111-133. Balossi R estelli F. 2006 The Development of ‘Cultural Regions’ in the Neolithic of the Near East: The ‘Dark Faced Burnished Ware Horizon’. Oxford (BAR Int. Ser. 1482). Banning E.B. 1996 2007a 2007b 2012 2014 Highlands and lowlands: problems and survey frameworks for rural archaeology in the Near East. Bulletin of the American Schools of Oriental Research 301: 25-45. Wadi Rabah and Related Assemblages in the Southern Levant: Interpreting the Radiocarbon Evidence. Paléorient 33,1: 77-101. Time and Tradition in the Transition from Late Neolithic to Chalcolithic: Summary and Conclusions. Paléorient 33,1: 137142. The Southern Levant. In: Potts D.T. (ed.), A Companion to the Archaeology of the Ancient Near East: 396-414. London: WileyBlackwell. Settlements, Surveys, and Site Visibility. Where are the Neolithic Sites? In: Finlayson B. and Makarewicz C. (eds.), Settlement, Survey, and Stone: Essays on Near Eastern Prehistory in Honour of Gary Rollefson: 91-99. Berlin: ex Oriente. Banning E.B., Blackham M. and Lasby D. 1998 Excavations at Wz 121, a Chalcolithic site at Tubna, in Wadi ziqlab. Annual of the Department of Antiquities of Jordan 42: 141-159. Banning E.B., Gibbs K. and K adowaki S. 2011 Changes in Material Culture at Late Neolithic Tabaqat al-Bûma, in Wadi ziqlab, Northern Jordan. In: L ovell J.L. and Rowan Y.M. (eds.): 36-60. 2008 Bourke S.J., Sparks R.T., McLaren B., Sowada K., Mairs L., Meadows J., Hikade T., R eade W. 2003 The Prehistory of Israel. Human Cultures before Writing. Jerusalem: Ariel Publishing & Museum of Yarmukian Culture at Sha’ar Hagolan. Bar-Yosef O. and Goren N. 1980 2012 2008 Bronk R amsey C., Higham T.F.G., Owen D.C., Pike A.W.G. and Hedges R.E.M. 2002 2007 Radiocarbon dates from the Oxford AMS System: Datelist 31. Archaeometry 44,3, Suppl. 1: 1-149. Caldwell J.R. 1964 Interaction Spheres in Prehistory. In: Caldwell J.R. and H all R.L. (eds.), Hopewellian Studies: 134-143. Springfield: Illinois State Museum (Illinois State Museum Scientific Papers 12). Campbell S. 2007 Rethinking Halaf Chronologies. Paléorient 33,1: 103-136. Campbell S. and Phillips C.S. 2003 The prehistoric pottery from trenches V, VI and VII. Part 1: A Commentary on the Earlier Prehistoric Pottery (Trenches V and VII). In: Parr P.J. (eds.), Excavations at Arjoune, Syria: 31-70. Oxford (BAR Int. Ser. 1134). Carr C. 2005 2010 Scioto Hopewell Ritual Gatherings. In: Carr C. and Case D.T. (eds.), Gathering Hopewell: Society, Ritual and Ritual Interaction: 463-479. New York: Kluwer Academic Press and Springer. 2010 Teleilat Ghassul: An appraisal of Robert North’s excavations (1959–1960). Levant 31: 19-64. Teleilat Ghassul: foreign relations in the Late Chalcolithic period. In: van den Brink E.C.M. and Levy T. (eds.), Egypt and the Levant: Interrelations from the 4th through the Early 3rd Millennium B.C.E.: 154-164. London: Leicester university Press. The Late Neolithic/Early Chalcolithic Transition at Teleilat Deconstructing the ubaid. In: Carter R.A. and P hilip G. (eds.): 1-22. Carter R.A. and Philip G. (eds.) Bourke S.J. 2002 Early Beth Shan (Strata XIX–XIII): G.M. Fitzgerald’s Deep Cut on the Tell. Philadelphia: university of Pennsylvania Museum. Deconstructing Y. Garfinkel’s “Beth Shean Ware” and “Middle Chalcolithic”: A realistic look at two archaeological chimeras. Journal of the Serbian Archaeological Society 24: 81-108. Carter R.A. and Philip G. The Northern Levant. In: Potts D.T. (ed.), A Companion to the Archaeology of the Ancient Near East: 375-395. London: WileyBlackwell. Blackham M. 1999 Excavations in the Plain of Antioch, I. The Earlier Assemblages, Phases A-J. Chicago: The university Press of Chicago (Oriental Institute Publications 61). Braun E. 2004 Afterthoughts Following Prehistoric Surveys in the Levant. Israel Exploration Journal 30,1-2: 1-16. Bartl K. Preliminary Report on the university of Sydney’s eighteenth and nineteenth seasons of excavations at Pella (Tabaqat Fahl) in 1996/97. Annual of the Department of Antiquities of Jordan 47: 335-388. Braidwood R.J. and Braidwood L.S. 1960 Bar-Yosef O. and Garfinkel Y. 2008 Ghassul: Context, Chronology and Culture. Paléorient 33,1: 15-32. The Chalcolithic. In: Adams R.B. (ed.), Jordan: An Archaeological Reader: 109-160. London and Oakville: Equinox. Beyond the Ubaid. Transformation and Integration in the Late Prehistoric Societies of the Middle East: 1-22. Chicago: The Oriental Institute of the university of Chicago (Studies in Ancient Oriental Civilization 63). Clarke J. 2010 Contextualising Neolithic Cyprus: Preliminary Investigations into Connections between Cyprus and the Near East in the Later Neolithic. In: Bolger D. and M aguire L.C. (eds.), The Development of Pre-State Communities in the Ancient Near Paléorient, vol. 42.2, p. 151-184 © CNRS ÉDITIONS 2016 Contacts between the Southern and Northern Levant in the first half of the 5th millennium BC East. Studies in Honour of Edgar Peltenburg: 197-206. Oxford and Oakville: Oxbow Books. Commenge C., Alon D., Levy T.E. and K ansa E. 2006 Gilat ceramics: Cognitive dimensions of pottery production. In: Levy T.E. (ed.), Archaeology, Anthropology and Cult: The sanctuary at Gilat, Israel: 394-506. London and Oakville: Equinox. de Contenson H. 1960 Three Soundings in the Jordan Valley. Annual of the Department of Antiquities of Jordan 4-5: 19-28. 1968 Rapport préliminaire sur les fouilles à Tell el Khazzami en 1967. Annales Archéologiques Arabes Syriennes XVIII : 55-62. 1992 Préhistoire de Ras Shamra : les sondages stratigraphiques de 1955 à 1976. Paris : ERC (Ras Shamra-Ougarit VIII). 1998 Ras Shamra et Mersin : état de la question. Paléorient 24,2 : 111113. 2000 Ramad. Site néolithique en Damascène (Syrie) aux VIIIe et VIIe millénaires avant l’ère chrétienne. Beyrouth : Institut Français d’archéologie du Proche-Orient (BAH 157). Gabrieli E. 2009 Sondages à Tell Ramad en 1963. Rapport préliminaire. Annales Archéologiques Arabes Syriennes XIV : 109-124. Copeland L. 1969 Neolithic Village Sites in the South Beqa’a, Lebanon. Mélanges de l’Université Saint-Joseph 45,5: 83-114. Coward F. 2013 Grounding the net: social networks, material culture and geography in the Epipalaeolithic and early Neolithic of the Near East (~21-6,000 cal BCE). In: K nappett C. (ed.): 247-280. Cruells W. 2013 La cerámica Halaf en Tell Halula (VII y VI milenios cal BC). Orígenes y desarrollo. In: Molist M. (ed.), Tell Halula: un poblado de los primeros agricultores en el valle del Éufrates, Siria: 59-211. Madrid: Ministerio de Educación, Cultura y Deporte. Diebold B. 2000 Preliminary Report on the Ceramic assemblage at Tell Kurdu. In: Yener K.A., E dens C., Casana J., Diebold B., Ekstrom H., Loyet M. and Özbal R., Tell Kurdu Excavations 1999. Anatolica 26: 58-65. Dollfus G. et K afafi z. 1988 Abu Hamid, village du 4e millénaire de la vallée du Jourdain. Amman : Centre culturel français et département des antiquités de Jordanie. Dunand M. 1973 Fouilles de Byblos V. L’architecture, les tombes, le matériel domestique, des origines néolithiques à l’avènement urbain. Paris : Adrien Maisonneuve (Études et documents d’archéologie 6). F ield J. and Banning E.B. 1998 Hillslope Processes and Archaeology in Wadi ziqlab, Jordan. Geoarchaeology 13,6: 595-616. F rankfort H. 1924 Studies in Early Pottery of the Near East, I. Mesopotamia, Syria and Egypt and their earliest interrelations. London: Royal Anthropological Institute of Great Britain and Ireland (Royal Anthropological Institute Occasional Papers 6). Paléorient, vol. 42.2, p. 151-184 © CNRS ÉDITIONS 2016 Aspetti e problemi del popolamento della Giordania in Età Calcolitica (V-IV millennio a.C.). unpublished Thesis. Milano: università degli Studi. Garfinkel Y. 1999 Neolithic and Chalcolithic Pottery of the Southern Levant. Jerusalem: The Hebrew university of Jerusalem (Qedem 39). Garfinkel Y., Ben-Shlomo D. and Kuperman T. 2009 Large-scale storage of grain surplus in the sixth millennium BC: The silos of Tel Tsaf. Antiquity 83,320: 309-325. Garfinkel Y., Ben Shlomo D., F reikeman M. and Vered A. 2007 Tel Tsaf: The 2004-2006 Excavation Seasons. Israel Exploration Journal 57: 1-33. Garfinkel Y., K limscha F., Shalev S. and Rosenberg D. 2014 de Contenson H. et van Liere W. 1964 181 The Beginning of Metallurgy in the Southern Levant: A Late 6th Millennium CalBC Copper Awl from Tel Tsaf, Israel. PLoS ONE 9,3 [http://www.plosone.org/article/info:doi/10.1371/journal. pone.0096882, accessed on 04/08/2014] Getzov N. 2011 Seals and Figurines from the Beginning of the Early Chalcolithic Period at Ha-Goshrim. ‘Atiqot 67: 1-26 (81*-83*). Getzov N., Lieberman-Wander R., Smithline H. and Syon D. 2009 Horbat Uza: The 1991 Excavations I: The Early Periods. Jerusalem: Israel Antiquity Authority (IAA Reports 41). Gibbs K. 2008 Understanding Community: A Comparison of Three Late Neolithic Pottery Assemblages from Wadi Ziqlab, Jordan. unpublished PhD Thesis. university of Toronto. Gibson C. 1994 The Area D pottery from the 1993 season. In: Baird D. and Philip G., Preliminary Report on the Third (1993) Season of Excavations at Tell esh-Shuneh North. Levant 26: 121-124. Gilead I. 2011 Chalcolithic Culture History: Ghassulian and Other Entities in the Southern Levant. In: Lovell J.L and Rowan Y.M. (eds.): 12-24. Godon M., Baldi J.S., Ghanem G., Ibáñez J.J. and Braemer F. 2015 Qarassa North Tell - Southern Syria: The Pottery Neolithic and Chalcolithic sequence. A few lights against a dark background. Paléorient 41,1: 153-176. Golden J. 2014 Who Dunnit? New Clues Concerning the Development of Chalcolithic Metal Technology in the Southern Levant. In: Roberts B.W., Thornton C.P. (eds.), Archaeometallurgy in Global Perspective. Methods and Syntheses: 559-578. New York: Springer. G opher A. (ed.) 2012 Village Communities of the Pottery Neolithic Period in the Menasheh Hills, Israel: Archaeological Investigations at the Sites of Nahal Zehora. Tel Aviv: Emery and Claire Yass Publications in Archaeology (Monograph Series of the Institute of Archaeology of Tel Aviv University 29). E. Gabrieli 182 Gopher A. and Gophna R. K aplan J. 1993 1960 Cultures of the Eighth and Seventh Millennia BP in the Southern Levant: A Review for the 1990s. Journal of World Prehistory 7: 297-353. The Relation of the Chalcolithic Pottery of Palestine to Halafian ware. Bulletin of the American Schools of Oriental Research 159: 32-36. Gophna R. and Sadeh S. K arsgaard P. 1988-1989 2010 Excavations at Tel Tsaf: An Early Chalcolithic Site in the Jordan Valley. Tel Aviv 15-16: 3-36. Goren Y. 1992 2006 K irkbride D. Petrographic study of the pottery assemblage from Munhata. In: Garfinkel Y. (ed.), The Pottery Assemblage of the Sha’ar Hagolan and Rabah Stages of Munhata (Israel): 329-360. Paris : Association Paléorient (Les Cahiers du CRFJ 6). The Technology of the Gilat Pottery Assemblage: A Reassessment. In: Levy T.E. (ed.), Archaeology, Anthropology and Cult: The Sanctuary at Gilat, Israel: 369-393. London: Equinox. Goren Y. and Cohen-Weinberger A. 2002 Petrographic Analyses of Selected Wares. In: K empinski A. (ed.), Tel Kabri. The 1986–1993 Excavations: 435-442. Tel Aviv: Tel Aviv university (Monograph Series of the Institute of Archaeology of Tel Aviv University 20). Gosselain O. P. 1992 2000 The bonfire of the enquiries. Pottery firing temperatures: what for? Journal of Archaeological Science 19,2: 243-259. Materializing Identities: An African Perspective. Journal of Archaeological Method and Theory 7,3: 187-217. Graham P. 2014 Archaeobotanical remains from late 6th/early 5th millennium BC Tel Tsaf, Israel. Journal of Archaeological Science 43,1: 105-110. 1969 1971 Tel Bet Yerah 2007, 2009. Hadashot Arkheologiyot. Excavations and Surveys in Israel 122 [http://www.hadashot-esi.org.il/ report_detail_eng.aspx?id=1553&mag_id=117, accessed on 04/03/2016]. Gustavson-Gaube C. 1986 Tell esh-Shuna North 1985: A Preliminary Report. Annual of the Department of Antiquities of Jordan 30: 65-113. Hanbury-Tenison J.W. 2011 Hitchings P.M.N., Abu-Jayyab K., Bikoulis P. and Banning E.B. 2013 A Bayesian Approach to Archaeological Survey in North-West Jordan. Antiquity 87,336, project gallery, http://antiquity.ac.uk/ projgall/hitchings336/, accessed on 25/04/2016. Entangled: An Archaeology of the Relationships between Humans and Things. New York: Wiley-Blackwell. Hours F., Aurenche O., Cauvin J., Cauvin M.-C., Copeland L. et Sanlaville P. 1994 Atlas des sites du Proche Orient (14 000-5700 BP). Lyon : Maison de l’Orient Méditerranéen (TMO 24). Network Analysis in Archaeology. New Approaches to Regional Interaction. Oxford: Oxford university Press. Köppel R., Mallon R. et Neuville R. 1940 Teleilat Ghassul II, compte rendu des fouilles de l’Institut biblique pontifical, 1932-1936. Rome: Institut biblique pontifical. Lave J. and Wenger E. 1991 Situated Learning. Legitimate Peripheral Participation. Cambridge: Cambridge university Press. Leenders R. 1989 The Red Wash Ware ceramic assemblage in Syria: A review. In: H aex O.M.C., Curvers H.H. and A kkermans P.M.M.G. (eds.), To the Euphrates and Beyond: Archaeological Studies in Honour of Maurits N. van Loon: 89-101. Rotterdam and Brookfield: A.A. Balkerna. Le Mière M. et Picon M. 1998 Les débuts de la céramique au Proche-Orient. Paléorient 24,2 : 5-26. Leonard A. Jr. 1989 1992 A Chalcolithic “Fine Ware” from Kataret es-Samra in the Jordan Valley. Bulletin of the American Schools of Oriental Research 276: 3-14. The Jordan Valley Survey, 1953. Some Unpublished Soundings Conducted by James Mellaart. Winona Lake: Eisenbrauns (The Annual of the American Schools of Oriental Research 50). Livingstone Smith A. 2001 2007 Hodder I. 2012 An Archaeology of Interaction: Network Perspectives on Material Culture and Society. Oxford: Oxford university Press. K nappett C. (ed.) 2013 1986 The Late Chalcolithic to Early Bronze I Transition in Palestine and Transjordan. Oxford (BAR Int. Ser. 311). Hauptmann A. 2007 The Archaeometallurgy of Copper: Evidence from Faynan, Jordan. Berlin: Springer. Early Byblos and the Beqa’a. Mélanges de l’Université SaintJoseph 45,5: 45-60. A Commentary on the Pottery Neolithic of Palestine. Harvard Theological Review 64,2-3: 281-289. K nappett C. Greenberg R. and Paz S. 2010 The Halaf-ubaid Transition: A Transformation without a Center? In: Carter R.A. and Philip G. (eds.): 51-68. Bonfire II: The Return of Pottery Firing Temperatures. Journal of Archaeological Science 28: 991-1003. Chaîne opératoire de la poterie. Références ethnographiques, analyses et reconstitution. Tervuren : Musée royal de l’Afrique centrale. Lovell J.L. 2001 2002 The Late Neolithic and Chalcolithic Periods in the Southern Levant. New Data from the Site of Teleilat Ghassul, Jordan. Oxford (BAR Int. Ser. 974). Shifting Subsistence Patterns: Some Ideas about the End of the Chalcolithic in the Southern Levant. Paléorient 28,1: 89-102. Paléorient, vol. 42.2, p. 151-184 © CNRS ÉDITIONS 2016 Contacts between the Southern and Northern Levant in the first half of the 5th millennium BC 2010 in the Era of State Formation: 149-179. Santa Fe: School of American Research Press. Community is Cult, Cult is Community: Weaving the Web of Meanings for the Chalcolithic. Paléorient 36,1: 103-122. Lovell J.L. and Rowan Y.M. (eds.) Noll W., Holm R. and Born L. 2011 1975 Culture, Chronology and the Chalcolithic. Theory and Transition. Oxford and Oakville: Oxbow Books (Levant Supplementary Series 9). Lovell J.L., Dollfus G. and K afafi z. 2004 2007 The Middle Phases at Abu Hamid and the Wadi Rabah horizon. Studies in the History and Archaeology of Jordan VIII: 263-274. The Ceramics of the Late Neolithic and Chalcolithic: Abu Hamid and the Burnished Tradition. Paléorient 33,1: 51-76. Lovell J.L., K afafi z. and Dollfus G. 1997 A Preliminary Note on the Ceramics from the Basal Levels of Abu Hamid. In: Gebel H.G.K., K afafi z. and Rollefson G. (eds.), Prehistory of Jordan II. Perspectives from 1997: 361-369. Berlin: ex Oriente. Lovell J.L., Meadows J. and Jacobsen G.E. 2010 upland olive domestication in the Chalcolithic period: new 14C determinations from el-Khawarij (Ajlun), Jordan. Radiocarbon 52: 364-371. Mallon A., Koeppel R. et Neuville R. 1934 Teleilat Ghassul I, compte rendu des fouilles de l’Institut biblique pontifical, 1929-1932. Rome : Institut biblique pontifical (Scripta Pontificii Instituti biblici). Kamid el-Loz, 13. The Prehistoric and Early Historic Context of the Site. Bonn: R. Habelt. Mathias V.T. 2003 The prehistoric pottery from trenches V, VI and VII. Part 2: The Later Prehistoric Pottery (Trenches VI). In: Parr P.J. (ed.), Excavations at Arjoune, Syria: 36-45. Oxford (BAR Int. Ser. 1134). Mellaart J. 1962 Preliminary Report of the Archaeological Survey in the Yarmouk and Jordan Valleys. Annual of the Department of Antiquities of Jordan 6-7: 126-157. Abu Ghosh, Jasmine Street: A Pre-Ghassulian Site in the Judean Hills. ‘Atiqot 82: 85-130. Nativ A., Rosenberg D. and Nadel D. 2014 The Southern tip of the Northern Levant? The Early Pottery Neolithic Assemblage of Tel Ro’im West, Israel. Paléorient 40,1: 99-115. Nigro L. e Panciroli B. 2001 Materiali archeologici dai primi scavi del Pontificio Istituto Biblico a Tuleilat el-Ghassul (1929-1938) nei Musei Vaticani. Bollettino dei Monumenti, Musei e Gallerie Pontificie XXI: 11-73. 2010 2011 2001 The Early History of the Ancient Near East, 9000-2000 B.C. Chicago: The university of Chicago Press. Cultural and Political Networks in the Ancient Near East during the Fourth and Third Millennia B.C. In: Rothman M.S. (ed.), Uruk Mesopotamia & Its Neighbors. Cross-Cultural Interactions Paléorient, vol. 42.2, p. 151-184 © CNRS ÉDITIONS 2016 The Emergence of ubaid Styles at Tell Kurdu: A Local Perspective. In: Carter R.A. and Philip G. (eds.): 293-309. The Chalcolithic of Southeast Anatolia. In: Steadman S. and McMahon G. (eds.), The Oxford Handbook of Ancient Anatolia 10,000-323 BC: 174-204. Oxford: Oxford university Press. Özbal R., Gerritsen F., Diebold B., Healey E., Aydın N., Loyet M., Nardulli F., R eese D., Ekstrom H., Sholts S., Mekel-Bobrov N. and Lahn B. 2004 Tell Kurdu Excavations 2001. Anatolica 30: 37-108. Perrot J., zori N. et R eich Y. 1967 Neve ur, un nouvel aspect du Ghassoulien. Israel Exploration Journal 17,4 : 201-232. Philip G. 2011 The Later Prehistory of The Southern Levant: Issues of Practice and Context. In: Lovell J.L. and Rowan Y.M. (eds.): 189-206. Philip G., Jabour F., Beck A., Bshesh M., Grove J., K irk A. and Millard A. 2002 Settlement and Landscape Development in the Homs Region, Syria: Research Questions, Preliminary Results 1999-2000 and Future Potential. Levant 34: 1-23. R ice P.M. 1987 Pottery Analysis: A Sourcebook. Chicago: university of Chicago Press. R ichter T., Garrard A.N., Allock S. and Maher L.A. 2011 Interaction before Agriculture: Exchanging Material and Sharing Knowledge in the Final Pleistocene Levant. Cambridge Archaeological Journal 21: 95-114. Rollefson G. 2008 The Neolithic period. In: A dams R.B. (ed.), Jordan. An Archaeological Reader: 71-108. London and Oakville, CT: Equinox. Rosen S.A. 2011 Desert Chronologies and Periodization Systems. In: Lovell J.L. and Rowan Y.M. (eds.): 71-83. Rosenberg D., Getzov N. and Assaf A. 2010 Nissen H.J. 1988 Sūkās 9: The Chalcolithic and Early Bronze Age Periods. Copenhagen: The Royal Danish Academy of Sciences and Letters (Publications of the Carlsberg Expedition to Phoenicia 11). Özbal R. Milevski I., Marder O., Mienis H.K. and Horwitz L.K. 2015 Painting of Ancient Ceramics. Angewandte Chemie 14,9: 602613. Oldenburg E. 1991 Marfoe L. 1995 183 New Light on Long-Distance Ties in the Late Neolithic/Early Chalcolithic Near East: The Chlorite Vessels from Hagoshrim, Northern Israel. Current Anthropology 51,2: 281-293. Rosenberg D., K limscha F., Graham P.J., Hill A., Weissbrod L., Ktalav I., Love S., Pinsky S., Hubbard E. and Boaretto E. 2014 Back to Tel Tsaf: A Preliminary Report on the 2013 Season of the Renewed Project. Journal of the Israel Prehistoric Society 44: 148-179. E. Gabrieli 184 Roux V. 2010 Technological innovations and developmental trajectories: social factors as evolutionary forces. In: O’Brien M.J. and Shennan S.J. (eds.), Innovation in Cultural Systems. Contributions from Evolutionary Anthropology: 217-234. Cambridge, MA and London: The MIT Press. Roux V., Courty M.-A., Dollfus G. and Lovell J.L. 2011 A Techno-Petrographic Approach for Defining Cultural Phases and Communities: Explaining the Variability of Abu Hamid (Jordan Valley) early 5th millennium cal. BC ceramic assemblage. In: Lovell J.L. and Rowan Y.M. (eds.): 114-132. Rowan Y. and Golden J. 2009 The Chalcolithic of the Southern Levant. A Synthetic Review. Journal of World Prehistory 22: 1-92. Santancreu D.A. 2014 Materiality, Techniques and Society in Pottery Production. The Technological Study of Archaeological Ceramics through Paste Analysis. Berlin: De Gruyter. Schaeffer C. F. A. 1962 Ugaritica IV. Découvertes des 18e et 19e campagnes, 19541955 : fondements préhistoriques d’Ugarit et nouveaux sondages, études anthropologiques, poteries grecques et monnaies islamiques de Ras Shamra et environs. Paris : Paul Geuthner. Schwarz G. and Weiss H. 1992 Syria, ca. 10000-2000 B.C. In: Ehrich R.W. (ed.), Chronologies in Old World Archaeology (3rd ed.): 221-243. Chicago: university of Chicago Press. Stein G. 2010 Local Identities and Interaction Spheres: Modeling Regional Variation in the ‘ubaid Horizon. In: Carter R.A. and Philip G. (eds.): 23-44. Streit K. and Garfinkel Y. 2015 Tel Tsaf and the Impact of the ubaid Culture on the Southern Levant: Interpreting the Radiocarbon Evidence. Radiocarbon 57,5: 865-880. Struever S. and Houart G.L. 1972 An Analysis of the Hopewell Interaction Sphere. In: Wilmsen E.N. (ed.), Social Exchange and Interaction: 47-79. Ann Arbor: Museum of Anthropology, university of Michigan. Sulaiman G. 2012 Tell al-Baharia: an important site in the Damascus Basin. In: Borrell T ena F., Bouso García M., G òmez Bach A., Tornero Dacasa C. and Vicente Campos O. (eds.), Broadening Horizons 3. Conference of Young Researchers Working in the Ancient Near East: 113-122. Bellaterra (Barcelona): universitat Autónoma de Barcelona (Congressos 8). Thuesen I. 1988 Hama. Fouilles et recherches de la Fondation Carlsberg 19311938. I. The Pre- and Protohistoric Periods. Copenhagen: Nationalmuseet (Nationalmuseets Skrifter Storre Beretringer XI). de Vaux R. 1966 Palestine during the Neolithic and Chalcolithic Periods. In: Edwards I.E.S., Gadd C.J. and Hammond N.G.L. (eds.), The Cambridge Ancient History, Volume I: 498-538. Cambridge: Cambridge university Press. Wengrow D. 2001 The evolution of simplicity: Aesthetic labour and social change in the Neolithic Near East. World Archaeology 33,2: 168-188. Wright G.E. 1937 The Pottery of Palestine from the Earliest Times to the End of the Early Bronze Age. New Haven: American Schools of Oriental Research. 1951 An important correlation between the Palestine and Syrian Chalcolithic. Bulletin of the American Schools of Oriental Research 122: 52-55. Yaroshevich A. 2016 `En zippori. Hadashot Arckheologyiot – Excavations and Surveys in Israel 128 [http://www.hadashot-esi.org.il/Report_ Detail_Eng.aspx?id=24979, accessed on 05/07/2016]. Yener K.A., Edens C., Casana J., Diebold B., Ekstrom H., Loyet M. and Özbal R. 2000 Tell Kurdu Excavations 1999. Anatolica 26: 31-117. Yoffee N. 2005 Myths of the Archaic State. Evolution of the Earliest Cities, States, and Civilizations. Cambridge: Cambridge university Press. Paléorient, vol. 42.2, p. 151-184 © CNRS ÉDITIONS 2016