The epi-Palaeolithic of Okuzini cave (SW Anatolia) and its mobiliary art.

Otte, Marcel ; Yalcinkaya, Isin ; Leotard, Jean-Marc 等

Late and Epi-Paleolithic sequence are well known from field work and publications in southeast Europe and the Levant. Current research in Anatolia promises to shed new light on the vast region that connects these two areas. At Okuzini cave a detailed sequence of Terminal Pleistocene and Early Holocene assemblages contributes greatly to our understanding.

Okuzini cave was found add first excavated during the 1950s by K. Kokten (1963). On the cave's interior, Kokten discovered a rock engraving which seemed to represent wild cattle, which gave the cave its name (Okuz means ox in Turkish). During these excavations, Kokten removed a large portion of the deposits from inside the cave which contained the remains of numerous occupations.

More recently, a small test excavation was carried out, limited to straightening a 1-m section by removing about 10-20 cm of the deposits from the face of the section, and carried out by a team from the University of Tubingen in co-operation with one of the authors (lY) from the University of Ankara (Albrecht et al. 1992). Since 1989, fieldwork has continued through a joint project conducted by the University of Liege and the University of Ankara.

Okuzini cave is situated only a few metres above the level of the alluvial plain in the foothills of the Taurus mountains (Figure 1). M. Pawlikowski's study (Yalcinkaya et al. in press) indicates that the cave was first opened during the Upper Pleistocene after a small river deviated from its main karstic channel, which now appears as a karstic spring. Following the drying of the cave, a major rockfall of large limestone blocks in the cave particularly affected the entrance and the terrace. The collapsed blocks and sloping walls limited the space available for human use and promoted certain spatial subdivisions which the current excavations hope to expose (Figure 2).

Stratigraphy

From the exposed stratigraphy, it seems that the accumulation inside the cave began on a level of collapsed rocks and was rather rapid (in Table 1 numerous [C.sup.14] dates provide the age of the deposits as recorded in the sections in Figures 3 & 5). The sediments were introduced mainly as detritus from the plateau and the slopes above the cave through a natural chimney at the back of the cave's roughly rectangular chamber, and through the numerous cracks in the bedrock. Human occupations, which seem to have been nearly continuous, contributed greatly to the rapid accumulation of sediment by the introduction of large quantities of organic matters such as firewood, as well as activities like knapping and dumping of animal bones. Numerous fire-places dot the stratigraphy from the lower levels to the top (Figures 3-4).

Schematically, the deposits represent three major ensembles. These probably mark changes in the rate of accumulation as well as shifts in human behaviour. Post-depositional effects include the climatic fluctuations which marked the late Pleistocene and early Holocene. In particular, one should note the increasing wetness at the top of the sequence.

* The lower ensemble seems to have accumulated in a relatively humid and cold environment. The sediment, mainly red in colour, due to the large amount of clay, contains some small angular limestone fragments. Hearths are not always well preserved; many seem to have undergone post-depositional processes and now appear as black and white striations encapsulated within the clayey deposits. * The middle ensemble is a major concentration and accumulation of angular rock fragments mixed with the remains of human activities including bones, lithics, mobiliary art, charcoal and land-snail shells. The heterogeneity in the size of the limestone blocks seems to indicate that, at least in part, they were transported to the site by humans. The large quantities of kitchen debris, containing numerous remains of Helix sp., resembles in a very general way shell middens of the latest European hunter-gatherers. There is only rare evidence for hearths, and one burial (no. 2) within this deposit. * The uppermost ensemble is dusty, grey in color, and about 30 cm in thickness. Within it, several pits have been recognized, and at least one burial of proto-historic age (no. 1). There is a clear association between the numerous sherds and a polished axe as well as a microlithic assemblage. This entity is related to the Neolithic or the Chalcolithic period of the region.

In general, the stratigraphy of most of the deposits seems quite horizontal and therefore relatively easy to excavate, although continuing water seepage has caused the accumulation of calcitic crusts and concretions.

Lithic industries and fauna

The Okuzini lithic and faunal sequence was sub-divided by combining stratigraphic evidence with the preliminary analysis of the lithic assemblages. While examining the techno-typological features, one of authors (J-ML) discerned a few shifts that enabled a tentative sub-division into four phases.

All the lithic assemblages were manufactured from cobbles of local flint collected in the environs of the site or brought from farther afield.

The following description of the four lithic phases from bottom to top, incorporates the preliminary results of faunal (see Tables 2-3 and Figures 6-7) and charcoal analysis (Yalcinkaya et al. in press). The faunal data were obtained on the material from a reference test (squares K5c and K5d) which provided around 10,000 bones. Further analysis confirmed the tendencies shown in tables and figures.

Phase 1, units XII to VII, dated to 16,000-14,000 BC (see Table 1)

[TABULAR DATA OMITTED]

The lithic industry is characterized by the production of elongated blades made of both imported and local radiolarite, shaped into straight-backed points with a retouched base. Other tool types made on blades include endscrapers, burins, perforators and truncated blades (Figure 8). Among the projectiles we have included the elongated triangles and arched backed points. The fauna comprises ovicaprines (mostly goat and also sheep).

Phase 2, units VI-V, dated to 14,000-13,000 BC (see Figure 8) Technically the blank production is the same as in the earlier assemblages but among the microlithic tools there are, in addition to the triangles, also trapezes and lunates. Again, ovicaprines dominate the fauna, but sheep is more present than in phase 1.

During the first two phases, Units XII-V, the fauna comprises mostly caprovines (80%) with secondary game (fallow deer) and smaller quantities of roe deer, red deer and hare. The macro-fauna reflects a landscape with gallery forests and steepcliffs (presence of goats), which is confirmed through charcoal analysis (Yalcinkaya et al. in press).

Grinding slabs and hammer-stones were found in this context. Of special interest are the elements of body decoration including stone beads and marine shells (Dentalium, Columbella rustica and Arcularia gibossula).

Phase 3, layers IV-Ia This is the gravelly units described above. The calibrated dates indicate an age in the range of 13,000-10,500 BC. The exhausted cores, made of local raw material, demonstrate blades removed from various directions. The blade blanks are generally short, about 50 mm in length. There is also evidence of the microburin technique used to make an oblique snap. Geometric microliths, mostly lunates, triangles, and trapezes, are very common. The other forms include end-scrapers, retouched blades, perforators, notched blades (Figure 8). Products of a bone industry are relatively abundant - awls, needles and spatulas. Fallow deer increases in frequency in the fauna, with the other forest animals. Ovicaprines represent 60% of the remains, with a more equal presence of goat and sheep.

To this phase we can attribute the incised pebbles collected by K. Kokten. These incisions are sometimes of geometric pattern and sometimes of a more figurative appearance (Figures 9-10; Marshack in press and see below). It seems that burial no. 2 is associated with this phase.

Phase 4, units 0-Ib This provided a range of calibrated radiocarbon dates from 90000 to 6500 BC. The industry seems to be a mixture of microlithic industries, plus a broken polished stone axe and numerous sherds (mainly associated with burial no. 1). Goat and sheep are equally Present, with a few forest animals.

Figure 9. 1 Early rendering of an incised Epi-Palaeolithic pebble from Okuzini (earlier excavations, no stratigraphic data) based on naked-eye observation, indicating an incorrect shape for the pebble and a crude representation of bovid morphology. The suggested presence of a female `buttocks' image is derived from female images of this type found in the west European Magdalenian. (After Anati 1968; 1972). 2 The incised composition on the same pebble as determined by microscopic study, indicating an accurate rendering of bovid morphology, an over-engraving or `renewal' of the muzzle, the addition of a second eye, and repeated symbolic `killing' of then by a series of schematic `weapons' incised into and around the animal. The composition suggests a well-developed tradition of representation and symbol use as well as hunting of the wild aurochs. 3 Schematic rendition of the engraved bovid muzzle, based on microscopic study. It indicates an original circular eye, the addition of an oval eye, a `stream' issuing from the original eye, and the over-engraving of an accurate muzzle by later crude, schematic additions. Comparable modes of image use are found among the mobiliary compositions and cave images of the late west European Upper Palaeolithic (cf. Lascaux).

The fauna of these two phases reflect an increased frequency of forest animals such as wild boar and fallow, red and roe deer. This trend is also recorded in the charcoal, which indicates the importance of oak and ash.

Iconography and symbol

Bostanci's early report (1959) describing an incised bovid on the wall at Okuzini and his discovery of a small pebble (5 cm) with an incised bovid in a level containing other forms of image and symbol (c. 13,000 BC) is significant. First, bovid bones were absent from the faunal remains. Second, early renderings of the bovid on the pebble (Figure 9.1) suggested an inaptitude for `art' and an unfamiliarity with the species. Third, the female `buttocks' image seen in the composition seemed to link it with similar female imagery in the contemporary Magdalenian cultures of western Europe. Microscopic analysis reveals that these early suggestions were in error. The head, posture and morphological outline of the bovid had been rendered accurately by a culture familiar with the species and having a well-developed tradition of depiction and image use (Figure 9.2).

The supposed rear line of the `buttocks' is, actually, a crack in the stone. The bent arm of the human, a slender male, is thrusting a spear into the chest of the bovid. Both surrounding and upon the animal are schematic weapons, and `wounds' of a type documented in the imagery of diverse historical hunting cultures as well as in the mobiliary materials and cave images of late Upper Paleolithic Europe. These abstract weapons and wounds, made in different styles, were apparently incised at different times, perhaps subsequent to the initial depiction of a killing by the human. The bovid was apparently `killed' symbolically a number of times.

The head and muzzle were interesting (Figure 9.3). The realistically rendered head and muzzle had been crudely, apparently hurriedly, over-engraved; the initial circular eye had had a second, oval eye added. The symbolic `renewal' or `re-use' of an animal image by adding parts of that animal is profusely documented in Upper Palaeolithic Europe (cf. Marshack 1969); Leroi-Gourhan & Allain 1979). The supposed `ladder' motif that was depicted on the muzzle are actually incised lines or 'tears' streaming from the circular eye.

The analysis therefore documents the conjoint use of realistic and schematic, abstract imagery as well as different modes of image use. The wild aurochs was clearly a symbolic animal, ritually killed and ritually renewed. It may be relevant that the bovid head is a crucial symbol at the Early Neolithic site of Catal Huyuk within the Taurus mountain range. A wild aurochs is also schematically depicted in a 'Palaeolithic' engraving within the coastal Anatolian cave of Beldibi, not far from Okuzini (Bostanci 1959). By contrast, the bovid is not often depicted during the Epipalaeolithic, Natufian or early Neolithic of Israel, to the south.

The complexity of these early Anatolian traditions is also documented on a limestone pebble found in the same level at Okuzini. Heavily hand-worn and polished along the edges, it is incised with different motifs and compositions on each face (Figures 10a, 10b). Microscopic analysis reveals that one face contains an accumulation of 12 circles (slightly triangular) incised within an encompassing circular containing line. That containing line is extended to meet a ladder-like, motif consisting of sets and subsets of marks. Similar `ladder-like' motifs are profusely accumulated on the obverse face, where three blocks of descending horizontal rows are carefully positioned on three different planes of the surface. Analysis of these `ladder-like' motifs indicates that they are composed of accumulated sets and sub-sets of marks, interspersed with `signs' and cueing-marks (Marshack in press). The sets, sub-sets, signs and cueing marks were apparently accumulated over a considerable period. On the first face, added to the initial accumulation of 12 circles are later single circles incised by different points. The association of circles with sets and sub-sets of marks, represents an entirely different form or mode of periodic ritual or symbolic marking than is found on the bovid pebble, or within the compositions found at Karain.

The `ladder-like' accumulation of sets and sub-sets occurs also in the Epipalaeolithic and Natufian of Israel. A huge block with this motif was excavated from an early Natufian level at Hayonim in the summer of 1994; a limestone plaque with this motif (still unpublished) was excavated on the Hayonim Terrace in the 1980s; a large limestone block with accumulations of the motif was excavated earlier within Hayonim cave (Belfer-Cohen 1991); a microscopic re-study of an Epipalaeolithic (Geometric Kebaran) pebble from the Israeli site of Urkane-rub II, c. 14,500 b.p. (Hovers 1990), has documented accumulations of the motif on both faces. Unstudied limestone fragments from Hayonim cave that contain accumulations of the `ladder-like' motif are in the Archaeological Institute of Hebrew University in Jerusalem.

A variable iconography and diverse modes of periodic image use and association are also documented at the Epipalaeolithic cave of Karain, not far from Okuzini, the images include anthropomorphs, `macaroni' motifs (possibly water), and a symbolic human hand (Kokten 1961; Marshack in press). Variants of these motifs also appear in the later Catal Huyuk imagery. The presence of a rich, variable iconography, and of diverse forms of periodic symbol use in the Mediterranean zone of the Near East before the beginnings of agriculture documents the presence of symbolling cultures that were far more complex than had been previously assumed on the basis of the numerical sparseness of artefacts and sites. That scarcity may be due to the sparsity of systematic excavation of Palaeolithic sites when compared to the profuse excavation of later cultures. It is of interest that recent excavations at Okuzini have uncovered incised motifs on bone (J.M. Leotard pers. comm.).

Discussion

The human occupation of Okuzini cave began when the main chamber dried. The sequence, dated to 16,000-7000 BC, is composed of anthropogenic remains, washed-in sediments and calcareous rock fragments detached from the walls and ceiling. The lithic industries represent shifts from microlithic, non-geometric assemblages made on both local and imported raw materials, to assemblages dominated by geometric microliths made from the local radiolarite. The terminal occupations include Neolithic-Alcolithic burials. The lithic industry of this phase begins with microlithic assemblages containing Neolithic elements. Preliminary analysis of cementum increments indicate that the site was probably occupied most often in the spring/summer season.

The Terminal Palaeolithic in the Antalya region is also known from earlier excavated sites such as Beldibi, Belbasi and Karain B. The reports from the first two indicate that the general regional sequence follows what is known from other areas. non-microlithic industries are followed by a dominance of the geometric forms. Long-distance comparisons with both the Balkans and the Levant reinforce this observation. However, since 13,000 bc, the Levant demonstrates a definite cultural change with the emergence of the Natufian culture. A complex settlement pattern of sedentary hamlets and seasonal camps characterizes the coastal ranges (e.g. Bar-Yosef & Belfer-Cohen 1992), and marginal areas favoured a more mobile settlement pattern (e.g. Henry 1989); Byrd 1989). The shift is also reflected in the material culture: the production of bladelets is mostly replaced by the appearance of short flakes, sickle blades, small picks, the elaboration of bone and horn industry and the use of mortar and pestles.

With regard to subsistence, the Natufian were gatherers of cereals, legumes, acorns and numerous other plants foods, as well as specialized hunters of gazelles and other species in more particular environments. During Late Natafian times we note in southeast Turkey early villages such as Hallam Cemi (Rosenberg & Davis 1992) that herald the beginning of the Neolithic.

Most of the layers at Okuzini are thus interpreted as the camp residues of foragers who hunted wild goat and sheep (and only rarely other species) and collected plant food, the remains of which are yet not identified. It seems that the advent of the Neolithic throughout the Anatolian plateau brought an end to the life-ways of these earlier foragers.

References

Albrecht, G. et al. 1992. Late Pleistocene and early Holocene finds from Okuzini: a contribution to settlement history of the Bay of Antalya, Turkey, Paleorient 18: 123-41. Anati, E. 1968. Anatolia's earliest art, Archaeology 21: 21-35. 1972. Arte Preistorica in Anatolia. Valcamonica: Edizioni del Centro. Studi Camuni 4. Bar-Yosef, O. and Belfer-Cohen A. 1992. From foraging to farming in the Mediterranean Levant, in A.B. Gebauer & T.D. Price (ed.), Transitions to agriculture in prehistozy: 21-8. Madison (WI): Prehistory Press. Belfer-Cohen, A. 1991. Art items from Layer B, Hayonim Cave: a case study of art in a Natufian Context, in O. Bar-Yosef & F.R. Valla (ed.), The Natufian Culture in the Levant: 569-88 Ann Arbor (Ml): International Monographs in Prehistory. Bostanci, E.Y. 1959. Researches on the Mediterranean coast of Anatolia: a new Palaeolithic site at Beldibi near Antalya, Anatolia 4(9): 129-67. Byrd, B. 1989. The Natufian: settlement variability and economic adaptations in the Levant at the end of the Pleistocene, Journal of World Prehistory 3(2): 159-98. Henry, D.O. 1989. From foraging to agriculture: the Levant at the end of the Ice Age. Philadelphia (PA): University of Pennsylvania Press. Hovers, E. 1990. Art in the Levantine Epi-Palaeolithic: an engraved pebble from a Kebaran Site in the Lower Jordan valley, Current Anthropology 31(3): 317-22. Kokten, I.K. 1961. Maras ve Antalya Vilayetleinde sureli Dip Tarih Arastirmalari hakkinda kisa bir Rapor, Turk Arkeologji Dergisi 11(1): plates XXXV, XXXVI. 1963. Die Stellung von Karain innerhalb der turkischen Vorgeschichte, Anatolia VII: 59-89. Leroi-Gourhan, A. & J. Allain. 1979. Lascaux inconnu. Paris: Editions du CNRS. Marshack, A. 1969. Polesini, a re-examination of the engraved Upper Paleolithic mobiliary materials of Italy by a new methodology, Rivista di Scienze Preistoriche In press. Okuzini: variabilite de catogorie dans l'imagerie symbolique, L'Anthropologie. Rosenberg, M. & M. Davis. 1992. Hallam Cemi Tepesi, an early aceramic Neolithic site in eastern Anatolia: some preliminary observations concerning material culture, Anatolica 18: 1-18. Yalcinkaya, I. et al. In press. Les occupations Tardiglaciaires du site d'Okozini (Sud-ouest de la Turquie). Resultats priliminaires des dernieres recherches. L'Aathropologie.