A Re-evaluation of the Upper Palaeolithic industries at Kebara Cave and their place in the Aurignacian culture of the Levant.

Abstract

This paper represents a re-evaluation of the Upper Paleolithic sequence (layers E, D2 and Dl) in Kebara Cave (Mt. Carmel) and its relation to the Aurignacian cultural complex of the Levant. The sequence of the cave was demonstrated previously by F. Turville-Petre (1932), D. Garrod (1954) and by M. Stekelis (1951-1967). The study was based on the collections made by Stekelis ; we used statistical and metrical data for analyzing the collections. In order to establish a stratigraphical frame for Stekelis'collection we used the original field-notes and the elevations of the different tool-complexes. By this method we established that the un-contaminated Upper Paleolithic assemblage lies between depth : 2.25 m and ca. 4.20 m, while the level of the uppermost uncontaminated Mousterian complex was at ca. 4.50 m (T. Schick, in press). The study of the material showed no sign of any typological or stratigraphie gap or any intermediate phase between the Upper Paleolithic and the Mousterian complexes. We arbitrarily divided Stekelis' collection into two phases : The upper-between 2.25 and 3.00 m ; The lower-between 3.00 and Ca. 4.20 m. The typological study of both divisions showed a clear difference between them ; the lower, being older than the upper, showed features similar to the lower Aurignacian phase (layer E of Garrod), while the upper showed more similarity to the Middle Aurignacian phase (layers D2 and Dl of Garrod) (all of Garrod's definitions were based on Turville-Petre's collections from the cave). We used the lower complex of Stekelis' collection to represent the lower Upper Paleolithic phase to be studied ; the D2 and Dl assemblages collected by Turville-Petre were used for the rest of the sequence. The typological data show a distinct difference between the material from layers E, D2 and Dl. The technological data point-out some-resemblance between the artifacts in D2 and Dl,but a distinct disimilarity to those in layer E. Both conclusions conflict with Garrod's views concerning the Levantine Upper Paleolithic ; she pointed-out the almost complete typological and technological similarities between D2 and Dl assemblages. We could correlate all three assemblages with the Levantine Aurignacian culture (šensu lato) - layer E being an early phase and Dl belonging to the complex previously called by Garrod and Neuville the "Atlitian" phase.