A late Middle Stone Age artifact assemblage from Sibudu (KwaZulu-Natal) : comparisons with the European Middle Paleolithic

International audienceIt has been suggested that many behavioral innovations, said to appear during the late Middle Stone Age in sub-Saharan Africa, facilitated the expansion of anatomically modern humans from Africa and the Near East into Europe at about 50 kyr; the process eventually led to the replacement of Neanderthals by modern humans and the emergence of the Upper Paleolithic. However, assemblages in this time range are little known in South Africa. In fact, the transition from Middle to the Later Stone Age in Southern Africa is controversial. The early appearance in South Africa of many innovations, such as sophisticated knapping techniques (e.g. the use of soft hammer or indirect percussion in blade production, of composite tools, of microlithic and bladelet technologies) remains to be established through technological analysis. We present here the first results of a project designed to carry out detailed technological studies of several lithic assemblages in South Africa and France dated to the transition period. At this time we have completed the study of a post-Howiesons Poort assemblage from the rock shelter site of Sibudu. The O2 m deep stratigraphic sequence of Sibudu extends from Howiesons Poort at its base to final Middle Stone Age, directly under Iron Age layers. We have analyzed in detail layer RSP (ca. 53 kyr, 1 m above the Howiesons Poort levels) which has provided a large assemblage of several thousand stone artifacts. Compared to published MSA assemblages this industry is unusual for the very high proportions of retouched pieces (15%). The technology is not very elaborate and there is no strong standardization of the end-products. There are no flakes of predetermined shapes; retouch is used to modify irregular flakes to obtain desired edges. Knapping of flakes and blades is done by hard hammer; soft hammer is used only for retouching tools. Interestingly the older Howiesons Poort blades were produced on the same raw materials by soft hammer. Raw material (hornfels and dolerite) was procured from distances of less than 20 km. Unifacial points are the dominant type and there is strong evidence of hafting and use as spear armatures. Detailed comparisons with Middle Paleolithic assemblages of Western Europe show that the late Middle Stone Age technology in South Africa is very similar to that of the Middle Paleolithic; in fact we see no fundamental differences between the two entities, as far as lithic technology is concerned. Implications for the Out of Africa hypothesis are discussed