THE LITHIC TECHNOLOGY OF HOLLEY SHELTER, KWAZULU-NATAL, AND ITS PLACE WITHIN THE MSA OF SOUTHERN AFRICA

While the majority of research on the Middle Stone Age (MSA) in southern Africa has been conducted in the southern and western Cape, studies of the east coast of South Africa have become increasingly important due to the existence of well-stratified sites such as Sibudu. Because of the scarcity of comparable localities, however, we still know little about the spatial and temporal variability of MSA lithic technology in this region. We therefore chose to expand our research focus to other, lesser-known sites in the eastern part of South Africa. One such site is Holley Shelter which was excavated by Gordon Cramb between 1950 and 1960. Since its archaeological material was only studied in a cursory manner, we conducted a detailed technological study of the MSA lithic artefacts from Cramb's excavations, including attribute analysis and examination of reduction sequences. Our first aim was to assess the degree of potential mixing and recovery bias among the lithic material. We then characterised the different assemblages and investigated their diachronic variation throughout the occupation sequence. In order to obtain a rough age estimate of the so far undated sequence of Holley Shelter, we compared its lithic technology to other MSA sites in the eastern part of South Africa. Our results indicate three different phases of MSA occupation that vary in terms of raw material composition, core reduction, and tool manufacture. The assemblages are characterised by a blade and point technology that mostly derives from platform cores as well the highest proportions of splintered pieces reported from a southern African MSA site. The sequence does not feature Later Stone Age (LSA), Howieson's Poort, Still Bay or final MSA industries. Compared to other sites in the general region, the assemblages are most similar to lithic technology post-dating the Howieson's Poort, suggesting that the occupations fall broadly into the earlier part of MIS 3.