Dating the stone age at Rose Cottage Cave South Africa - An exercise in optically dating cave sediments

Abstract

Results and analysis of Optically Stimulated Luminescence (OSL) dating of the Middle Stone Age (MSA) and Later Stone Age (LSA) sediments at Rose Cottage Cave (RCC) are presented. Seventeen luminescence samples taken over the last decade were used in this study. Fourteen of these samples were dated (eight in Pretoria and six in Risø, Denmark by A. S. Murray). The samples were taken from the entire sequence and gave age ranges from the bottom of the sequence up until the Oakhurst LSA industry. The protocol that was used is the conventional single-aliquot regenerative (SAR) protocol, due to its ability to correct for behavioural problems associated with OSL dating. This study is primarily concerned with testing the validity of the conventional SAR protocol applied to South African archaeological sites. RCC presents problematic sediments for OSL dating because of a high feldspar component in the sediments at the site (OSL dating is preferably done on quartz grains due to a better understanding of the mechanisms of OSL production). Assessing the radiation dose samples received during their burial period was problematic due to the large presence of potassium rich feldspars. Assessing the radiation dose was problematic because the measurement of potassium (K) returned different values using several independent techniques. The mode of sediment deposition via different depositional mechanisms such as, fluvial, and clast spalling present difficult challenges in assessing the zero age value of a sample. OSL ages were compared to a well defined radiocarbon chronology from RCC, and any inconsistencies would motivate closer sorting of the different dating techniques. In this study it was found that not all feldspar grains were removed from the quartz extracts during pretreatment procedures. The ability of the SAR protocol to pick out feldspar contamination was therefore not conclusive, and single grain measurements had to be used to differentiate quartz and feldspar grains. The likelihood of age contamination from problematic depositional events was not supported and the results suggest that aeolian deposition was the main mechanism at the site. The vertical separation of depositional events i.e. the varying archaeological events, is very dense at RCC and this introduces the possibility of sample mixing during collection. For a few samples it is shown that sample mixing has occurred as mixing is usually evident in the degree of scatter in the OSL results. After all the inconsistencies in OSL/Radiocarbon age correlations were worked out, a coherent OSL chronology was obtained. Certain issues surrounding dosimetry however, are still not resolved. These issues are beyond the scope of this study and so caution is advised when using OSL dates done with little or no dose-rate analysis. The resulting dates provide a useful dataset for archaeologists who now have added resources to asses the Middle Stone Age (MSA) and better compare synchronous evidence from different sites in order to contribute to the debate surrounding the origins of modern humans and modern human behaviour. These results combined with the well established radiocarbon chronology give age ranges as follows: The Pre-Howiesons Poort (Pre-HP) MSA IIb industry is between 94 and 68 thousand years ago (ka); the Howiesons Poort (HP) industry is between 68 and 55 ka; the Post –HP MSA III dates to between 55 and 48 ka; the ‘almost sterile sands’ (which include the MSA IV industry) are between 48 and >27 ka; the MSA/LSA transition is between 27 and 20 ka; the Robberg LSA industry is between 20 and 10.5 ka; the Oakhurst LSA industry ranges from 10.5 to 8.5 ka; and the Wilton LSA industries are <8.5 ka.