On the operation of retouch in southern Africa's Early Middle Stone Age

The southern African Early Middle Stone Age (~315-80 ka) is often regarded as a period of behavioural stasis. Though regionalised technological and behavioural adaptions are identified throughout this period, there appears to be a lack of coherent and regular turnover of technological systems that becomes common in later periods. Here we test if the perception of Early Middle Stone Age technological stasis may be influenced by the typological approaches to the retouched implements that are frequently used as markers of technological change. We deploy an attribute analysis on 498 retouched implements from three Early Middle Stone Age assemblages from the Doring River Catchment, South Africa to test three hypothetical explanations of variation in implement form: strict typology, reduction-mediated typology, or maximum expediency. We find strongest support for a maximum expediency model in which retouch was flexibly applied across multiple retouch episodes, facilitated by preferential selection of larger blanks, producing a range of outcomes that rarely conform to classic types. These results encourage an interpretation of Early Middle Stone Age technology as representing a flexible and widely effective technological system, the subtleties of which have been masked by an historical over-emphasis on the limited retouched component.The southern African Early Middle Stone Age (~315-80 ka) is often regarded as a period of behavioural stasis. Though regionalised technological and behavioural adaptions are identified throughout this period, there appears to be a lack of coherent and regular turnover of technological systems that becomes common in later periods. Here we test if the perception of Early Middle Stone Age technological stasis may be influenced by the typological approaches to the retouched implements that are frequently used as markers of technological change. We deploy an attribute analysis on 498 retouched implements from three Early Middle Stone Age assemblages from the Doring River Catchment, South Africa to test three hypothetical explanations of variation in implement form: strict typology, reduction-mediated typology, or maximum expediency. We find strongest support for a maximum expediency model in which retouch was flexibly applied across multiple retouch episodes, facilitated by preferential selection of larger blanks, producing a range of outcomes that rarely conform to classic types. These results encourage an interpretation of Early Middle Stone Age technology as representing a flexible and widely effective technological system, the subtleties of which have been masked by an historical over-emphasis on the limited retouched component.