This thesis presents new early Holocene excavations at Ntloana Tšoana in western Lesotho and makes the case for reintegrating method and theory in southern African Later Stone Age archaeology. It is argued that the distancing of method and theory relates to discipline-wide economic and material culture biases. An alternative is proposed which focuses on the everyday residues and temporal relations of hunter-gatherer life in order to build an understanding of attachment to place. This approach, which is more accepting of the multitemporal qualities of complex stratigraphies, is ideally suited to the abundant and wellstratified rockshelters of southern Africa. Importantly, tracing sequence in the construction of stratigraphic diagrams, the modelling of radiocarbon dates, as well as in studies of artefact production use and discard, remain central to the work presented here. At the same time, however, equal weight is placed on the durational and imbricated qualities of the archaeological record, which, it is argued, deserve to be repositioned as positive attributes. In this way of thinking sedimentation and erosion are vital to the human story and provide much of the movement and depth to the history-writing process. These insights are brought to bear on the 600-1700 calendar years of early Holocene stratigraphy at Ntloana Tšoana, during which time a complex history of social fragmentation and coalescence is traceable, sandwiched between, and partly overlapping with, two major flooding events. The site also becomes associated with a specific set of practices over many centuries involving the construction of large hearth features and the working of animal hides. Discussion revolves around the manner in which rockshelter residues may have acted as a stabilising force within the fluid socio-material world of hunter-gatherer groups and how this may have operated over an inter-generational timeframe. Also, a big thank you to Jess Meyer who took most of the site and landscape photos in this thesis. In Lesotho there are far too many people to mention individually, but to all of you, Kea leboha haholo! At home I had the constant support and love of Sarah and my two sons, Joseph and Rory, both born during the course of this project and who are a never-ending source of joy. I would also like to say a special thank you to my mum, who always loved unconditionally and never tried to make me do something more 'sensible'. I would like to dedicate this thesis to two women who made it all possible. Firstly, Annie Booth, and secondly, my mother, Hilary Bathurst, both of whom showed incredible strength at the end of their lives and will not be forgotten.