Assessing residue preservation and identification on stone tool assemblages from four different Late Pleistocene sites.

Residue analysis has become a widely applied procedure for reconstructing the lifecycle of prehistoric stone tools. The increasing popularity of the approach can be linked with methodological advances, which sometimes allow taxonomic identifications of observed residues. Initially, residue analysis was performed in combination with use-wear analysis to use multiple strands of evidence and filter out recent and environmental contamination. The last decade, a paradigm shift occurred, and residue analysis is now mostly performed without use-wear analysis based on the argument that the residue distribution on the stone tool surface allows distinguishing between functional and incidental residues. Moreover, these potentially ambiguous-looking residues are often only studied when adhering to the stone tool surface and are not extracted for further analysis. This paradigm shift in residue analysis has led to its disconnection from use-wear analysis, and the integration of the two approaches has become less common. In addition, these studies often deal with small samples usually limited to one layer within one site, hampering our understanding of the variability in residue preservation between geographical regions and across time. Here, we present a large-scale study of 1500 stone tools from four Late Pleistocene sites (Fumane (IT), Les Cottés (FR), Les Près de Laures (FR), Bushman Rockshelter (SA)) from two continents. The study aims at assessing existing residue analysis protocols against these assemblages and gaining insight into the mechanisms of residue preservation and alteration in various settings. Present-day residue analysis protocols were evaluated by confronting the residue data with data from independent use-wear analysis performed by different analysts. Residues were analysed using a wide range of techniques: stereo-, incident and transmitted light microscopy, biochemical staining and SEM-EDS. Residue interpretations were based on an extensive reference collection available at TraceoLab, University of Liège.