Faunal Evidence and the Duration of Occupation: A Statistical Approach to Neanderthal Archaeological Sites

Abstract The duration of Neanderthal occupations of caves and rock shelters is one of the most debated issues in prehistoric archaeology. Several disciplines, including lithic analysis, micromorphology, zooarchaeology, taphonomy, and spatial archaeology, have proposed criteria to estimate occupation length, but their reliability varies. This study focuses only on indicators derived from faunal assemblages and evaluates their effectiveness through statistical testing. Four features were analysed: number of taxa, hunting spectrum, anthropogenic modifications, and evidence of carnivore activity. These characteristics were applied to a dataset from Late Mousterian sites in Spain and south-eastern France (70–40 ka) that have previously been examined through multidisciplinary approaches. The results indicate that the four features do not provide equally robust signals. Anthropogenic modifications and the number of exploited taxa show the strongest correlation with occupation duration. In contrast, the number of taxa and carnivore activity, when considered alone, are less reliable indicators. Multivariate analyses reveal a clearer structuring of the faunal assemblages. Brief stopping places are consistently distinguished from longer occupations. However, long-term occupations and repeated short-term occupations tend to overlap, largely because of palimpsest effects. Overall, these findings demonstrate both the potential and the limits of faunal criteria. They highlight the need for integrated, high-resolution approaches to reconstruct the duration and intensity of Neanderthal occupations.