Measuring spatial structure in time-averaged deposits insights from Roc de Marsal, France

Abstract

The use of space, both at the landscape and the site level, is considered an important aspect of hominin adaptations that changed through time. At the site level, spatial analyses are typically conducted on deposits thought to have a high degree of temporal resolution. Sites with highly time-averaged deposits are viewed as inferior for these analyses because repeated site visits obscure individual behavioral events. To the contrary, here, we take the view that behaviors that repeat themselves in a spatially structured way through time are exactly the kinds of behaviors that are potentially significant at an evolutionary timescale. In this framework, time averaging is seen not as a hindrance but rather as a necessary condition for viewing meaningful behavior. To test whether such patterning is visible in time-averaged deposits, we use spatial statistics to analyze a number of indices designed to measure lithic production, use and discard behaviors in a multi-layer, late Neandertal cave site in southwest France. We find that indeed some such patterning does exist, and thus sites with highly time-averaged deposits have the potential to contribute to our understanding of how hominin use of space varied through time. This is useful because a great many archaeological sites have highly time-average deposits. Interpreting the spatial patterning will likely require modeling to create expectations in time-averaged and likely emergent contexts such as these.