Human agency on landscape modification and land use is often seen in terms of socio-economic opportunities vs. natural constraints. In the study of prehistoric cultures this is both a strong source of information about sustenance strategies and community behaviours, and a subject potentially easy to analyse within a limited set of physical and social parameters. The recent advancements in the use of spatial analysis tools in landscape archaeology allow to obtain ever more precise models. However, studies that compare at the same time the geological landscape and social elements are very scarce. We used Point Pattern Analysis and Modelling to investigate megalithic structures (nuraghes) in Bronze Age southwestern Sardinia (Italy) and identify correlations between their spatial patterns and a set of covariates encompassing both environmental (i.e. topography and geological resources) and cultural factors. The models which best represent pattern distribution come from the combination of covariates from both groups.The models highlight a close distance from known ore deposits and show a clear dependence of Nuragic populations to ore extraction and metallurgy. The availability of fertile soils with moderate permeability and moderately low pH is also significant, as well as a preference to prominent locations with a positive correlation with the Topographic Position Index and the Convexity Index. From a cultural standpoint, we observed a consistent aggregation of simple nuraghes around complex nuraghes at mid-short distances. The occurrence of polycentric patterns can be explained either by the former emerging from the presence of the latter or vice versa, and is typically associated with a loosely stratified social structure devoid of strong hierarchies. These results underscore the efficacy of spatial analysis in disentangling and juxtaposing the physical and social factors influencing the distribution of past culture, and offer new insight on the development of Bronze Age societies in their geographical context.
Read full abstract