Journal of Archaeological Science
Volume 55, March 2015, Pages 122–134
Radiocarbon date frequency distributions and archaeological site counts are two popular proxies used to investigate prehistoric demography, following the assumption that variations in these data reflect fluctuations in the relative size and distribution of past populations. However, the two approaches are rarely applied to the same data-set and their applicability is heavily conditioned by the archaeological record in question, particularly research histories, agendas, and funding availability. In this paper we use both types of data to examine the population history of the Upper Palaeolithic hunter–gatherers (∼40 000–12 000 cal BP) of Southwestern France, comparing the demographic signatures generated.
Both proxies produce similar signatures across the Upper Palaeolithic sequence of the region, strengthening the interpretation of relative demographic changes as the cause of the pattern. In particular, a marked population decline is seen in both datasets during the Late Gravettian (∼28 000 cal BP), as well as a population increase in the Late Solutrean (∼25 000 cal BP) supporting the notion that the region acted as a population refugium during the Last Glacial Maximum. Where the two proxies diverge in the demographic signatures they produce, the radiocarbon date distribution shows peaks compared to troughs in site counts; the opposite pattern expected given taphonomic issues surrounding cultural carbon. Despite differences in chronological resolution and sampling bias, our data suggest that the two proxies can be considered broadly equivalent; a finding which warrants the investigation of prehistoric demography in regions where either extensive survey data or radiometric dating programmes are unavailable. While some preliminary observations are made, the impact of changing mobility on diachronic patterns seen in both proxies remains, however, difficult to assess.