This article presents the initial results from the S2AGES data base of calibrated radiocarbon estimates from western Europe in the period 25,000–10,000 years ago. Our aim is to present a population history of this sub-continental region by providing a chronologically-secure framework for the interpretation of data from genetics and archaeology. In particular, we define five population events in this period, using dates-as-data, and examine the implications for the archaeology of Late Glacial colonization. We contrast this detailed regional approach to the larger project which we call the cognitive origins synthesis that includes historical linguistics in the reconstruction of population history. We conclude that only archaeology can currently provide the framework for population history and the evaluation of genetic data. Finally, if progress is to be made in the new interdisciplinary field of population history then both disciplines need to refrain from inappropriate agricultural thinking that fosters distorting models of European prehistory, and they should also pay less, if any, attention to historical linguistics.

With comments from L.G. Straus, J.-P. Bocquet-Appel, P.A. Underhill & R. Housley and followed by a reply from the author.