Radiocarbon dating the Aurignacian sequence at Isturitz (France): Implications for the timing and development of the Protoaurignacian and Early Aurignacian in western Europe. (February 2018)

• Record Type: Journal Article
• Authors: Barshay-Szmidt, Carolyn; Normand, Christian; Flas, Damien; Soulier, Marie-Cécile
• Is Part Of: Journal of archaeological science. Volume 17(2018)
• Page Start: 809
• Page End: 838
• Publication Date: 2018-02
• Subjects: Aurignacian -- Pyrenees -- AMS radiocarbon dating -- Sample selection-methodology -- Bayesian modelling; Archaeology -- Periodicals; Archaeology -- Research -- Periodicals; 930.1
• DOI: 10.1016/j.jasrep.2017.09.003
• Languages: English
• ISSNs: 2352-409X

Abstract

Abstract: The site of Isturitz is clearly important to discussions of the emergence and development of the Aurignacian. It bears a long stratigraphic sequence of this period and has benefited from recent excavation and analysis. In this paper we present 18 new AMS radiocarbon dates (Normand excavation), covering the majority of the Aurignacian sequence at this site. Our dating was aimed at addressing two key questions of this period (1) what is the date of occupation of each of the Aurignacian variants (Protoaurignacian, PA, and Early Aurignacian, EA) at Isturitz and (2) how do the dates of PA and EA occupation at Isturitz compare to those of other nearby sites? To achieve this we dated well-provenienced, species-identified, humanly-modified faunal remains from layers of each Aurignacian variant at Isturitz, most including an ultrafiltration step. We built a Bayesian model from these to determine start/end dates of each layer/industry at this site. We also compiled a list of all recently-dated, ultrafiltered/ABOx-SC, more carefully sampled, results from PA and EA layers in France, Italy and Spain to see where Isturitz fit into these groups. Results indicate that at Isturitz, the PA started at 42.8–41.3 modelled BP (95% confidence interval) and the EA at least as far back as 41.6–39.7 modelled BP (95% confidence interval). These are among the earliest dates for both of these industries in western Europe and it is one of the only sites to have multiple old dates. Our results and comparisons confirm that the PA was the earlier of the two variants, but show that the EA and later PA overlap statistically within the western European region (between sites). Thus the possibility that the two variants existed partially contemporaneously must remain open as a working hypothesis. Whether the EA was initiated as a result of HS4 is equivocal with the new data. Highlights: Dated Protoaurignacian and Early Aurignacian layers at Isturitz, France Obtained 18 new AMS radiocarbon dates on identified, humanly-modified, bone samples Produced Bayesian model: start dates among earliest in western Europe Showed that the variants overlap statistically in region Showed that the Early Aurignacian may have started before the onset of HS4