Did Gravettian apprentices make projectile elements in Tercis (Landes, France)?

1. Presentation of the site. The archaeological site of Tercis (Landes, France) has yielded several concentrations of lithic artefacts discovered since the 19th century (fig. 1). The majority of these substantial series were collected by C. Normand from the end of the 1970s onwards during the open cast mining of a large limestone quarry by the “Société des Ciments de l’Adour” (fig. 2). The spatial delimitation of these small concentrations scattered along Tercis ridge over a length of about one kilometer evokes rapid flint knapping activities, carried out in a few hours, providing evidence of short-lived installations linked to the presence of abundant flint on the anticlinal. Many human groups were thus attracted by this raw material since the early stages of the Palaeolithic. But the site was exploited on a larger scale during the Upper Palaeolithic, namely by Gravettians to whom most of the lithic industries can be attributed and notably the small assemblage presented in this paper. 2. Presentation of the assemblage with backed points. This lithic assemblage comes from a concentration which, unfortunately, was not entirely collected because of the short time allotted for rescue excavations by the Société des Ciments de l’Adour (fig. 3 and tab. 1). Tools make up a minority of the assemblage compared to flint knapping products, which is a logical consequence of the workshop context. Among tools, lithic projectile elements are largely dominant (fig. 4 to 6). The original aspects of this assemblage are a poorly executed flint reduction sequence associated with around thirty small backed pieces that are either unfinished and/or failed. These 34 unfinished and/or failed backed pieces are associated with seven cores (fig. 7 to 14), only one of which was productive (fig. 7). This core is also the only specimen which was not made in the grey-black Tercis flint used for other cores and all of the backed pieces. Moreover, a hundred relatively regular laminar blanks attest straight and regular blade production, with standardized dimensions around 2.5 inches long, 0.5 to 0.8 inches wide and 0.2 inches thick (fig. 16 to 19 and tab. 2). This production is also attested by the refitting of several flakes and laminar blanks which yield a reverse image of knappers’ intentions (fig. 15). 3. An example of projectile weapon elements apprenticeship ? Within the framework of this article, we wish to introduce a palaeo-sociological reflexion based on the accurate presentation of the data. Indeed, this lithic assemblage displays several paradoxes. Firstly, knappers used poor quality flint in a knapping workshop context which is paradoxically suitable for fine production. Secondly, two types of flint exploitation can be observed : on one hand, an exploitation of good quality blocks of raw material which supplied good blanks, such as cores 1, 8 and 9, as shown by the considerable proportion of laminar blanks (fig. 7, 15 and 16). On the other hand, the majority of refitted cores and products testify to the complementary existence of the local exploitation of low quality blocks or flakes (fig. 8 to 14). Finally, straight and regular laminar blanks are not used for backed pieces. What would have motivated knappers to use low quality flint and to select the less regular blades and bladelets to make their backed pieces ? The evidence provided by this assemblage point towards the appealing hypothesis of apprenticeship (fig. 20) which has been poorly documented archaeologically up until now (Pigeot 1988 ; Ploux 1991 ; Pelegrin 1995 ; Klaric 2006, in preparation). Although we cannot definitively advocate this latter hypothesis, we nevertheless propose to assess the archaeological facts, which are, as of yet, restricted, in order to propose a relevant methodology for the restitution of gestures and intentions.