Search

Bones collected on Palaeolithic archaeological sites are rarely found whole. While a range of natural factors and mechanisms can be the cause of this bone fragmentation (e. g. carnivores, weathering, and sediment compaction), human activity is often a major agent, as bone fracturing takes place in many operational sequences involving the transformation and exploitation of animal resources, particularly in terms of the use of bone marrow and the bone itself as a raw material. The bone marrow contained in the medullary cavity of long bones is an important food resource for many human groups, particularly in periglacial areas, such as among the Nunamiut of Alaska or the Siberian Koriak. These groups each use different processes to remove the bone marrow. This variability is partly explained by the technical traditions and modes of consumption specific to each group, which often lead to a standardization of fracturing techniques within the group. Can recurring patterns of this kind be observed in the Palaeolithic context ? And if so, is it possible to demonstrate the transmission of such processes in terms of the removal of the bone marrow ? To help answer such questions, we present in this article a detailed study of the percussion marks observed on saiga antelope (Saiga tatarica) long bones from the upper complex at Saint-Germain-la-Rivière (Gironde), attributed to the Middle Magdalenian. This includes an inventory of the number of impact marks on each fragment, their nature, their arrangement in relation to each other, and their location on the anatomical element. The degree of percussion has been calculated for the different parts of the anatomical elements in order to establish whether there were any preferred impact zones. The patterns observed were then compared with those described in publications in order to establish their specific characteristics and to assess the parameters that may have affected such distributions of marks. Saint-Germain-la-Rivière has already been the subject of a detailed taphonomic study (Costamagno, 1999) demonstrating the significant role of Magdalenian hunter-gatherers in the accumulation and modification of bone remains and the limited impact of carnivores. This homogeneity in the constitution of the assemblage reduces the risk of confusion between human and natural agents in the attribution of the marks observed. The corpus studied was composed of 640 bone remains, all of which were fragments of the main long bones containing bone marrow (humeri, radioulnae, femurs, tibias, and metapodials). The faunal remains at Saint-Germain-la-Rivière were intensely fractured and present numerous human percussion marks that can be attributed to this activity of fracturing the bone. Almost 19% of the corpus presented such marks. No correlation was observed between the distribution of the percussion marks and the conservation/ identification of the different skeletal parts. Recurrences in the location of the impact points were nonetheless identified. These recurrences varied in relation to the skeletal elements considered and differed in degree. In the case of the radii, for example, the vast majority of impact marks were found on the proximal parts, in contrast with the humeri and tibias for which the distribution of percussion marks, although more variable, appeared to indicate a selection predominantly in favour of the distal parts. The preferred percussion surface also varied according to the element, mainly the anterior surface in the case of the radii but the lateral or medial surfaces in the case of the humeri. The existence of such recurrences indicates that the arrangement of the impact points results from deliberate choice. This choice of location to perform the fracturing of the bone does not appear to be dictated solely by mechanical constraints (e. g. bone structure) or ergonomic constraints specific to each skeletal element. If that were the case, the patterns observed for a specific species would be relatively similar from one level to another and one site to another. However this is not the case, as at Saint-Germain-la-Rivière, for a single taxon, the saiga antelope, the pattern observed in the upper complex (Middle Magdalenian) differs from that observed in the lower complex (Lower Magdalenian). The most striking case is that of the radii, as between the two complexes the arrangement of the impact points is diametrically opposed. This variability is not specific to this species as at other sites attributed to different techno-complexes, differences in the choice of the location of the percussion marks have also been observed, particularly for reindeer and ibex. This is also the case for sites attributed to the same period. Thus, while we cannot completely exclude the possibility of a preferential choice in relation to the bone structure itself, the existence of such inter and intra-species variability suggests choices dictated in part by cultural factors. Furthermore, the analysis of percussion marks has allowed us to highlight in some cases the presence of several impact points on the same bone fragment. Thus, some of the bone fragments have multiple percussion marks (up to eight identified), sometimes located over a small area and in the form of chopmarks. This type of mark is generally attributed to the disarticulation stage of the limbs during the butchery process. Given their location on the bone element, this does not however appear to be the case here. Moreover, they do not appear to be of a technical nature and cannot be attributed to retouchers. One of the hypotheses that we can put forward here is that several attempts were made at fracturing the bone. This then raises the question of the tenacity and mastery of the technique and could imply that these bones were fractured by novices., La fracturation osseuse est une pratique couramment observée sur les gisements paléolithiques, celle-ci intervenant dans diverses chaînes opératoires de transformation et d’utilisation des ressources animales, notamment alimentaires. La moelle osseuse, contenue dans la cavité médullaire des os longs, est une ressource alimentaire non négligeable pour de nombreux groupes humains, en particulier dans les zones périglaciaires, tels les Nunamiuts d’Alaska ou les Koriaks de Sibérie. Différents procédés sont mis en oeuvre pour sa récupération ; leur variabilité s’explique en partie par l’existence de traditions techniques et de modes de consommation de la moelle, propres aux différents groupes. Ces traditions entraînent souvent une standardisation des procédés de fracturation à l’intérieur de chaque groupe. De tels schémas récurrents sont-ils perceptibles dans le registre paléolithique ? Si oui, est-il possible de mettre en évidence une transmission des gestes relatifs à la récupération de la moelle ? Afin d’apporter des éléments de réponse à ces questions nous proposons dans le cadre de cet article, une étude détaillée des traces de percussion observées sur les os longs d’antilope saïga (Saiga tatarica), provenant de l’ensemble supérieur Magdalénien moyen de Saint-Germain-la-Rivière (Gironde). Les schémas observés ont ensuite été comparés à ceux décrits dans la littérature afin d’apprécier leurs spécificités et de discuter des paramètres ayant pu influer sur cette répartition des traces. Les vestiges fauniques de Saint-Germain-la-Rivière sont intensément fracturés et présentent de nombreuses traces de percussion d’origine anthropique attribuables à cette activité de fracturation de l’os. Des récurrences dans la localisation des traces de percussion ont pu être mises en évidence. Celles-ci varient selon les éléments squelettiques considérés et sont plus ou moins marquées. L’existence de telles récurrences tend à prouver que la disposition des points d’impact dépend d’un choix délibéré. Ces choix ne semblent pas être dictés exclusivement par l’existence de contraintes mécaniques (e. g. la structure osseuse) ou ergonomiques spécifiques aux éléments squelettiques. Par ailleurs, l’existence d’une variabilité inter-et intra-spécifique, observable d''après les données de la littérature laisse à penser qu’il s’agit de choix dictés en partie par un facteur culturel. L’analyse des traces de percussion a également permis de mettre en évidence, dans certains cas, la présence de plusieurs points d’impact sur une surface réduite. L’une des hypothèses que l’on peut émettre ici serait qu’il y ait eu plusieurs tentatives pour fracturer l’os. Ces pièces posent alors la question de l’acharnement et de la maîtrise du geste et pourraient évoquer des os fracturés par des novices., Masset Caroline, Costamagno Sandrine, Cochard David, Laroulandie Véronique. La fracturation osseuse : du fait technique à l’essai d’interprétation sociétale : l’exemple de l’antilope saïga du gisement magdalénien de Saint-Germain-la-Rivière (Gironde). In: Bulletin de la Société préhistorique française, tome 113, n°4, 2016. pp. 691-712.

Mémoire numérisé par la Division de la gestion de documents et des archives de l'Université de Montréal.