Our thanks go to the French Ministry of Culture, DRAC-Service régional de l’archéologie des Hauts de France (Amiens site), Conseil général de l’Aisne, ASAVA, Archaeology Unit of Château-Thierry (Sébastien Ziegler), Cédric Salot for his logistical help, Françoise Bedel and Vincent Desabeau, owners, and all the 2011-2014 diggers.
Our warmest thanks also to Jean Clottes for his proofreading of the English translation and to the two reviewers for their useful advice.
1In 2012, two pieces related to elements of the bone industry were discovered at the «Le Colombier» site at Chézy-sur-Marne (Aisne, France) occupied during the Last Glacial Maximum. During the following campaigns, 2013 and 2014, three new artefacts were discovered and could be integrated into this set. To date, the series therefore consists of five pieces: four, made of reindeer antler, are waste products or proven tools; one, made of bone, raises questions about its status as a tool or butchery waste. All of them show technical and typological particularities that give, for this period in the north of France, a new character to this assemblage. It therefore seemed useful to us to make a detailed analytical presentation of it, which is the subject of this article.
2The « Le Colombier » open-air site is located in the south-east of Picardy, more precisely in the south of the department of Aisne, at an equal distance between Paris and Reims (fig. 1), in the municipality of Chézy-sur-Marne (Aisne) about 7km southwest as the crow flies from Château-Thierry. It is located at the top of a hillside, at an altitude of about 100m NGF on a slight flatness of a promontory overlooking the Marne valley, at the confluence of the latter with its tributary, the Dolloir. The archaeological diagnosis carried out in a preventive framework by Clément Paris in November 2010 confirmed the existence of an open-air occupation of the early Upper Palaeolithic and made it possible to limit, from a sedimentary point of view, the potential extension of the deposit over an area of approximately 1,300m2 (Paris et al. 2011). Four planned excavation campaigns (2011 to 2014) have made it possible to study nearly 70m2 and collect about 5,000 pieces, several of which are associated with anthropogenic layouts (fig. 2).
Figure 1. « Le Colombier », Chézy-sur-Marne (Aisne, France). Site localization (Computer graphics C. Montoya in Montoya et al. 2019).
« Le Colombier », Chézy-sur-Marne (Aisne, France). Localisation du site (infographie C. Montoya in Montoya et al. 2019).
Figure 2. « Le Colombier », Chézy-sur-Marne (Aisne, France). Plan and section (Computer graphics C. Montoya).
« Le Colombier », Chézy-sur-Marne (Aisne, France). Plan de la fouille et section-projection Ouest/Est. Les lignes matérialisent les remontages (infographie C. Montoya).
3Most of the sedimentation is represented by laminated brown silts of the Middle Pleniglacial (55-35 ka BP) capped by carbonate deposits of the Upper Pleniglacial (35-15 ka BP). These are kept only at the top of the slope. It is within these that the site of Chézy-sur-Marne «Le Colombier» is located, between 40 and 80 cm deep. Initially, two archaeological levels had been identified at the excavation on altimetric bases. It turns out out after analyses and reassembly that the small batch of pieces identified (50 pieces) as level A, is the probable result of a solifluction phenomenon. Level B is the main level of the site, which is distributed from a double depression: a small 10m2 basin with two equipped fireplaces, and a vast depression whose limits are not yet known (fig. 2). The two sectors have been linked by numerous refittings concerning several categories of remains (lithics, fauna, shells).
4Nine AMS dates - ranging from 23,582-22,952 intCal13 (19,320 ± 100 BP) to 20,711-20,303 Cal.BP intCal13 (17,010 ± 60 BP) - are currently available, making the Chézy-sur-Marne «Le Colombier» one of the few sites of the ancient Upper Palaeolithic in northern France dated to the Last Glacial Maximum (Montoya et al. 2019). At present, it is also the most northernly dated site in France for this period. It thus offers a rare opportunity to collect information on the behaviours developed by these poorly known societies from the early Upper Palaeolithic period in northern France (Montoya et al. 2019 ; Montoya et al. in press) and to reassess our knowledge of the human occupation of northern Europe at the Last Glacial Maximum, by completing the only site previously known of Wiesbaden-Igstadt in Germany (Street and Terberger 1999; Terberger 2013).
- 1 Studied by O. Bignon and published in Montoya et al. 2019
- 2 Studied by C. Montoya and published in Montoya et al. 2019
- 3 Studied by A. Pasquini and published in Montoya et al. 2019
- 4 Studied by C. Peschaux and published in Montoya et al. 2019
- 5 Studied by G. Dumarçay and published in Montoya et al. 2019
5The many faunal remains - where reindeer and horse predominate1 - were left between spring and autumn. Lithic production2, structured around operational sequences, essentially provided retouched tools with many transverse burins recognized as multifunctional tools (for cutting, scraping, grooving3), mainly involved in the processing of secondary animal resources (bone materials and treated skins). The shell ornament industry4 is divided into ornamental elements (gastropods and scaphopods) or tools (large cardium). Finally, the analysis of combustion structures5 indicates heating temperatures between 300 and 600°C and, for the main fireplace (KL/20-21), the existence of at least three operating episodes. In the end, the prehistoric occupation of « Le Colombier » appears to be closely correlated to the presence of many resources offered by the sector: flint, limestone, marl and sandstone deposits, shell outcrops of the Lutétien and Bartonien, migratory passages of mammalian fauna strategically observed by the dominant position of the site in the Marne valley (Montoya et al. 2019).
6In addition to these activities, there also is the processing of bone materials, as evidenced by the bone industry artifacts.
7Of the five pieces in the series, only one tool - a perforated baton fragment (CSM2014-n°6219-I19) – has been identified. Three other pieces, made of reindeer antler, are waste products : a block in the process of debitage (« matrice » CSM 2014 - n°7322 - D20), another possible extraction « matrice » (CSM2012-n °2710- GH20), a cut fragment of bez tine_ (CSM2014 - n°7117 - D20). The last (CSM2012), in bone, raises questions about its status as a tool or butchery waste.
8The pieces studied suffer from alterations of various origins: biological (vermiculations due to the action of plants), mechanical (fractures under the weight of sediments, cracks) and chemical (dissolution, manganese or iron deposition, concretion). There is a significant fragmentation of the two pieces found in the eastern part of the excavated area. The perforated baton has suffered a little less from this type of weathering and has a relatively good surface condition. The particular burial context of this tool -in a pit structure, framed on either side by large blocks of stone- will have provided the tool with a « protected environment » favourable to its better conservation.
9Despite the general poor surface condition of most of the artifacts, the clairvoyance of the excavation manager who, questioning the presence or not of stigmas of anthropogenic origin, ensured that careful sampling and packaging were carried out to minimise the risks of post-selection deterioration is commendable. Moreover, from the first campaign, these remains have benefited, since the dismantling in the field, from a restoration-conservation treatment carried out by a professional. It is also worth recalling that, in addition to the proper preservation of bone material, this rapid intervention by a specialist avoids the loss of information and guarantees the best conditions for reading all the technical and non-technical elements, including the most tenuous ones, that the exhumed parts still carry. The early implementation of this type of provisional measures in the context of planned excavations is sufficiently rare to be highlighted and welcomed here. The other consolidation steps (joining of newly formed fragments) can thus be carried out in complete serenity at a later date.
10Taking into consideration the entire industry, i.e. all the witnesses of the transformation (waste, blanks, roughouts and finished objects), the technological analysis makes it possible to identify and characterize from a technical and economic point of view the different operations of the transformation chains: acquisition and choice of raw material, technical means and methods used (for preparing, blank productioning, shaping, finishing). By combining these data with those collected on the use of finished objects, their repair and abandonment, the entire bone material exploitation chain can be traced, significantly enriching knowledge of the socio-economic systems of these prehistoric peoples. However, the small number of pieces available to date at the Chézy-sur-Marne « Le Colombier » gives an insufficient representation of two major categories of technological analysis -waste and finished objects- to carry out the refitting (by default) necessary for its realization. The absence of the two intermediate categories (blanks and roughouts) only reinforces this observation. It will therefore be necessary to wait for the discovery of other bone industry pieces to conduct such an approach. Nevertheless, each piece contains a great deal of information that can already be used to better define their integration into the site’s occupation (acquisition of raw materials, relationship with other industries produced, etc.) and their possible cultural attribution and, consequently, that of the groups that have settled at Chézy.
11The pieces of «Le Colombier» were therefore studied by means of a technological reading_ that made it possible to characterize, for each piece, its conservation state (taphonomic approach), its raw material (species and anatomical origin, state of freshness at the time of work), its technical stigmatas (type, location, position, extent, organization, orientation) to which are added the macro functional traces for the finished object. Each piece was observed with the naked eye, then under a stereoscopic microscope (Nikon SMZ 1500, magnifications ranging from x3.75 to x50). All the observations were carried out within the Imaging and Optical Microscopy Department of the Maison Archéologie et Ethnologie, René-Ginouvès, at Nanterre (CNRS USR 3225).
12Artifact n°6219 could unambiguously be identified as a mesio-distal fragment of a perforated baton (fig. 3-1) made from the right antler of a reindeer. Exhumed in 2013 within square I20 (on the edge of I19), this tool was located in a pit structure (IJK 19-21) and was associated with other remains (faunal remains, platelets, thermo-altered sandstone blocks, shell ornaments, flint) forming an elongated nappe that seemed to curve slightly northward (Montoya 2013: 15). The origin, anthropogenic or taphonomic, of the pit-like structure remains to this day undetermined (fig. 4).
Figure 3. « Le Colombier », Chézy-sur-Marne (Aisne, France). Fragmentary perforated baton of Chezy (part n°6219): general view after restoration, view of details of manufacturing and use stigmatas (photos and computer graphics N. Goutas).
« Le Colombier », Chézy-sur-Marne (Aisne, France). Bâton percé fragmentaire de Chézy (pièce n° 6219) : vue générale après restauration, vue de détails de stigmates de fabrication et d’utilisation (clichés et infographie N. Goutas).
Figure 4. « Le Colombier », Chézy-sur-Marne (Aisne, France). Location of the perforated baton within the pit-like structure (photo and Computer graphics C. Montoya).
« Le Colombier », Chézy-sur-Marne (Aisne, France). Localisation du bâton percé au sein de la structure fossoyée (clichés et infographie C. Montoya).
13The perforated baton fragment currently has a length of 339 mm but, depending on the estimated size of the antler from which it was made, it should not have exceeded 360 mm in length at the origin. With a subrectangular contour, it forms an elongated, wide and thick cylinder 21 and 19 mm in its mesial part with a maximum width and thickness in its distal part of 32 and 21mm. Its profile is slightly curved.
14The attachment of this perforated baton to one of the types defined by the nomenclature commission on the prehistoric bone industry (Peltier 1992a, b and c) is impossible given its fragmentary state. Indeed, a functional fracture, to which we will return later, deprives us of a significant part of the distal part. However, the anatomical portion selected for its manufacture makes it possible to exclude a connection to the category of perforated baton with « two divergent horizontal branches ». It could either be a « perforated baton with a short branch or without a branch » or « with one or two oblique branches ».
15The broken distal part (perforated area) can only be described partially, but it can be argued, in comparison with other parts and depending on the part still present, that the perforation should have a subcircular or subovalar contour. The profile of the perforation chimney (in axial section) is formed by two asymmetric wear planes (fig. 5). The lips of the perforation are blunt, rounded and deformed by use (see below). The mesial part is kept over a long length (nearly 300 mm). It is 21 mm wide and 19 mm thick. Its section is oval and its profile slightly curved. The proximal part is absent, eliminated by a post-depositional fracture still visible on the tip of the mesial part.
Figure 5. « Le Colombier », Chézy-sur-Marne (Aisne, France). Crushes observed on the perforation ring and chimney topography (photo and computer graphics N. Goutas).
« Le Colombier », Chézy-sur-Marne (Aisne, France). Ecrasements observés sur l’anneau de la perforation et topographie de la cheminée (cliché et infographie N. Goutas).
16The blank used corresponds to a segment of beam cut under the bez tine and above beam B (fig. 3-2). The perforation was made on the beam, at the junction area with the bez tine. In the absence of the basilar part, it is not possible to determine whether it is a massacre or a shed antler. The dimensions of the A/B beam (l: 348 mm ; circumference: 70 mm ; thickness of compact fabric : 3 mm) allow this antler to be classified as a medium size class_ (adult female or subadult male).
17The technical stigmatas created by the blank production process is no longer visible. As this is a «volume blank» (which kept the entirety of its original section), we can only rule on the fact that its production incorporates a transformation scheme by segmentation (Averbouh 2000). The manufacturing process of the perforation is equally imprecise: functional wear marks that have obscured the manufacturing stigmatas on both the chimney and the lips. However, it should be noted that there are a few small cupules (oval shaped depression) or curved nicks (1 to 2 mm) still slightly visible on the periphery of the perforation (fig. 3-3 and 3-4). These and some irregularity in the perforation appear to be consistent with the use of bifacial «picketing» (piquetage) to dig a circular depression by progressive material removal; the depression, drawing the future chimney, can then be enlarged and regularized by semi-rotary drilling as commonly observed on other objects of this type (Goutas 2004).
18Different types of macro-tracks - fracture and surface alteration - were found on the part.
19The fracture: only the one affecting the perforation is of functional origin. It is oblique to the axis of the beam and occurred on fresh antler; its amplitude is very important, since three quarters of the perforation are now missing. The violence of the fracture is observed in particular in the extent (36mm long) of the tongue-shaped tearing present on the posterior surface of the beam. According to the typology established by A. Rigaud (2001), this distal fracture is of subtype b2 (fig. 3-5 and 3-6). A careful check of the faunal remains will be necessary to ensure that the complementary fragment(s) of the perforation are not present. It is questionable whether the fracturing of the tool took place on the site; according to this hypothesis, the missing fragments would still have to be discovered or identified. Their absence on the site is also possible insofar as the perforation bears stigmatas suggesting that the tool should be recycled after fracture (see below).
20Surface alterations: these correspond to crushing and blunting of the perforation wall.
21Two main types of crashes were observed. They correspond to very distinct phases of use of the tool.
-
The former can be related to the primary function of a perforated baton and are located inside the perforation and on its periphery. The compression of the fibres indicates the passage of a «foreign body» inside the perforation. The inner and outer walls of the perforation have asymmetrical crushing forming two converging opposite planes: one, with a grazing incidence (inclination less than 10°), extends over the inside of the beam and extends over a length of 7mm; the other, forming a more or less horizontal wear plane, is limited to the chimney and the external wall of the perforation. Crushing results from progressive and repeated contact with a foreign body (compression and/or friction), with a relatively soft or semi-hard material, or in any case of a hardness lower than that of a reindeer antler. The location of the crushing indicates that the foreign body acted along a transverse or oblique axis within the perforation from the intern-outer surfaces of the boom (fig. 3-7). Considering the asymmetry of visible crushes, it is questionable whether they were formed simultaneously in opposite bifacial symmetry (symétrie bifaciale opposée) subsequently to the transverse passage of a foreign body, or whether they result from the oblique passage of a body that has generated the formation of wear in alternating symmetry (symétrie alterne). Finally, the presence of an «overstepping» fracture on the posterior side of the beam (left edge of the tool) also indicates that a «lateral» mechanical stress was also exerted on the perforated baton; i. e. by pressure from the antero/posterior sides of the beam.
-
The second types of crushing, referred to as secondary, cover the fracture plans affecting the «perforation ring» ; so they were created after it (fig. 6-1 and 6-2). On the fracture surfaces located on the posterior side of the beam, the bone fibres are compacted and relatively flat on one edge of the perforation (fig. 6- 3) and, thus, form a small convex bead on the other edge (fig. 6-4). On the anterior side of the beam (right edge of the tool), the obliqueness of the fracture plane forms a bevel on which the bone fibres are also compacted unidirectionally, towards the outside of the perforation (fig. 6-5 and 6-6). These crashes show that the pierced stick has been recycled; unlike the previous one, this second use did not induce a continuous or sudden traction on the perforation but testifies to a laid down gesture (made likely by friction).
Figure 6. « The Colombier », Chézy-sur-Marne (Aisne, France). Secondary crushing located on the fracture planes of the perforation and a zone of superficial striations on the periphery of the perforation (photos and computer N. Goutas).
« Le Colombier », Chézy-sur-Marne (Aisne, France). Écrasements secondaires localisés sur les pans de fracture de la perforation et plage de stries superficielles en périphérie de la perforation (clichés et infographie N. Goutas).
22Near an old glued fracture, there is a short zone of parallel and oblique striations, barely visible, located on the outer side of the beam, a few mms from the perforation (fig. 7, n° 7). This type of stigmata, generally observed on the walls of the perforation in association with crushing, has been interpreted as the result of an «action laid down, performed by an external material intended to contain abrasive elements, such as silica» wich would characterize «the etching bite of the material flowing (...) into the perforation»_. This material could be similar to a rope (Lompré 2003: 35).
Figure 7. « Le Colombier », Chézy-sur-Marne (Aisne, France). Block in the process of debitage on reindeer antler (artifact n° 7322): general views in situ (nos 1 and 3) and detailed views of a possible notching plane, after restoration (n° 2, 4 and 5) (photos N. Goutas).
« Le Colombier », Chézy-sur-Marne (Aisne, France). Bloc en cours de débitage sur bois de renne (pièce n° 7322) : vues générales de la pièce avant prélèvement (nos 1 et 3) et vues de détails d’un possible pan d’entaillage, après restauration (nos 2, 4 et 5) (clichés N. Goutas).
23In fact, among the best documented functional hypotheses at present, those considering pierced sticks as tools involved in lever or traction systems (Rigaud 2001; Lompré 2003) could be adapted to the first observations made on Chézy’s perforated stick. The experimental data available so far concern pieces discovered in the Magdalenian context; on these pierced sticks, the stigmatas, in particular the crasches, are organized in alternating symmetry on and around the perforation. However, other studies on older contexts (Gravettian) reveal greater variability in the arrangement and structuring of use-wear within the perforation (Goutas 2004).
24The first piece, No. 7322, is a left antler from a killed reindeer found in square D20. The anatomical portion preserved corresponds to the base + beam A + bez tine (fig. 7-1 and 7-2). According to the reference measurements taken by K. Gripp in 1935 on fossil branches of known age and sex from the Meiendorf site in Germany and according to the circumference of its beam (110 mm), the Chézy reindeer antler can come from an adult animal, either female (even aged) or male of 4 years and older. Indeed, its circumference is too large to refer to that of a young female or a subadult male antler (Table 1). Taking into account other criteria helps to refine our determination.
Table 1. Dimensions of fossil reindeer antlers of known age and sex from the Meiendorf site in Germany (after Bouchud 1959 in Goutas 2003).
Dimensions des bois de rennes fossiles d’âge et de sexe connus du site de Meiendorf en Allemagne (d’après Bouchud, 1959 in Goutas, 2003).
25The full development of the antler, the significant size of its bez tine and the thickness of the compact tissue (5 mm) measured at the top of the beam A allow to integrate it into the medium/large-size class, and possibly to attribute it to a young adult male (Averbouh 2000). The kept bez tine is about 105 mm long; the rest of the mesial part and its distal end are missing. The first tine is absent; it is, in fact, frequent that this andouiller does not develop and materializes by a simple tuberosity at its base, which would be the case here.
26Since it is a massacre reindeer antler, one can wonder if its acquisition can be linked to the reindeer hunted during the occupation of the site? Indeed, according to archaeozoological studies conducted by O. Bignon (Montoya et al. 2012, 2014, 2019), the remains of about nineteen individuals (NMIc) have been found in the LGM levels of Chézy, making reindeer the majority game. The hunting practiced in Chézy is called attritional insofar as it shows a preferential orientation towards young and old adults. The 3-6 year old class to which the massacre antler would relate «is significantly under-represented». According to the skeletal profiles, the skull boxes were also removed to the site. Antler no. 7322 could therefore have been brought at the end of a local hunting episode.
27In terms of typology and technology, artifact CSM-7322-D20 appears to correspond to an abandoned block in the process of debitage. Its comparison with the category of «extraction matrices» is also plausible, although not certified. The antler, when discovered (fig. 7-1 and 7-3), had a long, straight and narrow depression along beam A (anatomical anterior side) -stigma 1- which suggested a shallow grooving groove. In parallel, another depression stigma 2- of a somewhat different nature, of very limited length, and this time perpendicular to the first (fig. 1, n°. 3) had also been noticed. The observation of these stigmatas was difficult due to the poor surface condition of the antler. In addition, once out of the ground, a portion of the beam severely weakened by cracks, broke into 15 fragments 57 to 15 mm long (fig. 7-4).
28After observation under a stereoscopic microscope, the technical status of stigma 2 remains doubtful. It could be the remains of a gorge resulting from a shallow removal by direct percussion (fig. 1, n° 5 C-D). More precisely, it could be a short portion of one of the gorge walls. Its microtopography, stairs-like, evokes the controlled and continuous removal of material chips (removal scars). The precise length of this «depression» is no longer determinable, the beam is now broken at this point. However, according to the pictures of the antler before it was taken out of the ground, it was 20 to 30mm long. Now, only 10mm of its original length is still really observable. Further on, towards the base of the antler, small isolated and undoubtedly anthropogenic notches were also observed (fig. 7-5 E-F).
29This possible sectioning gorge is in direct association with stigma 1, a longitudinal grooving groove (fig. 8-1). Again, the multiple fractures of the beam allow us to observe one side of the groove, not the groove in full section. The preserved area is visible unevenly and discontinuously over a length of nearly 110 mm. The best preserved portion of the groove plane is about 35mm long (fig. 8-2). It has microstriations parallel to each other and slightly oblique (fig. 8-2 and 8-4) resulting from contact with a flint tool used according to a radial plane (fig. 8-5). The fineness of the groove indicates the use of a very fine burin tool; the very narrow grooving seems to have reached the spongy bone tissue in some places. The difficulty in reading the groove is due to the surface alterations of the compact bone tissue and the deformations of the beam after its restoration. Indeed, the sediment contained in the perch ensured its maintenance. Once removed, the cracks running through the antler widened and opened the beam in several places. However, one of these cracks will have been initiated within the grooving groove itself, the latter offering a zone of fragility in the material. Unlike fracture sections resulting from a natural cracking process, which clearly show a tearing of the bone fibres, the «open surfaces» that we refer to a vestigial grooving groove section crossed by a crack are regular and blunt, and in places, the bottom of the groove is still visible (fig. 9-1 and 9-3).
Figure 8. « Le Colombier », Chézy-sur-Marne (Aisne, France). Block in the process of debitage on reindeer antler (artifact n° 7322): detailed views after restoration of the grooving groove (photos N. Goutas).
« Le Colombier », Chézy-sur-Marne (Aisne, France). Bloc en cours de débitage de Chézy sur bois de renne (pièce n° 7322) : vues de détails après restauration du pan de sillon de rainurage (clichés N. Goutas).
Figure 9. « Le Colombier », Chézy-sur-Marne (Aisne, France). Block in the process of debitage on reindeer antler (artifact n° 7322): detailed views after restoration of the groove and location of possible extractions carried out on the beam (photos and computer graphics N. Goutas).
« Le Colombier », Chézy-sur-Marne (Aisne, France). Bloc en cours de débitage de Chézy sur bois de renne (pièce n° 7322) : vues de détails après restauration du pan de sillon de rainurage et localisation des hypothétiques extractions sur la perche (clichés et infographie N. Goutas).
30The interpretation of this groove is delicate, given its very residual nature; it could correspond to a tracing (fig. 9-4), the first step in the formalization of a grooving groove (« sillon de rainurage ») (Dauvois 1974). If the action of grooving stricto sensu does not require the mastery of complex skills, it does require the skill and safety of the hand (op. cit.). The initiation of grooving is therefore an important step because it is the one that must guide the flint tool during the first passages in order to prevent it from slipping and leaving its path, which would lead to the formation of a discontinuous groove. The initiation of grooving can easily be done with a small angle burin, or with the dihedral or trihedral with a small flake (Averbouh 2000). Once the steering groove has been created, the burin no longer slips, and the actual grooving can begin. The groove can then be widened in such a way as to obtain, technically, a real groove (with two sides and a flat bottom). However, this was not the case on this waste since the grooving was stopped very early on. The reasons for this discontinuation remain undetermined. The chronological arrangement of the longitudinal grooving and the possible notching can no longer be reconstructed with certainty. It can only be assumed that the latter was later than the grooving. It is possible that the realization of this shallow groove is linked to the detachment of rods located higher on the branch, on the internal side (in anatomical position) of the beam B/C (fig. 9-5). This hypothesis would allow for a better understanding of a technical detail. First, the portion of the grooving panel that is best preserved is directly associated with the possible transverse cutting groove (see fig. 8-1). Then, the obliquity of the microstriations on this grooving section indicates a relaxation of the pressure exerted on the flint tool (fig. 8-2 and 8-3), which generally occurs at the end of the procedure. In any case, the purpose of these various technical actions (notching? and grooving) remains very speculative for the moment.
31This piece (fig. 10-1 to 10-3) consists of the basilar + pole A portion of a left reindeer slaughter antler (presence of part of the skull and pedicle) probably belonging to an adult male given the first measurements taken. Indeed, the total length of the piece is about 300mm: pole A (first part of the pole) is present on its first third and its width taken above the first tine and then at the base of the bez tine are respectively 50mm and almost 40mm. In addition, the gap between these two tines is wide, which is more characteristic of a male than a female reindeer antler (see Averbouh 2000 reference collection, Averbouh 2015).
Figure 10. « Le Colombier », Chézy-sur-Marne (Aisne, France). The possible reindeer antler extraction matrice (photos D. Gilot).
« Le Colombier », Chézy-sur-Marne (Aisne, France). La possible matrice d’extraction en bois de renne (clichés D. Gilot)
32The poor state of conservation makes it difficult to read. It seems, however, that a line of necrosis is emerging between the pedicle and the medallion, which would indicate the next shedding of the antler. If this first observation is true, it would mean that it was obtained in late fall. But it will be difficult to determine whether the work carried out on this antler closely followed its acquisition: it is no longer possible to read the traces of deterioration that animals or plants could have left on the raw antler and would indicate that it remained more or less long in the open air before its debit.
33Several observations can be made at a first observation: the presence of a wide empty band on the inner surface of the beam (fig. 9, n° 1-A), the absence of a first tine eliminated at its base and the presence of the first third of the bez tine (fig. 9, n° 1-B and C). This set of features is commonly observed on a single rod extraction « matrice ».
34The empty strip can, in fact, refer to the extraction of a support of the strip type, generally wide and roughly rectangular. This rod, if extracted, could have a minimum length (due to breakage) of 140 mm, a maximum width of between 35 and 40 mm and a maximum thickness of 10mm. It would have largely consisted of compact bone tissue, which would once again argue in favour of an adult male antler. This does not solve the question that remains: has a rod been extracted or not? In fact, to delimit this future support and prepare its extraction, it is necessary to draw its outline and initiate the future fracture lines sufficiently so that the detachment of the rod follows them exactly.
35In the Magdalenian, this preparation for extraction is done, at least for the longitudinal edges of the future support, with a very commonly used wear technique -grooving- which simultaneously creates characteristic planes on the support and on the «matrice». However, only the straight edge (looking at the piece in anatomical position, fig. 10-4) evokes a groove plane by its inclination (steep/with worked surface) and by its rectangular morphology (hardly resulting from an action other than anthropogenic). However, the fine reading of the characteristic striations was not possible during this first examination and must be carried out under a scanning microscope before making any well-founded hypothesis, especially since the opposite edge (left) is much less well preserved and has a general appearance more reminiscent of a broken edge than a groove plane.
36Similarly, the fact that the first tine is removed at its base (fig. 10-1B) suggests that this is due to an anthropogenic action, possibly a transverse sectioning action by removal by direct percussion quite common during this period. But no element, even vestigial, has so far been identified to corroborate this hypothesis. Again, a second, more detailed examination is required.
37Finally, the presence of the first third of the bez tine is also a common feature observed on single rod extraction matrices. But the reading of the fracture zone is, once again, not easy and deserves to be examined at a higher magnification, in particular to verify that the «pacifier» shape (fig. 10-5) is not linked to animal action (after the waste had been abandoned).
38In the end, this piece presents several criteria characteristic of a single rod extraction matrice but it remains to be demonstrated whether this hypothesis based essentially on morphological analogies is valid.
39A last piece of reindeer antler was discovered during the 2014 season (n° 7117, fig. 11). Considering its dimensions (139 mm long, 30 mm wide and 18mm thick), associated with its very flattened oval cross-section, its overall elongated shape and the ratio between compact and spongy bone tissue, we identify this fragment as a mesial part of a bez tine of a medium-large sized antler. At about fifty centimetres, artifact n° 7322 has been abandoned. As these two pieces belong to the same class-size, we wonder about the possibility that these two anatomical elements belong to the same antler. But these two parts cannot be physically (concretely) reassembled together and this hypothesis, such as the fact that this fragment of ice andouiller belongs to the group of worked parts, cannot be confirmed.
Figure 11. « Le Colombier », Chézy-sur-Marne (Aisne, France). Fragment of a bez tine, internal and external surface (photos N. Goutas).
« Le Colombier », Chézy-sur-Marne (Aisne, France). Fragment d’andouiller de glace, vues face interne et externe (clichés N. Goutas).
40This artifact (fig. 12, fig. 13) consists of the joint portion (neck and glenoid cavity) and the beginning of the pallet of a horse scapula. The alteration of its surface, very marked by vermiculations left by the roots of plants, makes it particularly difficult to read. Two questions were asked to determine whether this part could have been used as a hammer, a hypothesis put forward by C. Montoya: one concerned the natural or anthropogenic origin of the scapula fracture from the end of the cervix in order to improve the grip of the supposed tool; the other concerned the presence of stigmatas associated with a percussion around the glenoidal cavity and on the supraglenoidal tuber.
Figure 12. « Le Colombier », Chézy-sur-Marne (Aisne, France). The artifact on horse scapula (lateral and medial view and detailed view of the perforated area, photos C. Montoya).
« Le Colombier », Chézy-sur-Marne (Aisne, France). La pièce sur scapula de cheval (vue latérale et médiale, Clichés C. Montoya).
Figure 13. « Le Colombier », Chézy-sur-Marne (Aisne, France). The horse scapula object and the possible reindeer antler « matrice » at the time of discovery, overall view (photo C. Montoya) and close up view (photo V. Miclon)
« Le Colombier », Chézy-sur-Marne (Aisne, France). La pièce sur scapula de cheval et la possible matrice d’extraction en bois de renne lors de leur découverte, vue générale (cliché C. Montoya), vue détaillée (cliché V. Miclon).
41In the current state of observations, we cannot reliably answer the first question, but we have nevertheless observed localized notches near the glenoidal cavity, on the supraglenoidal tuber, and more scattered on the surfaces of the scapula. These nicks (or notches) are short (a few millimetres at most), shallow, marked at the bottom by a double groove and they mostly have an oblique profile with a flat bottom. All these elements appear to be consistent with the action of a lithic cutting edge, and therefore with an anthropogenic action. The preliminary examination has shown that there may also be two distinct subtypes of nicks that have yet to be precisely characterized. For the time being, it is only noted that the nicks present do not correspond to the main stigmatas we see on the archaeological organic hammers used for flint knapping (cupules, V-notches, compaction, etc. cf. Averbouh and Bodu 2002). Their structuring (organized, unorganized) should be characterized by their fine mapping on the piece in order to be able to determine its origin: were they created by butchery activities or by processing activities? In the first case, the piece will join all the faunal remains; in the second, it will be part of the bone industry by taking the status of a tool whose use will first have to be determined if it has been active or passive.
42The Chézy-sur-Marne site undoubtedly contains elements from an osseous material industry. If among the 5 exhumed pieces, the one on bone cannot be integrated in a certain way into the series of industry out of hard animal material, this small set still presents a great coherence, if only from the raw material of the pieces and from the presence of elements, certainly sometimes tenuous but potentially attributable to a rod production by extraction by grooving.
43While the use of notching is attested in Badegoulian contexts (Averbouh 2004), the use of the longitudinal grooving technique is more relevant. Indeed, the data currently available suggest that this technique was not used by Badegoulian groups to work with reindeer antlers (Averbouh, comm. SPF 2006, Averbouh and Pétillon 2011), except perhaps in the Rond du Barry whose archaeostratigraphy is unfortunately not safe (Rémy 2013). For the time being, based on data from reliable archaeological contexts, it seems that Badegoulians have exclusively produced their supports on reindeer antlers by blank production by fracturing, marking a major difference with Magdalenian groups (Averbouh comm. SPF 2006; Averbouh and Pétillon 2011; Pétillon and Averbouh 2011; Pétillon and Ducasse 2012; Averbouh 2018).
44Because of their poor conservation, the matrices from « Le Colombier » at Chézy-sur-Marne are probably not the best candidates to establish that the grooving technique and more, a debitage by extraction, were used on reindeer antler in a Badegoulian context. But these are witnesses we have to take into account. Nevertheless, we insist on the importance of dissociating the punctual identification of the grooving technique, and the integration of this same technique into the «Badegoulian technical tradition»; by this we mean all the skills and know-how (sensu Pelegrin 2000) repeatedly mobilized and codified by the Chézy group in order to produce its equipment. Thus, the data currently available from the Chézy’s bone material -and in particular the difficulty of reintegrating this technical observation into a global operating scheme for the block- do not allow us to put forward the hypothesis that Chézy’s human groups carried out sawing by extracting rods, using the double longitudinal grooving process, as commonly practiced by Magdalenian groups.
45If this reading is confirmed and if new evidence proves it, it can then be argued that the transformation of osseous materials, and more particularly that of reindeer antler, the major raw material of this period, did not follow the same principles as those of the potentially contemporary Badegoulian groups in southwestern France.