Navigation – Plan du site

AccueilNuméros10PARTIE I - Le référentiel expérim...The reference collection of cutmarks

PARTIE I - Le référentiel expérimental

The reference collection of cutmarks

Sandrine Costamagno, Marie-Cécile Soulier, Aurore Val et Sébastian Chong
Cet article est une traduction de :
Le référentiel de stries de boucherie [fr]

Plan

Haut de page

Texte intégral

1The butchery experiments conducted as part of the PCR belong to a long tradition of studies initiated in the early 20th century by L. Henri-Martin (1910). With the exception of this early example however, the first butchery experiments really began in the 1970s, with the goal of testing the effectiveness of stone tools but also to define the constraints related to the removal of certain animal animals (see Frison, 1978 on the processing of a bison carcass: 318-328; Poplin, 1972 on a hyena carcass; followed by Jones, 1980; Johnson 1985; Mitchell, 1995; Bracco, Morel, 1998; Machin et al., 2007; Braun et al., 2008). The pioneering work of P. L. Walker and J. C. Long (1977) established a new line of research focused on the morphological characteristics of cutmarks on bone based on the raw material of the tools employed (see below). The study of cutmarks took on a new significance with the work of L. R. Binford (1981). The reference collection that he proposed established a clear causal relationship between the placement and orientation of cutmarks and specific butchery activities. The study of butchery cutmarks on well preserved faunal assemblages has since been used to reconstruct all or part of the butchery chaînes opératoires (for example, Delpech, Villa, 1993; Castel, 1999; Costamagno, 1999; Chaix, 2004; Fiore et al., 2004; Niven, 2006; Leduc, 2010; Müller, 2013; Soulier, 2013; Street, Turner, 2013; Feyfant et al., 2015; Cueto et al., 2016). Beyond revealing the resources that were exploited at a given site, the detailed study of butchery operations and the traces related to them provides the opportunity to examine potential markers of cultural differences in the exploitation of animals (see Part I, chapter 3.A). The consideration of the social implications of different butchery chaînes opératoires has been primarily advanced by the French school of thought (see Birouste, 2018 for a review), while the quantitative approach has been privileged in the Anglophone literature (see below).

2In the PCR, the butchery experiments conducted on animal carcasses relate to three fields of research: connecting cutmarks to motions, connecting cutmarks to tools, and, finally testing the effects of different variables on the formation of cutmarks related to butchery. This reference collection of cutmarks was applied in several academic theses (Chong, 2011; Costamagno, 2012; Soulier, 2013) and is the subject of several articles (Soulier, Costamagno, 2017; Val et al., 2017). This chapter 3 presents a synthesis of those results.

1 - A history of the study of butchery cutmarks

S. Costamagno, A. Val

A - A critical review of actualistic approaches to butchery cutmarks

a - Connecting cutmarks to butchery activities

3In spite of its heuristic value, the experimental reference collection proposed by L. R. Binford (1981) based on his observations of the Nunamiut has several limitations. The observations made on reindeer and sheep bones processed at three Nunamiut residential campsites were conducted a posteriori and, because the carcasses had been subjected to intensive butchery, it is difficult to tie any one trace to a particular butchery activity with certainty. This is especially relevant in the case of the articular extremities of bones, which may bear cutmarks related to disarticulation but also to defleshing An examination of the templates documenting cutmarks related to disarticulation (Binford, 1981: figures 4.25, 4.26, 4.30 and 4.32) and defleshing of long bones (Binford 1981: figures 4.37, 4.38 and 4.39) shows a clear dichotomy between the first ones – mostly transverse and present at or near the epiphyses – and the second ones – mostly oblique and located farther from the epiphyses. There are legitimate reasons to question the grounds on which Binford made these attributions. The low variability of cutmarks described raises some questions, because Binford does not provide any details on the number of pieces analysed in relation to the overall skeletal elements available or the method of observation. Furthermore, Binford’s reference collection does not include the placement and form of cutmarks on shaft fragments, as these elements were long considered to be of little informative value for archaeozoological studies in the Anglo-Saxon school of thought (see for example Klein, Cruz‑Uribe, 1984). It was only with the work of L. E. Bartram (Bartram, 1993; Bartram, Marean, 1999) and then of C. W. Marean et al. (Marean, Frey, 1997; Marean, Kim, 1998) that the informative potential of diaphyseal fragments was recognized, in particular in assemblages that had been subjected to carnivore activity, which demonstrated the need for new interpretive frameworks. Another limitation of Binford’s data relates to the tools used by the Nunamiut for the butchery of carcasses. These hunters used metal knives which, according to several experiments (Frison et al., 1976; Milo, 1994 cited in Nilssen, 2000; Potter, 2005), have proven to be much more effective than lithic tools and, besides, allow for cutting gestures that are not possible with stone tools (Nilssen, 2000). This raises the following question: are the cutmarks recorded by Binford representative of the entire range of possibilities? Finally, several external factors (climate) and internal factors (number of animals killed, culinary practices) influenced the butchery practices of the Nunamiut and, by extension, the types of cutmarks produced during butchery. The question of the relevance of this framework to the interpretation of fossil bone assemblages has been particularly raised in African contexts (Yellen, 1977b; Bunn et al., 1988; O’Connell et al., 1988; Gifford-Gonzalez, 1989a; Bartram, 1993; Lupo, 1994).

4In spite of these limitations, few studies have attempted to enrich the available reference collections. The work of Y. Abe (2005) on a group of Evenki living in the taiga demonstrated how much influence cultural practices, whether symbolic or culinary, can have on the formation of cutmarks. In this specific case, the degree of care shown by the group during the butchery is a sign of the respect for the animals hunted. For reindeer, for example, this takes the form of the systematic disarticulation of the carpals / tarsals from the adjacent long bones (see above). Additionally, the favoured method of cooking is boiling, which results in extremely rare cutmarks on the meatier bones. Similar observations were made on another Evenki group living in the Amour region (Costamagno, David, 2009). Nonetheless, these uncontrolled studies of butchery practices have not provided sufficient data to complete Binford’s reference collection.

5In addition to these ethnoarchaeological observations, P. J. Nilssen (2000) favoured an experimental approach with the aim of controlling the butchery operations on the carcasses analysed. One of the objectives was to compile a list of the cutmarks that could result from several different butchery operations (for example disarticulation and defleshing The resulting framework is, to date, the most complete because it is based on the study of seventeen small and large bovid carcasses. The butchery was performed by a professional butcher and by hunters to produce dried meat, with metal and sometimes flint tools. All cutmarks were reported following Binford’s codes or given new ones, especially in the case of shaft portions. This reference collection demonstrated that the interpretation of butchery marks is not as straightforward as Binford’s work suggests; cutmarks related to defleshing can be superimposed on cutmarks related to disarticulation. Nonetheless, the interpretive framework provided by P. J. Nilssen, based on carcasses that were butchered primarily with modern knives, suffers some of the same limitations as that of Binford. Moreover, the extremities of the legs were not included in the study and cutmarks related to skinning are little documented, while those related to the removal of tendons are absent. J.-D. Vigne (2005) is the only author to mention cutmarks related to this sequence during experiments conducted on red deer and roe deer (Capreolus capreolus). According to him, the removal of the tendons of the toe flexors and of the foreleg left no traces. Even so, longitudinal cutmarks on the anterior surface of the radius were recorded and are related to the severing of the distal tendon from the carpal extensor muscle before its removal.

6With regards to the number of experiments performed with flint tools (see below), very few have provided detailed information about the placement of butchery marks that would connect them to specific activities. Amongst them, several are based on the butchery of small prey (Laroulandie, 2001; Cochard, 2004; Willis et al., 2008; Mallye, 2011). M. Padilla (2008) conducted experiments with stone tools on calf (Bos taurus) carcasses in order to document the variability of gestures according to the butcher’s dexterity. Even though the cutmarks are reported on templates, no distinction was made between different activities. The experiments of J.-F. Bez (1995) were based on the butchery of a goat carcass and a horse head. To the best of our knowledge, the experiments conducted by A. B. Galán and M. Domínguez-Rodrigo (2013) are the only ones conducted with flint tools (unmodified flakes, retouched flakes, bifaces) on a wild animal (Cervus elaphus). The butchery was performed by an experienced hunter who, according to the selected protocol, used exclusively oblique gestures for disarticulation and transverse gestures for skinning and defleshing. The cutmarks recorded on the templates are attributed to different activities on the basis of orientation. In butchery that is not constrained by such protocol, the butcher can employ highly variable gestures, which limits the relevance of this framework to archaeological material (Soulier, Costamagno, 2017).

b - Connecting the cutmark to the tool

7A number of publications are concerned with the relationship between the raw material of stone tools and the morphology of cutmarks. The research of H. Greenfield (Greenfield 1999; Greenfield 2006) and of R. Jones (2011) confirms certain results of previous studies (Walker, Long, 1977; Walker, 1978; Olsen, 1988) by demonstrating, through SEM analysis of experimentally-produced reference collections, that it is possible to distinguish the cutmarks left by stone tools of unidentified raw materials from those left by metal ones. Employing the same method, other studies have demonstrated the possibility of identifying traces left by bamboo and those left by chert (West, Louys, 2007). However, it currently seems impossible to identify cutmarks made by different lithic raw materials, such as obsidian, flint, chert, quartz and quartzite (Olsen, 1988; Greenfield 2006; de Juana et al., 2010). In addition to studies of raw materials, the potential of distinguishing between different tool types has been examined. S. de Juana et al. (2010) attempted to differentiate the cutmarks produced by a biface from those produced by a retouched flake and an unmodified flake (the raw materials used were flint and quartzite). Their method, inspired by the work of M. Domínguez‑Rodrigo et al. (2009) for distinguishing butchery cutmarks from trampling damage, combines the observation of 16 variables using light microscopy at low levels of magnification (between 20 × and 40 ×) with multivariate analyses. In an experimental reference collection, the tool employed was correctly identified in 80 % of cases, supporting the application of the method to the archaeological record. New techniques of microscopic analysis such as laser profilometer (Stemp, Stemp, 2003; Stemp et al., 2009; Stemp, 2013) and confocal microscopy have recently been applied to the study of butchery cutmarks. These two techniques provide topographic information that can be used to conduct fine-grained and quantitative analyses of different kinds of cutmarks. However, with a few exceptions (Bello et al., 2009; Yravedra et al., 2010), such approaches are infrequently applied to the archaeological record and are primarily dedicated to the analysis of experimental products. Therefore, questions remain regarding the reproducibility by other scientists as well as their applicability to archaeological fauna. For example, concerning the question of retouched / unretouched flakes, we are not aware of any application to archaeological assemblages. Furthermore, if the expansion of criteria, combined with multivariate analyses, has proven successful on experimental examples (Dominguez‑Rodrigo et al., 2009; de Juana et al., 2010), how explain that only a relatively limited range of cutmarks is considered when the method is applied to the study of archaeological fauna? Finally, it is important to bear in mind that although the primary goal of the archaeologist is to recover the maximum amount of information possible from remains of the past, the value of the information recovered must be considered in terms of the time investment required. In other words, extensive microscopic analysis of butchery cutmarks must above all respond to the actual questions and problematics raised by an archaeological assemblage, a site and its context, and the chronological period.

c - Quantifying cutmarks

8Cutmarks quantification has been the subject of numerous studies and reflections. Generally, the frequency of cutmarks is considered to be indicative of the intensity of skeleton exploitation by human groups. This approach is founded on the implicit assumption that the proportion of fragments recovered from an excavation bearing cutmarks is representative of the proportion initially present in the bone assemblage (Lyman, 1992, 1994b). All studies concerning the quantification of cutmarks rest on this fundamentally unverifiable assertion. The fact that the cutmarks could be epiphenomenal is another limitation of this type of approach (Lyman, 1992, 1994b). Indeed, cutmarks form accidentally in the course of butchery activities and it is legitimate to question the relevance of quantitative analyses while comparing bone assemblages and inferring potentially different techniques of butchery. The question of factors that might cause variations in the frequency of cutmarks remains, in this sense, far from resolved (see for example Lyman, 2005; Domínguez‑Rodrigo, Yravedra, 2009).

9Many factors can be invoked to explain variability in the frequency of cutmarks on archaeological bone, including:

  • intrinsic factors: size of the carcass (Lyman, 1992), quantity of meat on the bones (Shipman, Rose, 1983a; Binford, 1984), rigidity of the joints (Lyman, 1987), state of the carcass (Binford, 1981; Lupo, 1994);

  • external factors: skill of the butcher (Guilday et al., 1962), precision of defleshing (Binford, 1988 ; Delpech, Villa, 1993), type of tool used (Potts, 1982; Domínguez-Rodrigo, 2002), culinary practices (Gifford-Gonzalez, 1989a; Costamagno, David, 2009);

  • taphonomic factors: preservation of bone surfaces (Thompson, 2005; Domínguez-Rodrigo, Yravedra, 2009), degree of fragmentation (Bartram, 1993; Otárola-Castillo, 2010);

  • methodological factors: the degree of investment in observing cutmarks on the part of the archaeozoologist, levels of magnification used.

10Investigating the effects of these various factors on the frequency of cutmarks, on the basis of ethnoarchaeological observations, is not without problems. The work of K. D. Lupo and J. F. O’Connell (2002) amongst the Hadza seems to indicate that the size of the animal and the quantity of meat on the bones does not influence the frequency of cutmarks. The ribs and the pelvic bones are nonetheless characterized by higher frequencies of cutmarks than the other skeletal elements. The studies conducted by D. Gifford-Gonzalez (1989a) amongst the Daassanach provide contrasting evidence as the long bones of goats bear more cutmarks than the bones of zebras or cattle. According to this author, “it related less to the greater intensity of cutting on caprine limb segments than to the role soft tissues play in fortuitously protecting the bone of larger animals from the impacts of cutting implements” (Gifford-Gonzalez, 1989a: 202). This assertion, however, cannot be verified ethnoarchaeologically due to the variety of factors that could come into play in the formation of cutmarks. Y. Abe (2005: 185) notes, following studies amongst a group of Evenki, that “There were variations in butchery mark frequency and placement that were identified to various causes observed during the butchery process, but these variations were strictly case-by‑case and not identifiable quantitative patterns emerged overall”. Several experimental approaches developed in recent years have attempted to evaluate the effects of different variables on the frequency of cutmarks on bone surfaces (Egeland, 2003; Domínguez-Rodrigo, Barba, 2005; Pobiner et al., 2005; Potter, 2005; Dewbury, Russell, 2007; Braun et al., 2008; Padilla, 2008; Otárola-Castillo, 2010).

The size of the animal

11Does the size of the animal influence the frequency of cutmarks? In the available literature, the results are mixed. For some, there is a negative relationship between the size of the carcass and the frequency of butchery marks (Crader, 1983; Gifford-Gonzalez, 1989a; Haynes, 1991), while others report a positive relationship (Lyman, 1992). The experimental approaches developed in recent years do not always provide coherent results. In the experiments conducted by B. L. Pobiner et al. (2005), the hindquarters of six goats, six calves and six zebras were defleshed by two professional butchers using basalt flakes. The numbers of resulting cutmarks increased with carcass size. M. Domínguez‑Rodrigo and R. Barba (2005) arrived at the same conclusions by comparing the frequency of cutmarks (%NISPcut) on small and medium size ungulate carcasses that were defleshed, then disarticulated, and finally fractured for marrow. C. P. Egeland (2003) reports contrasting results but the data have yet to be published (Egeland, Byerly, in prep. cited in Egeland, 2003). The experimental protocol employed by M. Domínguez‑Rodrigo and R. Barba is problematic because, having chosen to fracture the bones, they introduced two variables that could account for the recorded variations: the size of the carcass but also the intensity of fragmentation (Otárola‑Castillo, 2010). Having also decided to perform complete butchery sequences, they may have produced frequencies of cutmarks that are not necessarily comparable to those produced by “regular” butchery.

The quantity of meat

12The question of the quantity of meat present on the bones and its influence on the frequency of cutmarks has been the subject of much debate. For H. T. Bunn and E. M. Kroll (1986: 450), the high frequencies of cutmarks recorded on meaty bones “are most likely to occur when it is difficult or impossible to see where the hone is, as when a complete, meaty limb bone is being defleshed”. At the site of FLK Zinjanthropus, the abundance of cutmarks on meaty bones led the authors to propose the hypothesis of access to carcasses still rich in meat for the early hominids. In contrast, for L. R. Binford (1986: 446), the high intensity of butchery marks on these elements indicated “extreme difficulty in processing already partially desiccated limb parts that had been previously ravaged by carnivores, leaving little usable meat available to the hominids” and thus the butchery of carcasses poor in meat. The experiments conducted by B. L. Pobiner et al. (2005) show that the number of cutmarks is neither related to the amount of meat available on a skeletal element nor to the quantity of meat removed, casting doubt on both of the preceding hypotheses. C. P. Egeland (2003) conducted an experiment to see whether the intensity of butchery activities conducted on a carcass had a positive correlation with the frequency of cutmarks observed on the bones. The number of blows with the tool required for defleshing was considered by Egeland (2003) as representative of the intensity of butchery, following Y. Abe et al. (2002: 657): “A key assumption that all zooarchaeologists make in this type of analysis is that more intensive cutting (more cutting actions) results in higher frequencies of cutmarks on the bone surface”. The experimental sample comprises 16 carcasses (nine horses and seven cows) that were defleshed with unmodified flakes of obsidian and flint. The results show no relationship between the number of cutmarks and the number of blows. According to Egeland (2003), this variability is not due to the quantity of meat initially present on the bones because there was no correlation between the number of cutmarks observed by skeletal element and the mass of meat removed. However, with the exception of a “hyper-standardized” approach to butchery, the latter is not necessarily proportional to the quantity of meat available on the bones. Thus, in our opinion (Costamagno, 2012), the quantity of meat still adhering to the bones could explain in part the variation observed in the formation of cutmarks relative to the number of blows. In Egeland (2003), the data concerning the defleshing of the frozen hind leg of a horse is an excellent illustration of this problem. Indeed, as predicted (Binford, 1981, 1984), this limb is the one that required the greatest effort, even though the number of cutmarks is particularly low: “even though the topmost layers of flesh on the frozen limb allowed a stone tool to cut through them, the bottom layers remained in a condition such that the tool edges had almost no chance of coming into contact with the bone surface” (Egeland, 2003: 47-48). This raises the question of the relevance of this variable for determining the intensity of butchery (Costamagno, 2012). Let us consider two animals of different sizes: a roe deer and a bison. While defleshing the roe‑deer carcass required fewer blows than for the bison carcass, this does not mean that the roe-deer carcass was less intensively butchered than the bison carcass. “Thus, the number of blows is more a reference to the energy expended by the butcher – which can vary with the size of the animal, the state of the carcass, the techniques employed – than to the intensity of butchery, which […] is a measure of defleshing extensiveness” (Costamagno, 2012: 15; translated from the original French). This definition approaches the one proposed by Binford (1988: 127): “the number of cutmarks, exclusive of dismemberment marks, is a function of differential investment in meat or tissue removal”.

The tool

13The influence of the type of tool on the formation of butchery marks has received little attention. A. G. Dewbury and N. Russell (2007) hypothesized that obsidian tools would leave fewer cutmarks than flint tools but although the half-carcasses of red deer defleshed with obsidian tools did bear fewer cutmarks than those defleshed with flint tools, the observed differences were not significant.

14In contrast, in most of the studies related to butchery, it is implicitly admitted by the archaeozoologists that Palaeolithic butchers avoided touching the bones with their tools as much as possible, in order to avoid damaging the tools. For instance, H. T. Bunn (2001: 207) writes: “butchers with any interest in preserving the sharpness of their knife blades are not going to repeatedly hack into the visible bone surfaces when adhering meat can be shaved free without hitting the bone directly enough to produce cutmarks. Cutmarks are mistakes; they are accidental miscalculations of the precise location of the bone surface when muscle masses obscure it. As soon as the butcher can see the bone surface, few if any cutmarks will be inflicted thereafter in that area”. On the basis of a study of the defleshing of 18 hind legs (see above), D. R. Braun et al. (2008) demonstrated that the number of cutmarks observed on the leg‑bones was not reflected in the extent of use-wear on the edges of the tools. According to these authors, it is therefore improbable that Palaeolithic butchers carefully avoided contact between the bone surfaces and the tool edges. As these butchery sequences were conducted with basalt tools, one might ask whether these results apply to tools in other materials, such as obsidian, which is more fragile (Dewbury, Russell, 2007), and flint. Nonetheless, even if the experiment conducted by M. Padilla (2008) shows that more experienced butchers leave fewer cutmarks than novices, and although the results obtained by Braun et al. (2008) may not apply to all types of lithic raw materials, the hypothesis that Palaeolithic butchers carefully avoided touching bones with the edges of their tools remains suspect (Costamagno, 2012). Indeed, potential damage to the cutting edge is of greater concern to butchers using metal knives than it would have been for Palaeolithic people. Unmodified flakes make excellent tools for butchering a carcass. At residential sites, these tools are abundant amongst the debitage from knapping. In contexts where the availability of raw materials was not a constraining factor, these flakes could have been easily put to use and rapidly sharpened. The abundant cutmarks on bones recovered from Palaeolithic sites also support this hypothesis.

B - Methods of quantifying and recording butchery cutmarks

15In addition to the terminology (Lyman, 1994a), the diversity of methods employed has hindered the comparative study of cutmarks (Abe et al., 2002; Domínguez-Rodrigo, 2002). Two distinct approaches are preferred. The first, based on the number of pieces bearing cutmarks, is the most widely used at present (for example, Guilday et al., 1962; Frison, 1970; Bunn, Kroll, 1986; Marean et al., 1992; Lyman, 2005; Costamagno et al., 2006). This method consists of counting the number of fragments that bear butchery marks and comparing them to the total number of fragments, accounting for the legibility of the surfaces (%NISPcut) (Delpech, Villa, 1993; Lyman, 1994b: 306; Thompson, 2005). This frequency can also be calculated using the MNE (%MNEcut) (Binford, 1984; Grayson, Delpech, 1994; Milo, 1998). The second approach is based on the number of cutmarks present on a given element or portion of the skeleton. These data can be given raw (Stiner, 1999) or in the form of frequencies, using either the number of remains or the minimum number of elements (Bartram, 1993; Milo, 1998). Given the difficulty of counting the number of actual cutmarks in cases of overlapping cutmarks, Lyman (1987; 1992; 1994b) recommends counting non‑adjacent cutmarks or groups of cutmarks. This procedure has nonetheless rarely been used and counts of all cutmarks are widely preferred. L. E. Bartram (1993) and more recently Y. Abe et al. (2002) have demonstrated that the frequency of cutmarks based on the total numbers of remains varies strongly according to the degree of fragmentation of the osseous assemblages. Fragmentation actually reduces the overall frequency. E. Otárola-Castillo (2010), using computer‑simulated butchery, was able to show that the %NISPcut yielded inconsistent results for extensively fragmented assemblages, especially when the anatomical portions were taken into account. Like Bartram (1993), he recommends the use of quantitative units based on MNE for this type of analysis. Y. Abe et al. (2002), for their part, propose a unit of measure (CNC) based on the number of cutmarks observed on the surface of a given element or portion, in order to mitigate problems of differential preservation that could affect the frequencies of cutmarks based on the MNE. However, in addition to its inherent complexity, this method is founded on the assumption that the missing zones possess the same number of cutmarks as the preserved ones (Lyman, 2005). With the exception of the experiments conducted by M. Domínguez-Rodrigo (1997; Domínguez‑Rodrigo, Barba, 2005) and Egeland (2003), all experimental approaches aiming at understanding the factors that influence the formation of cutmarks use units of measurement based on MNE. The number of cutmarks is also favoured, with some counting all of the marks (Egeland, 2003; Padilla, 2008) and others using a slight modification of the method proposed by Lyman (Pobiner et al., 2005; Dewbury, Russell, 2007; Braun et al., 2008).

16Overall, the methods for recording butchery cutmarks have not been much discussed. Cutmarks are rarely reported individually and, in publications, they are usually reported by “types” related to specific activities and indicated on diagrams representing the complete skeleton (for example, Romandini et al., 2014a; Gabucio et al., 2014b; Soulier et al., 2014; Kuntz et al., 2016). While they are being studied individually, these cutmarks are recorded on anatomical templates representing the various surfaces of the bones, either manually on sheets of paper (for example, Delpech, Villa, 1993; Valensi, 1991; Castel, 1999; Costamagno, 1999), or with the aid of digital templates made in Adobe Illustrator® (for example, Val, Mallye, 2011; Soulier, 2013; Chevallier, 2015). The quantitative study of cutmarks is often limited by numbers that are too high to be processed manually. Nonetheless, thanks to GIS, quantitative approaches to the question of cutmarks orientation are being developed (Stiner et al., 2009, 2011; Soulier, Morin, 2016). Cutmark variability in the levels attributed to the Lower Palaeolithic at Qesem Cave has for instance been interpreted in terms of number of butchers that defleshed the carcasses, raising questions, according to Stiner et al. (2009), about models of food-sharing amongst “archaic” human groups. More recently, M.-C. Soulier and E. Morin (2016) used the reports of cutmarks from thirty Late Pleistocene sites as well as those from the actualistic studies of P. J. Nilssen (2000) and Y. Abe (2005). Analysis of the orientation and length of the cutmarks confirmed the existence not only of variable gestures employed in defleshing in the Late Pleistocene but also of the practice of meat‑drying during the Last Glacial Maximum. Experimental archaeology has also benefited from the application of GIS (Nilssen, 2000; Abe, 2005; Egeland et al., 2014), but only C. P. Egeland et al. (2014) conducted a quantitative study aimed at determining whether the number of butchers on a carcass affects the orientation of the cutmarks.

2 - Butchery experiments

S. Costamagno, M.-C. Soulier, S. Chong [†], A. Val

A - Experimental protocol

17The protocol established for the butchery of carcasses was the result of collaborative discussions between archaeozoologists, technologists and use-wear specialists. As one of the objectives was to develop a reference collection that would allow us to connect the cutmark to the related butchery gesture, we limited the number of operations performed on the limbs as much as possible. During dismemberment of whole carcasses, limbs were detached from the trunks at the scapular and pelvic girdles while the elbow, knee, and ankle were not disarticulated (red deer 1-6 and 8). In the case of disarticulated carcasses (red deer 7 and 9), the limbs were defleshed superficially in order to avoid contact between the tools and the bones. For the right side of red deer 7 and the left side of red deer 9, the leg bones were disarticulated prior to defleshing; these procedures were reversed for the left side of red deer 7 and the right side of red deer 9. The limb extremities, less fleshy, were treated with a less strict protocol. On some carcasses the tendons were removed, while on others the phalanges were disarticulated (table 7). The same is true of the axial skeletons, which were subjected to a variety of different butchery operations during the experimental sessions (table 8). For sanitation reasons, the carcasses were generally eviscerated prior to skinning. However, in order to document the range of possible functions of Mousterian triangular lithic pieces, some carcasses were first speared and then skinned prior to evisceration (table 7).

18As one of our objectives was to connect specific cutmarks to specific tools using morphological criteria, each red deer half-carcass was processed with the same type of tool and each individual tool was replaced by another of the same type once the experimenter found it to be ineffective. The tools used in the butchery activities are described in table 4. Other types of tools were sometimes used on the skull and the post-cranial axial skeleton. They were systematically noted on the recording sheets and are listed in table 8.

19In order to limit the number of factors that could potentially influence the formation of cutmarks and to test whether the number of cutmarks present on a carcass was influenced by the type of tool and/or the butcher, the archaeozoologists recommended that the entire experimental corpus be processed by only two experimenters. As the technologists wished to test the functionality of their tools, these parameters could not be met. Nonetheless, each half-carcass was systematically treated by a single individual (table 5).

20For each experiment, all operations were timed. The operational sequence for each side was recorded on separate forms. The date, name of experimenter, and tool used were noted on each form. Data recording was performed by the archaeozoologists. The first phases of butchery (skinning, evisceration) were recorded on anatomical templates representing complete skeletons (figure 100a). The operations performed on the axial skeleton were recorded on three different forms: the first one for the cervical vertebrae, the second one for the thoracic vertebrae, ribs and sternebrae, and the third one for the lumbar vertebrae and sacrum (figure 100b-d). Two additional forms were used for the anterior and posterior limbs, respectively. The motions are described chronologically following a continuous numerical notation (figure 100e-f). The positions of actions that could potentially leave cutmarks on the bones were meticulously noted, as well as the orientation and type of gesture (unidirectional incision, bidirectional incision) (figure 101).

Figure 100 - Forms used for recording the gestures used and the instances of tool contact with bone during the experiments

Figure 100 - Forms used for recording the gestures used and the instances of tool contact with bone during the experiments

a: form for skinning activities; b: form for the neck; c: form for the ribcage; d: form for the lumbar vertebrae and the sacrum; e: form for the forelimb; f: form for the hind limb

CAD: M.-P. Coumont

Figure 101 - Example of the form used to record instances of tool contact with bone, the duration of the butchery activities, and comments from the experimenter during the experiment

Figure 101 - Example of the form used to record instances of tool contact with bone, the duration of the butchery activities, and comments from the experimenter during the experiment

CAD: M.-P. Coumont

21In order to avoid altering the morphology of the butchery marks, we decided to bury the bones directly after the butchery process. Each portion of the skeleton was buried separately in a metallic screen while the date of the experiment was carefully noted. After twelve months, the bones were unearthed, cleaned with water and a soft brush, and analysed.

B - Experimental butchery proceduress

22Due to the similarity of the butchery operations performed on the carcasses, only two of the experimental procedures are described in detail. Experiment 1 is exemplary of the procedure followed in the butchery of carcasses that were defleshed and experiment 7 is exemplary of those that were disarticulated. The different operations conducted on each carcass are summarized in tables 7-8. The duration of the operations is provided for informational purposes, but given our inexperience in these activities, especially in the case of the first carcasses, they should in no way be considered representative of the actual time required in prehistoric contexts and were not factored in to our interpretations.

a - Butchery of red deer 1

23This butchery experiment was conducted in October 2007 on the headless carcass of a four years old red deer. The right side was processed with three flint denticulates almost entirely by C. Thiébaut, with the exception of the removal of the tendons, which was performed by L. Streit. The left side was processed with two Mousterian points by A. Coudenneau and then by S. Costamagno. The whole butchery lasted 6 hours.

Skinning and evisceration

24Skinning took approximately two hours per side. The experimenters were confronted with the strong adherence of the hide to the muscular tissues. Without doubt, our lack of experience prolonged the operation considerably. Additionally, not all tools were equally effective, the Mousterian point proving more suited to the activity than the denticulate.

25The animal had been castrated and then eviscerated prior to our experiment. The incision of the skin, already initiated prior to our intervention, was extended to the anus with the use of a Mousterian point. The incision was extended in the opposite direction with a denticulate as far as the atlas. On the anterior and posterior limbs, a circular incision was made in the skin at the distal end of the metapodial shafts (figure 102a). From there, a longitudinal incision was made along the internal face of the limb, joining the evisceration incision (figure 102b). The hide could then be detached from the limbs (figure 102c) and removed from the ribcage (figure 102d), the back, and the neck. The contacts of the tool with the bone as well as the primary gestures employed are presented in figure 103. They are located mostly on the metapodials, the calcaneus and the innominate.

Figure 102 - Skinning of red deer 1

Figure 102 - Skinning of red deer 1

a: circular incision on the metapodial and start of the longitudinal incision on the skin of the medial surface (denticulate); b: incision on the medial surface of the limb (Mousterian point); c: detachment of the hide from the ribcage (denticulate); d: detachment of the skin from the tibia (Mousterian point)

Photographs: PCR Des Traces et des Hommes

Figure 103 - Form for recording the gestures used in the skinning of red deer 1

Figure 103 - Form for recording the gestures used in the skinning of red deer 1

a: left side with the location of the incision of the skin and the instances of contact with bone (Mousterian points); b: right side (denticulates). The bones indicated in grey made contact with tools, without precise contact location

CAD: M.-P. Coumont

Disarticulation and defleshing of the anterior limb

26The anterior limbs were detached from the axial skeleton at the scapula. On both sides, contacts with the tools were noted on the first ribs and the scapular cartilage.

27Defleshing of the scapula was primarily achieved with longitudinal gestures on the medial and lateral faces (figure 104) and a similar approach was used on both sides of the carcass. On the humerus, the gestures and points of contact with the tools were relatively similar on both sides: the gestures were mostly oblique and the contacts with the tool were noted on the major tubercle, the deltoid tuberosity, the teres major tuberosity, and the epicondyles. The muscles attached to the radius and ulna were detached at the olecranon (flexor carpi ulnaris, flexor digitorum profundus), then at the anterior surface of the radius (biceps brachii, brachialis, extensor digitorum, extensor carpi ulnaris), and finally on the posterior surface along the length of the ulna (extensor digitalis, flexor digitorum profundus) (figure 105a). At the distal end of the radius, the intersections of the radial extensor muscles and the extensor carpi ulnaris muscle were incised with unidirectional gestures from the left side and bidirectional gestures from the right side.

Figure 104 - Form for recording defleshing and tendon removal on the forelimbs of red deer 1

Figure 104 - Form for recording defleshing and tendon removal on the forelimbs of red deer 1

a: right side (denticulates); b: left side (Mousterian points)

CAD: M.-P. Coumont

Disarticulation and defleshing of the posterior limb

28For the dismemberment of the posterior limbs, contacts with tools were noted at the femoral head and around the acetabulum of the pelvis. The defleshing of the pelvis occurred with disarticulation as the meat detached from the posterior limb (figure 105b). The position and especially the orientation of the gestures are not identical on the two sides of the carcass (figure 106). In spite of the variety of gestures, the predominance of longitudinal motions for removing muscles from the femoral shaft is worth noting, particularly on the lateral surface. Bidirectional gestures proved especially effective on the right side, butchered with a denticulate. For the tibia, bidirectional gestures were more frequently used on the left side, in particular for cutting through the tendons and the muscle masses at the distal end of the shaft. As was the case for the femur, the removal of muscles from the mid-section of the left tibial shaft (peroneus tertius, tibialis posterior, anterior and posterior flexor digitorum profundus) was primarily achieved with longitudinal gestures. For the right side, the gestures used were primarily transverse or oblique to the longitudinal axis of the bone.

Figure 105 - a: defleshing of the left radius and ulna of red deer 1; b: disarticulation of the coxofemoral joint of the left hind limb of red deer 1

Figure 105 - a: defleshing of the left radius and ulna of red deer 1; b: disarticulation of the coxofemoral joint of the left hind limb of red deer 1

Photographs: PCR Des Traces et des Hommes

Removal of tendons

29At the metapodials, the tendons were removed from the anterior and posterior surfaces. The extensor and flexor tendons were sectioned at the proximal and distal ends of the diaphysis, but the bone did not appear to come into contact with the tool systematically. On the right side, contacts between bone and tool were observed on the posterior surface on either side of the central trough and longitudinal contacts were observed on the medial and lateral surfaces (figure 104a). On the left side, the points of contact observed were primarily related to longitudinal gestures on the anterior and posterior surfaces (figure 104b). On the metatarsals, the gestures employed to remove the tendons differed substantially from one side to the other. On the right side, the cutting of the extensor digitorum tendon created points of contact at the proximal end of the diaphysis. On the posterior surface, a longitudinal contact developed in the interior of the central trough (figure 106a). On the left side, after the tendons were cut at the proximal end of the diaphysis, the tool was used to progressively detach the tendons from the bone with bidirectional movements transverse to the long axis of the bone. On the posterior surface, the tendons were cut through at the distal end of the diaphysis, whereas longitudinal gestures were employed on the anterior surface (figure 106b).

Figure 106 - Form for recording the defleshing and tendon removal on the hind limbs of red deer 1

Figure 106 - Form for recording the defleshing and tendon removal on the hind limbs of red deer 1

a: right side (denticulates); b: left die (Mousterian points)

CAD: M.-P. Coumont

Disarticulation of the phalanges and metapodials

30The disarticulation of the phalanges was only performed on the posterior limbs: the first phalanges were separated from the metatarsals. An incision was made all the way around the left metatarsal epiphysis and only on the posterior surface on the right metatarsal (figure 107). The first phalanx was then detached with incisions at the proximal end.

Defleshing and segmentation of the post-cranial axial skeleton

31These operations proceeding in several stages and the same approach was not followed for the two sides (figure 108).

32- Right side:

  • removal of the sirloin: longitudinal incisions on the spinous processes of the thoracic vertebrae and incisions on the transverse processes of the lumbar vertebrae;

  • removal of the muscles of the neck with oblique incisions on all the cervical vertebrae;

  • removal of the meat on the ribs by long incisions perpendicular to the long axis of the ribs;

  • removal of the tenderloin with points of tool contact observed on the internal surfaces of the ribs and vertebrae;

  • separation of the ribs between the eighth and ninth ribs;

  • separation of the ribs between the fourth and fifth ribs. Percussion with a quartz cobble (figure 109d). Hyperextension at the ventral extremity of the ribs to detach them completely. Prior to percussion, the zone of impact was prepared by scraping to remove the meat using one of the notches of a denticulate.

Figure 107 - Gestures related to the disarticulation of the phalanges of a red deer 1

Figure 107 - Gestures related to the disarticulation of the phalanges of a red deer 1

CAD: M.-P. Coumont

Figure 108 - Gestures related to the defleshing and disarticulation of the post-cranial axial skeleton

Figure 108 - Gestures related to the defleshing and disarticulation of the post-cranial axial skeleton

a: right side; b: left side

CAD: S. Costamagno

Figure 109 - Defleshing and segmentation of the axial skeleton of red deer 1

Figure 109 - Defleshing and segmentation of the axial skeleton of red deer 1

a: defleshing of the neck; b: removal of the sirloin; c: defleshing of the ribs; d: disarticulation of the ribs by percussion; e: incision between the ribs; f: detachment of the ribs by flexion and sawing

Photographs: PCR Des Traces et des Hommes

33- Left side:

  • removal of the muscles of the neck with longitudinal incisions on all the cervical vertebrae (figure 109a);

  • removal of the sirloin: longitudinal incisions on the spinous processes of the lumbar and thoracic vertebrae (figure 109b). Short incisions on the lumbar vertebral bodies. Sawing between the vertebral bodies of the two penultimate lumbar vertebrae. Incisions perpendicular to the long axis of the ribs on the final four ribs;

  • removal of the meat on the ribs by long incisions perpendicular to the long axis of the ribs (figure 109c);

  • separation of the ribs from the sternum: incisions, percussion and transverse sawing of the fourth through eighth ribs to detach the sternum;

  • removal of the tenderloin with tool contact on the internal faces of the ribs and vertebrae;

  • disarticulation of the vertebral column: incisions along the length of ribs 8 and 9 and ribs 5 and 6 up to the vertebrae (figure 109e), incisions on the vertebral bodies, segmentation between the eighth and ninth thoracic vertebrae by hyperextension (figure 109f).

b - The butchery of red deer 7

34After skinning, the carcass was disarticulated and meat removal was performed with precaution in order to avoid contact between the tool and the bone to the greatest extent possible. The tendons were not removed. All incidents of contact were noted on the plates, with a distinction made between cutmarks related to defleshing and cutmarks related to disarticulation. Unmodified flakes in fine quartzite were used for the removal of the skin and the disarticulation of the limbs; unmodified flakes in flint were used on the axial skeleton.

Skinning

35Skinning was performed on the left side by G. Chácon and on the right side by S. Costamagno. On the anterior limbs and on the posterior limbs, a circular incision was made in the skin at the mid-shaft of the metapodials (figure 110b), then the skin was sliced longitudinally along the internal surface of the legs (figure 110a).

Figure 110 - Skinning of red deer 7

Figure 110 - Skinning of red deer 7

a: incision of the skin on the internal surface of the right hind limb; b: circular incision of the skin at the shaft of the right metatarsal; c: incision of the skin on the sternum

Photographs: PCR Des Traces et des Hommes

36On the sternum, a longitudinal incision was made to join the circular incision around the neck (figure 110c). The head was skinned once separated from the vertebral column, but the points of contact were not recorded.

Disarticulation and defleshing

37In order to limit the potential variables that might influence the formation of cutmarks, disarticulation was performed by a single experimenter: S. Costamagno. The bones of the left limbs were defleshed and then disarticulated, while the bones of the right limbs were disarticulated and then defleshed.

38- Left side

39The forelimb was defleshed after separation from the trunk (figure 111a). The muscles were incised at the sites of tendon attachments (figure 111b). On the humerus, the muscles were then sliced with careful attention to avoid contact with the bone. On the scapula and the radius-ulna, the muscle masses could be pulled from the bones without use of the flake. In spite of the precautions taken, some incidents of tool contact with the scapula were noted on the lateral surface of the supraspinous fossa and around the supraglenoid tubercle on the medial and lateral surfaces. On the humerus, a point of contact was recorded on the proximal end of the diaphysis on the lateral surface and on the major tubercle on the anterior surface. Finally, on the ulna, the tool made contact with the medial surface of the olecranon. Once it was defleshed, the disarticulation of the limb was relatively easy, as the recorded duration of the process shows. In less than one minute, the scapulohumeral joint was disarticulated and no tool contact was observed (figure 111c). The elbow was disarticulated in six minutes: once the ligaments were segmented, the disarticulation was performed by torsion (figure 111d). Several incidents of contact were noted on the distal epiphysis of the humerus: the lateral and posterior surfaces of the condyle, the medial and posterior faces of the trochlea and the interior of the olecranon fossa. On the ulna, tool contact was recorded on the medial and lateral surfaces of the olecranon and in the trochlear notch. The radius-ulna and the metacarpal were disarticulated with incisions to the mediocarpal ligaments of articulation followed by torsion (figure 111e). The process took five minutes and points of tool contact were indicated on all faces of the carpus. We then attempted to separate the antebrachiocarpal joint, but after four minutes we gave up, as it proved too difficult to reach the interosseous ligaments (figure 111f) on the posterior face in order to cut them. Tool contact was recorded on the anterior and lateral faces of the distal radius and on the anterior and lateral faces of the carpus.

40The hind leg was defleshed without being detached from the pelvis (figure 112a). As with the foreleg, the muscles were only incised at the location of the tendon attachments. At the pelvis and the femur, the muscles were carefully cut in order to avoid contact with the bones (figure 112b). On the tibia, it was possible to pull the muscles away from the bone without using the flake. Points of tool contact related to defleshing were indicated on the lateral face of the pelvis near the ischium and ilium. On the femur, the intertrochanteric crest, the lateral surface of the distal diaphysis and the lateral lip of the trochlea made contact with the tool. The coxofemoral articulation was then disarticulated in four minutes (figure 112c) and tool contact was noted on the femoral neck (anterior and posterior surfaces), in the trochanteric fossa, and on the anterior surface of the femoral head. Disarticulation of the knee took eight minutes. The ligaments were cut, which created several points of tool contact on all surfaces of the distal femur, and then the joint was disarticulated by torsion. For the disarticulation of the zeugopod and autopod, incisions were made at the crurotarsal joint. Under torsion, the talus detached with the tibia and the rest of the tarsus remained attached to the metatarsal (figure 112e). The process took 17 minutes. Points of tool contact were noted on the posterior and lateral surfaces of the distal tibia, on all surfaces of the talus, and on the lateral surface of the calcaneus.

Figure 111 - Defleshing and disarticulation of the left forelimb of red deer 7

Figure 111 - Defleshing and disarticulation of the left forelimb of red deer 7

a: removal of the limb; b: cutting the tendons and ligaments for removal of the muscles; c: disarticulation of the scapula and the humerus; d: disarticulation of the elbow; e: disarticulation of the “wrist”; f: disarticulation of the radius and the first row of carpal bones

Photographs: PCR Des Traces et des Hommes

41- Right side

42Disarticulating the bones that had not first been defleshed was highly challenging, not only in terms of locating the right point of disarticulation, but also in terms of accessing the ligaments for incision and attempting to disarticulate by torsion. Disarticulation of the right side therefore took considerably longer than disarticulation of the left side (59 minutes as opposed to 19 minutes).

43On the forelimb, disarticulation of the shoulder resulted in points of tool contact on the lateral surface of the proximal humerus and the gleniod fossa of the scapula (figure 113a-b). Tool contacts on the medial and posterior surfaces of the distal humerus resulted from the trial-and-error approach to locating the elbow joint (figure 113c-e). Points of tool contact were also noted on the anterior surface of the trochlea, the olecranon fossa, the lateral face of the humeral condyle, and the medial and anterior surfaces of the radius-ulna. The radius was then disarticulated from the lower leg, and tool contacts were noted on the lateral, medial, and posterior surfaces of the radius and the lateral surface of the ulna (figure 113f). The approach to defleshing the bones was similar to that taken on the left foreleg.

Figure 113 - Defleshing and disarticulation of the left forelimb of red deer 7

Figure 113 - Defleshing and disarticulation of the left forelimb of red deer 7

a-b: disarticulation of the shoulder; c-e: disarticulation of the elbow; f: disarticulation of the radius from the first row of right carpal bones

Photographs: PCR Des Traces et des Hommes

44On the hind leg, points of tool contact were noted on the posterior and medial surfaces of the femoral head and on the intertrochanteric crest (figure 114a). The points of tool contact indicated on the anterior and posterior surfaces of the distal femoral shaft are a result of the trial-and-error process of finding the knee joint (figure 114b). Additional points of contact directly related to the disarticulation of this element were noted on all of the faces of the distal femur, including the inferior surface of the condyles. On the tibia, tool contact was noted on the lateral surface and intercondylar eminences. Twenty‑four minutes were required to disarticulate the right knee, compared to the eight minutes required for the (defleshed) left knee. During separation of the tibia from the lower leg, the tarsus detached completely from the tibia with disarticulation by rotation. Points of tool contact were observed on the posterior surface of the tibia, the lateral surface of the calcaneus, and the lateral, anterior, and medial surfaces of the talus. Disarticulation of these two segments took eight minutes for the right leg, compared to 17 for the left: in the latter case, it was the detachment of the talus with the tibia that complicated the process. We then proceeded to disarticulate the tarsometatarsal joints (figure 114c,e). This difficult process took 15 minutes and ultimately resorted to disarticulation by force. Tool contact was noted on all surfaces of the proximal metatarsal and the tarsus. The bones were then defleshed (figure 114d).

Figure 114 - Defleshing and disarticulation of the right hind limb of red deer 7

Figure 114 - Defleshing and disarticulation of the right hind limb of red deer 7

a: disarticulation of the coxofemoral joint of the right limb; b: disarticulation of the right knee; d: detachment of the muscles inserted into the tibia; c, e: disarticulation of the tarsal and metatarsal

Photographs: PCR Des Traces et des Hommes

45- Axial skeleton

46The sirloin and tenderloin were removed primarily by hand. Two points of tool contact were recorded on the lumbar vertebrae related to removal of the sirloin. The neck and the ribs were also defleshed and some points of tool contact were noted on the third, fifth, and sixth cervical vertebrae and some of the ribs. The skull was crudely defleshed after separation from the vertebral column at the atlas, but incidents of tool contact were not noted. One each side, the three final ribs were disarticulated with an incision of the intercostal muscle connecting the tenth and eleventh ribs followed by forceful extension. Ribs 8, 9, and 10 were separated together, as were ribs 4, 5, 6, and 7. After incision of the intercostal muscles and disarticulation of the costal and sternal cartilage by incision or forceful fracture, the ribs were separated from the thoracic vertebrae by forceful extension. The first three ribs were not separated from the vertebrae. The vertebral column was separated into six sections: cervical 6-cervical 7, thoracic 2-thoracic 3, thoracic 8- thoracic 9, lumbar 3-lumbar 4, lumbar 6-sacrum.

47Contrary to our expectations, the bones defleshed during butchery of this carcass yielded very little meat in the end, as can be seen in the photos (figure 115). The wind had dried the epimyseum of the muscles, permitting us to pull the majority of them away from the skeleton without making use of the tools.

Figure 115 - State of the bones after the defleshing of red deer 7

Figure 115 - State of the bones after the defleshing of red deer 7

a: left forelimb; b: left hind limb; c: right forelimb; d: right hind limb

Photographs: PCR Des Traces et des Hommes

Details regarding the limb skinning and the lower extremities processing

48There are few available references that address the extremities of the limbs in any detail. Yet, with regard to skinning and the extraction of tendons, these elements are highly important sources of information. Several different operations were therefore performed on the various carcasses with the aim of defining the stages and gestures involved (table 7).

49With regard to skinning, generally speaking, only one circular incision was made on the limbs. The skin was most often incised on the metapodials, either on the diaphyses (distal end n=11, medial section n=11) or at the distal extremity (n=4). On red deer 6, the incisions were made on the first phalanges: the experimenters noted that separating the skin from this starting point was the most difficult, due to the presence of the vestigial phalanges. On the right side of red deer 2, the starting incision was made on the distal end of the tibia. On the right hind leg of red deer 5 and on all of the limbs of red deer 8, incisions were also made closer to the trunk, ultimately separating the skin of the limbs from that of the trunk. On these limbs, the skin left on the metapodials was removed. From these starting incisions, in order to facilitate removal of the skin, longitudinal incisions were made on different surfaces: one on the posterior surface (red deer 8 right foreleg), two on the medial surface (red deer 2 right hind leg, red deer 5 right hind leg), three on the anterior surface (red deer 8 right metacarpal; red deer 8 right and left metatarsal), one on the lateral surface (red deer 2 right hind leg). On red deer 8, a final incision was made closer to the hooves of the four limbs. In these cases, from the incisions on the metapodials, longitudinal incisions were made on the posterior surface, with the exception of the phalanges of the left hind leg.

50The disarticulation of the metapodials and the phalanges was performed on eight limbs (hind leg of red deer 1, red deer 2, left foreleg of red deer 5, right foreleg of red deer 6). No disarticulation was performed between the proximal phalanx and the intermediate phalanx. The only case of disarticulation of phalanges was the separation of the proximal phalanges on the right hind limbs of red deer 6 and 8 with longitudinal gestures.

51The extensor tendons were removed from nine metacarpals and nine metatarsals; the flexor tendons from eight metacarpals and nine metatarsals. Transverse or longitudinal gestures were employed. These tendons were most frequently sawed near the proximal and distal extremities of the metapodial diaphyses. On red deer 2, right side, the sectioning of the tendons was performed at the articular extremities of the metapodials. On red deer 6, these gestures were localised on the first phalanges.

3 - Analysis of the location of cutmarks: relating the cutmarks to the activities and gestures of the butcher

S. Costamagno, M.-C. Soulier

52As we have already noted above, in all human societies, food and foodways are at the heart of identity practices (Havelange, 1998; Lalhou, 1998; Fischler, 2001; Serra Mallol, 2010). The ethnographic literature also demonstrates the crucial role that food plays in economic systems as well as in social and symbolic systems (Yellen, 1977a; Lyman, 1987; Simoons, 1994; Vialles, 1998; Chenal‑Velarde, Velarde, 2004; Russel, 2012). Numerous rules, rituals, and taboos dictate the way in which food are consumed (by whom, what, when, and how?) (for example, Guevara, 1988; Malaurie, 1989; Simoons 1994; Stefansson, Palsson, 2001; Politis, Saunders, 2002; d’Iatchenko et al., 2007). These rules can take various forms, sometimes extending through the final stages of the animal carcass processing. Even if the social and symbolic elements encoded in butchery practices are clearly difficult to perceive for archaeozoologists interested in the deep past, a detailed reconstruction of the chaîne opératoire of the exploitation of carcasses is the only means of potentially identifying butchery traditions. These constitute cultural expressions that are just as informative as the knapping of stone or the shaping of stone tools. Fully realizing the informative potential of faunal remains requires the development of new methodological approaches to complete existing interpretive frameworks (see Part I, chapter 3.1). More accurate and precise decoding of butchery cutmarks will allow researchers to relate specific cutmarks unequivocally to different phases of the butchery chaîne opératiore and connect them to specific butchery gestures as potential indications of the strength of cultural traditions in the animal carcass processing. As the constraints that accompany the use of certain tool types can also influence the gestures of butchers, experimental approaches must adhere strictly to a particular archaeological point of reference, in this case the Mousterian, with the use of different types of tools directly related to the assemblages included in the study.

A - Study material

53The material studied is comprised exclusively of the nine complete red deer carcasses (table 5).

B - Method of study

54The bone surfaces from the red deer carcass processed in October 2017 were analysed by three different individuals: two with the aid of a hand lens (10 × magnifying power) and the third with the aid of a stereo microscope (30 × magnifying power). This triple analysis confirmed that the effects of inter-observer variation were negligible in the identification of cutmarks. Furthermore, it demonstrated that the stereo micropscope provided relatively few advantages over the hand lenses. On certain bones, use of the microscope did not result in the identification of additional cutmarks, and when new cutmarks were observed, they usually accompany sets of marks that had already been identified with the hand lens (see Coumont et al. in Thiébault et al., 2008 for further details). These initial observations were confirmed by a second analysis conducted as part of the Master 2 research of S. Chong (2011). Thus, though the use of a stereo microscope does allow for a more complete count of cutmarks, it does not yield sufficient informative results with regard to the research questions posed and the amount of time required to justify its systematic use. The remaining carcasses were therefore analysed by two observers, with the use of a hand lens and under oblique light.

55Once identified each cutmark was recorded on graphic templates depicting each of the surfaces of each bone. Rather than create a template unique to each carcass, we opted to use distinct layers in Adobe Illustrator®, which allowed us to superimpose the notations carcass by carcass.

56In order to facilitate discussion of the location of cutmarks, the long bones were subdivided into six segments for the templates. For a single skeleton, all of the cutmarks are illustrated on the right side. The reports include the total count of cutmarks for the entire carcass. The templates are presented by activity type, with the exception of the cranial and pelvic elements. In fact, during butchery, tool contacts with these elements were not recorded, and it was therefore not possible to relate the cutmarks to specific activities (skinning, defleshing, disarticulation). With few exceptions, the protocol followed did not permit the differentiation of disarticulation cutmarks from defleshing cutmarks on the pelvis.

C - Results

a - Skinning

57Skinning of the limbs required the use of a range of different gestures: circular incisions, longitudinal incisions, and oblique gestures for the detachment of the skin from the underlying tissue. The cutmarks related to the circular incision were primarily transverse, relatively deep, and occurred most often in clusters. They occurred principally on the medial and lateral surfaces (red lines; figure 116). Based on their placement, it is therefore possible to precisely locate the starting-points.

Figure 116 - Cutmarks related to skinning of a: the metacarpal; b: the metatarsal; c: the tibia; d: the phalanges; e: the vestigial phalanges; f: pyramidal, lunate and pisiform; g: the calcaneus and the cubonavicular bones

Figure 116 - Cutmarks related to skinning of a: the metacarpal; b: the metatarsal; c: the tibia; d: the phalanges; e: the vestigial phalanges; f: pyramidal, lunate and pisiform; g: the calcaneus and the cubonavicular bones

Abbreviations: M = medial, A = anterior, L = lateral, P = posterior, Ab = abaxiale

CAD: M.-C. Soulier and M. Coutureau

58When the circular incision was made on the articular extremities of the metapodials, cutmarks could be present on the vestigial phalanges. As the second phalanges are partially embedded in the hoof, incisions made at this location can leave cutmarks on them. Even so, such cutmarks are rare because the hoof often protects the surface of the bone.

59The longitudinal incision – which connects the circular incision on the limb to the ventral incision – generates long longitudinal cutmarks, often in continuous succession (blue lines; figure 116). Due to the presence of the extensor tendon on the posterior surface of the limb, longitudinal incisions made in this position left no cutmarks. When the longitudinal incision is made on the medial surface, cutmarks can develop on the scantly muscled distal end of the tibia, calcaneus, cuboideonavicular joint, and metapodials. On the lateral surface, cutmarks are present on the metapodials, and the triquetral and pisiform. On the anterior surface, longitudinal cutmarks were only observed on the lunate.

60The detachment of the skin from the underlying tissue with tools requires fairly acute oblique gestures, in order to avoid damaging the skin with the blade. These gestures produce cutmarks that are oblique and superficial and often isolated on the lateral and medial surfaces of the metapodials (green lines, figure 116). Oblique cutmarks were also noted amongst the clusters left by the initial circular incision. These two types of cutmarks are nevertheless relatively easy to distinguish from one another based on their general appearance (deep or superficial, clustered or isolated) (figure 117). On the mandible, oblique cutmarks on the vestibular surface of the horizontal ramus are likewise related to the skinning process (figure 118a). The cutmarks observed on the cheek teeth correspond to the incision of the skin, as do those located on the mandibular symphysis and incisors. No such cutmarks were observed on any other bones – but recall that they were not recorded for the skull.

Figure 117 - Cutmarks from skinning on the metatarsal

Figure 117 - Cutmarks from skinning on the metatarsal

Photographs: M.-C. Soulier

b - Defleshing

61Cutmarks related to defleshing are present on all of the meaty bones. On the mandible, they are mostly observed on the vestibular surface of the ascending ramus. Cutmarks observed on the lingual surface of the horizontal ramus are related to removal of the tongue (figure 118b). On the vertebrae, cutmarks are present on the axis, the dorsal surfaces of the cervical vertebrae, the spinous process of the thoracic vertebrae, the transverse processes and spinous process of the lumbar vertebrae as well as on the wing of the sacrum (figure 119). On the thoracic and lumbar vertebrae, cutmarks related to the removal of the sirloin appear on the craniocaudal axis. On the ribs, cutmarks from defleshing are oblique or transverse and present on the lateral surfaces. At the proximal end, they result from the removal the sirloin and/or meat from the ribs, while at the distal end they relate only to the latter operation. On the coxal bone, the cutmarks related to defleshing could not be distinguished from those related to dismemberment of the hind leg, with the exception of those localized on the wing of the ilium.

Figure 118 - Cutmarks from butchery on the mandible

Figure 118 - Cutmarks from butchery on the mandible

a: cutmarks related to skinning; b: cutmarks related to defleshing; c: cutmarks related to disarticulation. Abbreviations:V = vestibular, L = lingual

CAD: M.-C. Soulier and M. Coutureau

62On long bones, cutmarks are primarily transverse or oblique (figure 120). Longitudinal cutmarks were also observed, primarily on the scapula and ulna. Cutmarks related to defleshing are also present at the articular extremities: intercondylar plateau and eminence of the tibia, humeral head, trochlear notch of the ulna, intertrochanteric crest of the femur. Cutmarks related to the sectioning of the tendon of the superficial flexor digitorum muscle, necessary for removal of the muscles of the tibia, are present on the calcaneus.

Figure 119 - Cutmarks related to defleshing on a: the axis; b: the other cervical vertebrae; c: the sacrum; d: the thoracic vertebrae; e: the lumbar vertebrae

Figure 119 - Cutmarks related to defleshing on a: the axis; b: the other cervical vertebrae; c: the sacrum; d: the thoracic vertebrae; e: the lumbar vertebrae

Abbreviations: Cr = cranial, D = dorsal, L = lateral, V = ventral

CAD: M.-C. Soulier and M. Coutureau

Figure 120 - Cutmarks related to defleshing on a: the scapula; b: the calcaneus; c: the humerus; d: the radioulnar/ ulna; e: the femur / the patella; f: the tibia / the fibula

Figure 120 - Cutmarks related to defleshing on a: the scapula; b: the calcaneus; c: the humerus; d: the radioulnar/ ulna; e: the femur / the patella; f: the tibia / the fibula

Abbreviations: M = medial, A = anterior, L = lateral, P = posterior

CAD: M.-C. Soulier and M. Coutureau).

c - Disarticulation

63On the mandible, disarticulation from the skull left cutmarks near the mandibular condyle as well as on the coronoid process (figure 118c).

64The separation of the scapula from the trunk did not leave any cutmarks. In fact, due to the absence of an articular surface, dismemberment of the foreleg only required severing the muscle masses. For the pelvis, the cutmarks present on the acetabulum could be attributed to the separation of the coxal bone and the femur. The other cutmarks on the pelvis, though, are more difficult to interpret. In any event, given the difficulty of manipulating an entire ungulate carcass, the defleshing of the hind leg of the ungulates can only occur once the limbs are separated from the trunk. Thus, disarticulation and defleshing of the coxal region are closely interrelated activities.

65On the long bones, the cutmarks from disarticulation are primarily short, with transverse or oblique orientations; they are all located near the articular surfaces. Longitudinal cutmarks were also noted on the distal extremities of the humerus, femur, and metapodials as well as on the pisiform and near the articular surface of the malleolus and the lateral surface of the calcaneus (figure 121). No cutmarks were observed on the scapulohumeral joint. The tibia bears no tool-traces related to disarticulation, but there are cutmarks present on the fibula and malleolus. At the distal ends of the limbs, cutmarks are recorded at the proximal end of the first and second phalanges as well as the sesamoid and first vestigial phalanges.

66The long bones of two carcasses were disarticulated and on these, the cutmarks are distant from the articular surfaces (blue lines; figure 121). Noted on the limbs that were disarticulated before being defleshed (right side of red deer 7, right side of red deer 9), they are related to a phase of trial-and-error in locating the joint (see figure 113c-e). Due to their location, we can say that these cutmarks were not directly related to the disarticulation of these bones.

67On the vertebrae as well, signs of trial-and-error could be seen on the atlas, the axis, the cervical vertebrae, and the thoracic vertebrae (blue lines, figure 122). These cutmarks are mostly ventrodorsal in orientation but, unlike the cutmarks related to the disarticulation of the vertebrae (black lines, figure 122), are not located near the articular zones. Cutmarks present on the ventral or caudal edges of the ribs are related to the separation of the ribs.

d - Removal of the tendons

68The removal of the nuchal ligament left no tool-traces on the dorsal surfaces of the spinous processes on the thoracic and lumbar vertebrae.

69For the removal of tendons at the extremities of the limbs, cutmarks were primarily observed on the metapodials, phalanges, and sesamoids (figure 123). The cutmarks related to the segmenting of the tendons are located exclusively near the articular extremities of the bones, as the aim was to obtain the longest possible sections of tendon. The cutmarks are short, transverse or oblique, and usually deep. The longitudinal cutmarks observed on the posterior surface of the metatarsal condyles result from the separation of the large sesamoids, which was performed in order to remove the entire length of the interosseous muscle. On the anterior surface, the longitudinal cutmarks indicate that the removal of the extensor tendons continued on the phalanges.

70The small, short, transverse cutmarks that formed on either side of the central troughs indicate that the tool was held in a transverse position while the tendons were removed (figure 124a,c). Longitudinal cutmarks indicate that the tool passed under the tendon to detach it (figure 124b,d). These cutmarks are especially prevalent on the posterior surface because, on deer, the flexor tendons are inserted in relatively deep troughs. Longitudinal cutmarks present on the abaxial surface of the distal extremity of the first phalanges indicate removal of the deep flexor digitorum tendons at the phalanges, as do oblique cutmarks on the posterior surface. On the anterior surface, they indicate the removal of the extensor digitorum tendons.

Figure 121 - Cutmarks related to the disarticulation of the limbs on a: the humerus; b: the radioulnar/ ulna; c: the femur; d: the fibula, the sesamoids, the malleolus bone; e: the metacarpal; f: the metatarsal; g: the carpal bones; h: the tarsal bones; i: the first phalanges; j: the second phalanges and the vestigial phalanges. The cutmarks from testing for the location of the joint are in blue

Figure 121 - Cutmarks related to the disarticulation of the limbs on a: the humerus; b: the radioulnar/ ulna; c: the femur; d: the fibula, the sesamoids, the malleolus bone; e: the metacarpal; f: the metatarsal; g: the carpal bones; h: the tarsal bones; i: the first phalanges; j: the second phalanges and the vestigial phalanges. The cutmarks from testing for the location of the joint are in blue

Abbreviations: M = medial, A = anterior, L = lateral, P = posterior, Ab = abaxial

CAD: M.-C. Soulier and M. Coutureau

Figure 122 - Cutmarks related to the segmentation of the vertebral column on a: the atlas; b: the axis; c: the other cervical vertebrae; d: the thoracic vertebrae; e: the lumbar vertebrae. The cutmarks from testing for the location of the joint are in blue

Figure 122 - Cutmarks related to the segmentation of the vertebral column on a: the atlas; b: the axis; c: the other cervical vertebrae; d: the thoracic vertebrae; e: the lumbar vertebrae. The cutmarks from testing for the location of the joint are in blue

Abbreviations: Cr = cranial, Ca = caudal, D = dorsal, M = lateral, V = ventral

CAD: M.-C. Soulier and M. Coutureau).

Figure 123 - Cutmarks related to the extraction of the tendons on a: the metacarpal; b: the metatarsal; c: the first phalanges

Figure 123 - Cutmarks related to the extraction of the tendons on a: the metacarpal; b: the metatarsal; c: the first phalanges

Abbreviations: M = medial, A = anterior, L = lateral, P = posterior, Ab = abaxial

CAD: M.-C. Soulier and M. Coutureau).

Figure 124 - Removal of the tendons

Figure 124 - Removal of the tendons

a: transverse gesture for removal of the tendon; b: longitudinal gesture for removal of the tendon; c: cutmarks characteristic of transverse gestures used for tendon detachment; d: cutmarks characteristic of longitudinal gestures for tendon detachment

CAD: M.-C. Soulier

D - Discussion

a - The T&H reference collection in comparison to other available reference collections

71As P. J. Nilssen (2000) has emphasized, the gestures of the butcher – and by extension the cutmarks they leave – can differ with the type of tool used in butchery (metal vs lithic tool). Morphological differences in butchery cutmarks have also been documented (Greenfield, 1999). Several studies conducted on butchery operations conducted with metal knives have noted the presence of “shave marks” (Nilssen, 2000; Abe, 1005; Costamagno, David 2009; Soulier personal observation 2015, Binford Numamiut collection), a type of cutmark that has not been observed in the T&H reference collection or in any of the Palaeolithic faunal assemblages we have analysed. No cutmark related to skinning were present on the posterior surface of the phalanges in the T&H reference collection because the powerful flexor tendons prevent tool contact with the bone (figure 116). In contrast, longitudinal cutmarks related to skinning performed with metal knives have been observed on the posterior surface of the phalanges (Costamagno, David, 2009: 16). Unlike Mousterian lithic tools, long and pointed knives seem to pass through these sturdy tissues easily. Butchery conducted with metal knives also tends to produce cutmarks that are longer than those observed in Palaeolithic faunal assemblages (Soulier, Morin, 2016). The tools used for the butchery of carcasses therefore seems to have an influence on the gestures employed by the butcher and, as follows logically, the traces they leave on bone. All of the experimental butchery performed in the PCR was performed with lithic tools. This reference collection therefore constitutes a more relevant example than those of L. R. Binford (1981) or P. J. Nilssen (2000) for the interpretation of butchery cutmarks on faunal assemblages from the Palaeolithic, regardless of the period.

72In contrast to the reference collections of Binford (1981) and Nilssen (2000), the butchery for the T&H reference collection was performed by archaeologists. According to the experimental work of M. Padilla (2008), novice butchers employ different gestures than do professional butchers. However, the novice butchers observed by Padilla had little to no knowledge of animal anatomy, which is not the case for the experimenters performing butchery in the PCR sessions. M. Padilla (2008) also emphasizes that seasoned butchers tend to leave fewer cutmarks on bone compared to novice butchers, a point with which we concur. The professional butchers who participated in our experiments clearly stipulated that tool‑bone contact is inevitable when one removes the deeper muscle tissues. In the T&H reference collection, a high number of cutmarks, with varied orientations (from transverse to longitudinal), was generated and this distribution closely matches that documented on Palaeolithic faunal assemblages (for example, Cho, 1998; Costamagno, 1999; Castel, 2011; Soulier, 2013; Soulier, Morin, 2016).

73For the T&H reference collection all butchery activities were controlled, with annotation of each instance of contact between tool and bone as well as the associated gesture. Furthermore, with the exception of the pelvofemoral joint, the butchery activities were conducted in isolation from one another, allowing us to unequivocally attribute each cutmark to a specific activity. This is not the case in the work of P. J. Nilssen, for example, in which numerous cutmarks could not be attributed to a specific activity. As can be seen in figures 119-120, the cutmarks from disarticulation and defleshing present a wide range of orientations, from transverse to longitudinal, confirming the observations of Nilssen (2010). The archaeological data display similar characteristics (Soulier, Morin, 2016). Thus, the reference collection proposed by A. B. Galán and M. Dominguez-Rodrigo (2013) in which the butcher was instructed to make oblique gestures for disarticulation and transverse gestures for defleshing is not representative of butchery activities conducted in reality. Moreover, our experiments show that numerous cutmarks attributed by Galán and Dominguez‑Rodrigo (2013: 1137, 1140) to disarticulation based on their oblique orientation were, based on their location, more likely produced by defleshing. What is more, the attribution of the activity performed on the sole basis of cutmark orientation is sometimes erroneous. Indeed, the oblique cutmarks interpreted as disarticulation traces on portions 2 and 3 of the ulna, 2 and 5 of the femur, and 2, 3, and 5 of the humerus in Galán and Dominguez-Rodrigo (2013) are far too distant from the points of articulation to be related to disarticulation (figure 125). Therefore, the butcher could not possibly have impacted the bones in these positions during disarticulation of the long bones; these cutmarks must be related to defleshing in spite of their obliquity.

Figure 125 - Illustration of the gestures relative to some cutmarks attributed by A.B. Galán and M. Dominguez‑Rodrigo (2013) to cutmarks from disarticulation

Figure 125 - Illustration of the gestures relative to some cutmarks attributed by A.B. Galán and M. Dominguez‑Rodrigo (2013) to cutmarks from disarticulation

The double arrows indicate the gestures associated with these cutmarks

CAD: M.-C. Soulier

b - Reevaluation of the interpretation of certain cutmarks documented in other reference collections

74As the number of half-carcasses disarticulated was relatively low (n=4), it is probable that the T&H reference collection has not documented the entire range of variability in cutmarks generated by disarticulation. Cutmarks from disarticulation are to be expected at the articular extremities. Nonetheless, as P. J. Nilssen (2000) has emphasized, these zones are also regularly impacted by cutmarks from defleshing. As also documented by other references (Binford, 1981; Nilssen, 2000), the zones adjacent to the articulations can also present cutmarks associated with disarticulation. The Hp-3 cutmarks on the neck of the humerus are considered by L.R. Binford (1981) to be related to disarticulation. However, cutmarks in this location are too far removed from the joint to facilitate scapulohumeral disarticulation (figure 126). In L. R. Binford’s reference (1981), these Hp-3 cutmarks are also very similar to the Hp-4 cutmarks that he assigns to defleshing. In accordance with P. J. Nilssen (2000), this type of cutmark was obtained in our experiments only during defleshing activities. Our reference collection and that of P. J. Nilssen both benefited from the rigorous recording of butchery activities, which allowed for the association of specific cutmarks to specific activities. L. R. Binford’s (1981) reference collection, however, was the product of the interpretation of cutmarks observed on archaeological assemblages on the basis of butchery activities observed amongst the Nunamiut (Binford, 1978). These circumstances lead us to reject Binford’s attribution of the Hp-3 cutmarks to disarticulation and to relate them exclusively to defleshing.

75In his reference, Nilssen (2000: 245) attributed the S-12 cutmarks on the neck of the scapula to disarticulation. Again, these cutmarks are situated too far from the scapulohumeral joint to be related to disarticulation (figure 126). Moreover, the separation of these joints is fairly easy, and does not require the kind of forceful gestures suggested by the numerous S-12 cutmarks (Nilssen, 2000: 188). These cutmarks appear to be related to defleshing.

Figure 126 - Location of the cutmarks related to codes S-12 and Hp-3 (white lines) interpreted by L.R. Binford (1981) as cutmarks from disarticulation

Figure 126 - Location of the cutmarks related to codes S-12 and Hp-3 (white lines) interpreted by L.R. Binford (1981) as cutmarks from disarticulation

The dotted line indicates the position of the joint and the arrows indicate the gestures associated with these cutmarks

Photograph and CAD: M.-C. Soulier

76A certain caution must thus be exercised in the interpretation of cutmarks situated at or near the articular extremities because a number of cutmarks interpreted as signs of disarticulation by Binford (1981) are ubiquitous (see figures 135-146). On this basis, the claim that disarticulation marks are frequently documented in European Palaeolithic faunal assemblages (Costamagno, David, 2009) should be re‑evaluated. For example, at Vogelherd, all of the cutmarks interpreted as signs of disarticulation (Niven, 2006) could just as well have been produced by defleshing. The same is true of the cutmarks present on the fragments of distal humerus found at Site BK (Upper Bed II) and Olduvai Gorge (Dominguez-Rodrigo et al., 2013). These marks are interpreted as the oldest evidence of secondary disarticulation of the elbow but, with regard to their orientation, the cutmarks on the medial surface of the trochlea (Hd-e, see figure 135) could just as well be interpreted as defleshing cutmarks.

77Based on the protocol followed in Nilssen’s reference, it is not possible to determine the activities (skinning/ defleshing) that generated the transverse cutmarks on the distal shaft of the radius and tibia (codes Rcs-2 and Td-5, Nilssen, 2000: 255, 257). In the T&H reference, no cutmarks were observed on carcass 2, on which the circular starting incision on the hide was made at the distal end of the tibia. Still, it seems entirely possible that the cutmarks identified by P. J. Nilssen (2000) could have been produced during the skinning of the distal section of the zeugopod because the skin is in direct contact with the bone in this region (figure 127). The cutmarks observed by P. J. Nilssen appear longer than those that we produced by defleshing and, inversely, very similar to those we produced in making the circular incisions on the metapodials and phalanges in the T&H reference collection, with the presence of a cluster of numerous sub‑parallel cutmarks. Sub‑transverse cutmarks present at the distal end of the zeugopod could therefore be the result of several different operations, but their length and general distribution could constitute diagnostic criteria for distinguishing defleshing cutmarks from skinning cutmarks.

Figure 127 - Red deer limbs showing the parts of the bones in direct contact with the skin (dotted areas)

Figure 127 - Red deer limbs showing the parts of the bones in direct contact with the skin (dotted areas)

Photographs and CAD: M.-C. Soulier

78In L. R. Binford’s reference, codes Mtd-4 and Mcd-4 are reported as deriving from defleshing activities. Even though no meat is present on the metapodials, cutmarks situated in this zone have been interpreted as defleshing marks in numerous archaeological assemblages: at Schöningen (Voormolen, 2008), at Flageolet (Deplano, 1994) and at Grotte du Lazaret (Valensi, 1991). In other cases, such as at Pataud (Sekhr, 1998), Moulin‑Neuf, Saint‑Germain‑la-Rivière (Costamagno, 1999), Voghelherd (Niven, 2006), Monruz (Müller, 2013) and Gönnesdorf (Street, Turner, 2013), these traces were not interpreted at all. According to the results of the PCR T&H experiments, these cutmarks correspond to oblique movements made with a sharp cutting-edge to detach the skin from the underlying tissue, which supports the previous hypothesis of J.-F. Bez (1995).

c - A new system for the notation of cutmarks

79Beyond the doubts raised about certain interpretations (see above), the re-examination of the references established by Binford (1981) and Nilssen (2000) revealed certain fundamental flaws these interpretive models. In the work of Binford (1981), inconsistencies were noted between the illustrations and descriptive table 4.04. The illustration of code Rcd-2 (Binford, 1981: 133) differs from its description on page 141. Mtd-1 is illustrated on two different pages, 107 and 121, and the example on page 107 is actually listed as Mtd-2 in table 4.04 (p. 140). On the other hand, the illustration labelled Mtd-1 on page 121 does not correspond to any of the descriptions in the table.

80The system of notation for cutmarks originally developed by Binford (1981) was later completed by Nilssen (2000). The latter established new codes for the same activity, which complicated its application. Furthermore, several of the new codes make reference to pre-existing codes in Binford’s reference. For example, Nilssen divided Binford’s code Hd-1 into three distinct codes (Nilssen, 2000: 191), all related to disarticulation, as was code Hd-1 initially. We also noted ambiguous usage of some of L. R. Binford’s codes. For example, the cutmarks Hp-1 correspond to “marks along the border of the ‘lip’ of the ball” (Binford, 1981: 140) but Nilssen also used this code to refer to longitudinal cutmarks related to defleshing (Nilssen, 2000: 190).

81In order to complete and clarify the reference established by Binford (1981), we propose a new system for the notation of cutmarks that combines the results of the T&H reference collection with those published by J.-D. Vigne (2005), L. R. Binford (1981), P. J. Nilssen (2000), and S. Costamagno and F. David (2009). For the reasons outlined in the introduction to this volume, the data published by Y. Abe (2005), A. B. Galán and M. Domíguez‑Rodrigo (2013), M. Padilla (2008), and J.-F. Bez (1995) have been excluded. The system of notation (Annex 4; figures 128-146) was simplified by grouping together the codes that refer to a single butchery activity while maintaining the distinctions between cutmarks that result from different activities. The cutmarks present on the diaphyses of meaty long bones are, according to current observations, systematically related to defleshing activities. In contrast, the cutmarks located at or near the articular extremities are the most equivocal. In order to evaluate our codes over the course of the experiments, we opted to employ a finely subdivided set of codes for these areas, noted from to x. The orientation of the cutmarks presents another crucial course of information on the processing of animal carcasses (Annex 4) and was also considered in our system of notation. For example, a longitudinal cutmark located on the femoral diaphyses is coded as Fs-a’ (“Fs” = femoral shaft, the “ ˊ ” indicates that the cutmark is longitudinal) while an oblique or transverse cutmark located on the proximal end of the metatarsal is coded as Mtp-a (“Mtp” = metatarsal, proximal). In certain cases, oblique and transverse cutmarks located in the same position can result, respectively, from different butchery activities. A supplementary subdivision was established for these cases. For example, on the medial surface of the proximal radius, oblique cutmarks related defleshing are coded Rp-f while transverse cutmarks resulting from disarticulation are coded Rp-fˊˊ (the ˊˊ signifying a transverse cutmark). On the templates, the ambiguous zones in terms of butcher activities are represented in light grey. When the orientations of the cutmarks allow for discrimination between activities, these zones are indicated in medium grey. Dark grey indicates zones of unequivocal interpretation. An interactive database that integrates the variables described here – with potential for the addition of new variables – is currently in development. This program, based on a touch-screen interface, will facilitate recording of the position and orientation of cutmarks and simplify their interpretation.

Figure 128 - Illustration of the new codes for cutmarks related to butchery on the mandible

Figure 128 - Illustration of the new codes for cutmarks related to butchery on the mandible

The dark grey zones correspond to the zones where only a single activity was documented; the very light grey zones correspond to zones where cutmarks could not be attributed to a particular activity
a Abbreviations: dir. = direction of the cutmark; T = transverse; O = oblique; L = longitudinal. c Abbreviations: Vest. = vestibular; Ling. = lingual. d The activities are taken from Binford (1981), Nilssen (2000) and the T&H experimental reference (see Annex 4 for details). Abbreviations: DP = skinning; DS = disarticulation; DC = defleshing. Caution, the axis considered is different between the vertical and the horizontal branch..

CAD: M.-C. Soulier and M. Coutureau

Figure 129 - Illustration of the new codes related to cutmarks from butchery on the atlas

Figure 129 - Illustration of the new codes related to cutmarks from butchery on the atlas

The dark grey zones correspond to the zones where only a single activity was documented; in the light grey zones, the orientation of the cutmark allowed identification of the corresponding activity, the very light grey zones correspond to ambiguous zones where cutmarks cannot be attributed to a particular activity
a Abbreviations: dir. = direction of the cutmark; T = transverse; O = oblique; L = longitudinal. c Abbreviations: Dors. = dorsal; Vent. = ventral; Cra. = cranial; Cau. = caudal. d The activities are taken from Binford (1981), Nilssen (2000) and from the T&H experimental reference (see Annex 4 for details). Abbreviations: EV = evisceration; DS = disarticulation; DC = defleshing, tât. = testing for joint

CAD: M.-C. Soulier and M. Coutureau

Figure 130 - Illustration of the new codes related to cutmarks from butchery on the axis

Figure 130 - Illustration of the new codes related to cutmarks from butchery on the axis

The dark grey zones correspond to the zones where only a single activity was documented; in the light grey zones, the orientation of the cutmark allowed identification of the corresponding activity, the very light grey zones correspond to ambiguous zones where cutmarks cannot be attributed to a particular activity
a Abbreviations: dir.= direction of the cutmark;T = transverse; ST = sub-transverse; O = oblique; L = longitudinal. c Abbreviations: Med. = medial; Lat. = lateral; Dors. = dorsal; Vent.= ventral. d The activities are taken from Binford (1981), Nilssen (2000) and the T&H experimental reference (see Annex 4 for details). Abbreviations: EV = evisceration; DS = disarticulation; DC = defleshing; tât. = testing for join

CAD: M.-C. Soulier and M. Coutureau

Figure 131 - Illustration of the new codes related to cutmarks from butchery on the other cervical vertebrae

Figure 131 - Illustration of the new codes related to cutmarks from butchery on the other cervical vertebrae

The dark grey zones correspond to the zones where only a single activity was documented; in the light grey zones, the orientation of the cutmark allowed identification of the corresponding activity
a Abbreviations: dir. = direction of the cutmark; T = transverse; ST = sub-transverse; O = oblique; L = longitudinal. c Abbreviations: Med. = medial; Lat. = lateral; Dors. = dorsal; Vent. = ventral; Cra. = cranial. d The activities are taken from Binford (1981), Nilssen (2000) and the T&H experimental reference (see Annex 4 for details). Abbreviations: EV = evisceration; DS = disarticulation; DC = defleshing; tât. = testing for join

CAD: M.-C. Soulier and M. Coutureau

Figure 132 - Illustration of the new codes related to cutmarks from butchery on the thoracic vertebrae

Figure 132 - Illustration of the new codes related to cutmarks from butchery on the thoracic vertebrae

The dark grey zones correspond to the zones where only a single activity was documented; in the light grey zones, the orientation of the cutmark allowed identification of the corresponding activity, the very light grey zones correspond to ambiguous zones where cutmarks cannot be attributed to a particular activity
a Abbreviations: dir. = direction of the cutmark; T = transverse; O = oblique; L = longitudinal. c Abbreviations: Med. = medial; Lat. =lateral; Dors. = dorsal; Vent. = ventral; Cra. = cranial; Cau. = caudal. d The activities are taken from Binford (1981), Nilssen (2000) and the T&H experimental reference (see Annex 4 for details). Abbreviations: EV = evisceration; DS = disarticulation; DC = defleshing

CAD: M.-C. Soulier and M. Coutureau

Figure 133 - Illustration of the new codes related to cutmarks from butchery on the lumbar vertebrae

Figure 133 - Illustration of the new codes related to cutmarks from butchery on the lumbar vertebrae

The dark grey zones correspond to the zones where only a single activity was documented; in the light grey zones, the orientation of the cutmark allowed identification of the corresponding activity
a Abbreviations: dir. = direction of the cutmark; T = transverse; O = oblique; L = longitudinal. c Abbreviations: Med. = medial; Lat. = lateral; Dors. = dorsal; Vent. = ventral; Cra. = cranial; Cau. = caudal. d The activities are taken from Binford (1981), Nilssen (2000) and the T&H experimental reference (see Annex 4 for details). Abbreviations: EV = evisceration; DS = disarticulation; DC = defleshing

CAD: M.-C. Soulier and M. Coutureau

Figure 134 - Illustration of the new codes related to cutmarks from butchery on the sacrum

Figure 134 - Illustration of the new codes related to cutmarks from butchery on the sacrum

The dark grey zones correspond to the zones where only a single activity was documented
a Abbreviations: dir. = direction of the cutmark; T = transverse; O = oblique; L = longitudinal. c Abbreviations: Med. = medial; Lat. = lateral; Dors. = dorsal; Vent. = ventral; Cra. = cranial. d The activities are taken from Binford (1981), Nilssen (2000) and the T&H experimental reference (see Annex 4 for details). Abbreviations: EV = evisceration; DS = disarticulation; DC = defleshing

CAD: M.-C. Soulier and M. Coutureau

Figure 135 - Illustration of the new codes related to cutmarks from butchery on the humerus

Figure 135 - Illustration of the new codes related to cutmarks from butchery on the humerus

The dark grey zones correspond to the zones where only a single activity was documented; in the light grey zones, the orientation of the cutmark allowed identification of the corresponding activity, the very light grey zones correspond to ambiguous zones where cutmarks cannot be attributed to a particular activity
a Abbreviations: dir.= direction of the cutmark;T = transverse; O = oblique; L = longitudinal; ST = sub-transverse; SL = sub-longitudinal. b The long bones were subdivided into six sections. c Abbreviations: M = medial; A = anterior; P = posterior; L = lateral. d The activities are taken from Binford (1981), Nilssen (2000), Vigne (2005), Costamagno, David (2009) and the T&H experimental reference (see Annex 4 for details). Abbreviations: DP = skinning; DS = disarticulation; DC = defleshing

CAD: M.-C. Soulier and M. Coutureau

Figure 136 - Illustration of the new codes related to cutmarks from butchery on the radius

Figure 136 - Illustration of the new codes related to cutmarks from butchery on the radius

The dark grey zones correspond to the zones where only a single activity was documented; in the light grey zones, the orientation of the cutmark allowed identification of the corresponding activity, the very light grey zones correspond to ambiguous zones where cutmarks cannot be attributed to a particular activity
a Abbreviations: dir.= direction of the cutmark;T.= transverse; O.= oblique; L = longitudinal; ST = sub-transverse; SL = sub-longitudinal. b The long bones were subdivided into six sections. c Abbreviations: M = medial; A = anterior; P = posterior; L = lateral. d The activities are taken from Binford (1981), Bez (1995), Nilssen (2000), Costamagno, David (2009) and the T&H experimental reference (see Annex 4 for details). Abbreviations: DP = skinning; DS = disarticulation; DC = defleshing; TN = removal of the tendons. XX+XX indicates that both activities are documented; XX(+XX?) indicates that the first activity mentioned is attested but that the second, in parentheses, is uncertain; XX?/XX? indicates that the protocol used does not allow discrimination between both activities

CAD: M.-C. Soulier and M. Coutureau

Figure 137 - Illustration of the new codes related to cutmarks from butchery on the ulna

Figure 137 - Illustration of the new codes related to cutmarks from butchery on the ulna

The dark grey zones correspond to the zones where only a single activity was documented; in the light grey zones, the orientation of the cutmark allowed identification of the corresponding activity, the very light grey zones correspond to ambiguous zones where cutmarks cannot be attributed to a particular activity
a Abbreviations: dir.= direction of the cutmark;T.= transverse; O.= oblique; L = longitudinal; ST = sub-transverse; SL = sub-longitudinal. b The long bones were subdivided into six sections. c Abbreviations: M = medial; A = anterior; P = posterior; L = lateral; Palm. = palmar. d The activities are taken from Binford (1981), Bez (1995), Nilssen (2000), Costamagno, David (2009) and the T&H experimental reference (see Annex 4 for details). Abbreviations: DP = skinning; DS = disarticulation; DC = defleshing

CAD: M.-C. Soulier and M. Coutureau

Figure 138 - Illustration of the new codes related to cutmarks from butchery on the carpal bones

Figure 138 - Illustration of the new codes related to cutmarks from butchery on the carpal bones

The dark grey zones correspond to the zones where only a single activity was documented; in the light grey zones, the orientation of the cutmark allowed identification of the corresponding activity
a Abbreviations: dir. = direction of the cutmark; T = transverse; O = oblique; L = longitudinal. c Abbreviations: M = medial; A = anterior; P = posterior; L = lateral; Palm. = palmar. d The activities are taken from Binford (1981), Bez (1995), Nilssen (2000), Costamagno, David (2009) and the T&H experimental reference (see annex 4 for details). Abbreviations: DP = skinning; DS = disarticulation; DC = defleshing; TN = removal of the tendons, ANT = anterior; POST = posterior; RAD = radius; CAR1 = proximal row of carpals; CAR2 = distal row of carpals

CAD: M.-C. Soulier and M. Coutureau

Figure 139 - Illustration of the new codes related to cutmarks from butchery on the metacarpal

Figure 139 - Illustration of the new codes related to cutmarks from butchery on the metacarpal

The dark grey zones correspond to the zones where only a single activity was documented; in the light grey zones, the orientation of the cutmark allowed identification of the corresponding activity, the very light grey zones correspond to ambiguous zones where cutmarks cannot be attributed to a particular activity
a Abbreviations: dir. = direction of the cutmark; T = transverse; O = oblique; L = longitudinal. b The long bones were subdivided into six sections. c Abbreviations: M = medial; A = anterior; P = posterior; L = lateral; Palm. = palmar. d The activities are taken from Binford (1981), Bez (1995), Nilssen (2000), Costamagno, David (2009) and the T&H experimental reference (see Annex 4 for details). Abbreviations: DP = skinning; DS = disarticulation; DC = defleshing; TN = removal of the tendons, ANT = anterior; POST = posterior

CAD: M.-C. Soulier and M. Coutureau

Figure 140 - Illustration of the new codes related to cutmarks from butchery on the scapula

Figure 140 - Illustration of the new codes related to cutmarks from butchery on the scapula

The dark grey zones correspond to the zones where only a single activity was documented; in the light grey zones, the orientation of the cutmark allowed identification of the corresponding activity ().
a Abbreviations: dir. = direction of the cutmark; T = transverse; O = oblique; L = longitudinal; ST = sub-transverse. c Abbreviations: M.= medial; A = anterior; P = posterior; L.= lateral. d The activities are taken from Binford (1981), Bez (1995), Nillsen (2000), Costamagno, David (2009) and the T&H experimental reference (see Annex 4 for details). Abbreviations: DS = disarticulation; DC = defleshing

CAD: M.-C. Soulier and M. Coutureau

Figure 141 - Illustration of the new codes related to cutmarks from butchery of the femur

Figure 141 - Illustration of the new codes related to cutmarks from butchery of the femur

The dark grey zones correspond to the zones where only a single activity was documented; in the light grey zones, the orientation of the cutmark allowed identification of the corresponding activity, the very light grey zones correspond to ambiguous zones where cutmarks cannot be attributed to a particular activity
a Abbreviations: dir. = direction of the cutmark; T = transverse; O = oblique; L = longitudinal. b The long bones were subdivided into six sections. c Abbreviations: M = medial; A = anterior; P = posterior; L = lateral; Palm. = palmar. d The activities are taken from Binford (1981), Nillsen (2000), Vigne (2005), Costamagno, David (2009) and the T&H experimental reference (see annex 4 for details). Abbreviations: DP = skinning; DS = disarticulation; DC = defleshing

CAD: M.-C. Soulier and M. Coutureau

Figure 142 - Illustration of the new codes related to cutmarks from butchery of the tibia

Figure 142 - Illustration of the new codes related to cutmarks from butchery of the tibia

The dark grey zones correspond to the zones where only a single activity was documented; in the light grey zones, the orientation of the cutmark allowed identification of the corresponding activity, the very light grey zones correspond to ambiguous zones where cutmarks cannot be attributed to a particular activity
a Abbreviations: dir. = direction of the cutmark; T = transverse; O = oblique; L = longitudinal. b The long bones were subdivided into six sections. c Abbreviations: M = medial; A = anterior; P = posterior; L = lateral; Palm. = palmar; Cra. = cranial. d The activities are taken from Binford (1981), Nilssen (2000), Vigne (2005), Costamagno, David (2009) and the T&H experimental reference (see Annex 4 for details). Abbreviations: DP = skinning; DS = disarticulation; DC = defleshing

CAD: M.-C. Soulier and M. Coutureau

Figure 143 - Illustration of the new codes related to cutmarks from butchery of the fibula, patella and malleolus

Figure 143 - Illustration of the new codes related to cutmarks from butchery of the fibula, patella and malleolus

The dark grey zones correspond to the zones where only a single activity was documented
a Abbreviations: dir. = direction of the cutmark;T = transverse; O = oblique. b Abbreviations: Prox. = proximal. c Abbreviations: A = anterior; L = lateral. d The activities are taken from Binford (1981), Nilssen (2000), Vigne (2005), Costamagno, David (2009) and the T&H experimental reference (see annex 4 for details). Abbreviations: DS = disarticulation; DC = defleshing

CAD: M.-C. Soulier and M. Coutureau

Figure 144 - Illustration of the new codes related to cutmarks from butchery on the tarsal bones

Figure 144 - Illustration of the new codes related to cutmarks from butchery on the tarsal bones

The dark grey zones correspond to the zones where only a single activity was documented
a Abbreviations: dir. = direction of the cutmark; T = transverse; O = oblique; L = longitudinal. c Abbreviations: M = medial; A = anterior; P = posterior; L = lateral. d The activities are taken from Binford (1981), Bez (1995), Nilssen (2000), Costamagno, David (2009) and the T&H experimental reference (see Annex 4 for details). Abbreviations: DP = skinning; DS = disarticulation; TN = removal of the tendons, ANT = anterior; POST = posterior; TIB = tibia; TAR1 = proximal tow of tarsals; TAR2 = distal row of tarsals; MTM = metatarsal; MET = metapodial; PH1 = first phalanx; PH2 = second phalanx; exten. = extension; flex. = flexion. XX+XX indicates that both activities are documented; XX(+XX?) indicates that the first activity mentioned is attested but that the second, in parentheses, is uncertain

CAD: M.-C. Soulier and M. Coutureau

Figure 145 - Illustration of the new codes related to cutmarks from butchery of the metatarsal

Figure 145 - Illustration of the new codes related to cutmarks from butchery of the metatarsal

The dark grey zones correspond to the zones where only a single activity was documented; in the light grey zones, the orientation of the cutmark allowed identification of the corresponding activity
a Abbreviations: dir.= direction of the cutmark;T = transverse; O = oblique; L = longitudinal; ST = sub-transverse. b The long bones were subdivided into six sections. c Abbreviations: M = medial; A = anterior; P = posterior; L = lateral; Palm.= palmar. d The activities are taken from Binford (1981), Bez (1995), Nilssen (2000), Costamagno, David (2009) and the T&H experimental reference (see Annex 4 for details). Abbreviations: DP = skinning; DS = disarticulation; TN = removal of the tendons, ANT = anterior; POST = posterior; TIB = tibia; TAR1 = proximal row of tarsals; TAR2 = distal row of tarsals; MTM = metatarsal; MET = metapodial; PH1 = first phalanx; PH2 = second phalanx; exten. = extension; flex. = flexion

CAD: M.-C. Soulier and M. Coutureau

Figure 146 - Illustration of the new codes related to cutmarks from butchery on the phalanges and sesamoids

Figure 146 - Illustration of the new codes related to cutmarks from butchery on the phalanges and sesamoids

The dark grey zones correspond to the zones where only a single activity was documented; in the light grey zones, the orientation of the cutmark allowed identification of the corresponding activity, the very light grey zones correspond to ambiguous zones where cutmarks cannot be attributed to a particular activity
a Abbreviations: dir.= direction of the cutmark;T = transverse; O = oblique; L = longitudinal; ST = sub-transverse, SL = sub-longitudinal. c Abbreviations: M = medial; A = anterior; P = posterior; L = lateral; Cra. = cranial; Ab = abaxial. d The activities are taken from Binford (1981), Bez (1995), Nilssen (2000), Costamagno, David (2009) and the T&H experimental reference (see Annex 4 for details). Abbreviations: DP = skinning; DS = disarticulation; TN = removal of the tendons, ANT = anterior; POST = posterior; TIB = tibia; TAR1 = proximal row of tarsals; TAR2 = distal row of tarsals; MTM = metatarsal; MET = metapodial; PH1 = first phalanx; PH2 = second phalanx

CAD: M.-C. Soulier and M. Coutureau

d - Implications of the new system of notation for improved interpretation of modes of prey exploitation and the culinary and domestic strategies of human groups

82During our experiments, the muscles were each removed individually. P. J. Nilssen’s (2000) reference is based on butchery for the production of dried meat, which requires the removal of muscles in as large a mass as possible and is generally performed on entire limbs (Binford, 1978; Odgaard, 2007; Pasda, Odgaard, 2011). Only 5.6 % (171/3050) of the defleshing cutmarks produced during our experiments are longitudinal in orientation, which is far below the values reported by P. J. Nilssen (see the discussion in Soulier, Morin, 2016). Thus, it is clear that “classic” defleshing and butchery for drying differ in terms of the percentage of longitudinal cutmarks. The values obtained in the T&H experiments are close to those documented on the majority of the faunal assemblages from the Upper Palaeolithic of southwestern France (Soulier, Morin, 2016). The T&H reference collection also shows that the cutmarks produced during “classic butchery” have a different distribution than those documented for defleshing that follows boiling. As a matter of fact, the cutmarks from “classic” butchery are distributed along the entire diaphyses of the long bones, while the cutmarks on previously boiled bones are concentrated at the ends of the diaphyses (see Abe, 2005: Annexes). The orientation and location of the cutmarks are therefore important factors to consider in the interpretation of cutmarks on meaty long bones because they can reveal different approaches to processing carcasses and preparing meats.

83The experiments conducted as part of the PCR T&H show that the cutmarks related to disarticulation can also be longitudinal. The orientation of the cutmarks depends on the position in which the leg is held during disarticulation (i.e. flexed / extended). This factor can also influence the location of the cutmarks (figure 147). Indeed, it is unlikely that cutmarks would form above the olecranon fossa and the upper half of the olecranon of the ulna when the leg is held in flexion. In this case, the cutmarks are more likely to appear on the anconeal process of the ulna (Up-g). At the trochlea, longitudinal cutmarks can appear when the leg is held in flexion (Hd-dˊ), while they will be transverse if the leg is held in extension (Hd-dˊˊ). Transverse cutmarks on the edge of the trochlea are documented in faunal assemblages attributed to the Aurignacian at La Quina aval and Roc-de-Combe (Soulier, 2013), and in the Magdalenian at Saint-Germain-la-Rivière (layers 3-4: Costamagno, 1999), which indicates that the leg was held in extension during disarticulation. This differs from the evidence from Lazaret (Valensi, 1991), which indicates that the leg was held in flexion. On the ones of the tarsus, the zones Tal-a, Tal-e, Cal-a et Cal-b can only come into contact with the tool if the leg is held in flexion. Cutmarks on the zones Tal-c and Cal-e show different orientations depending on whether the leg is held in flexion or extension. For example, the oblique cutmarks observed on the Tal-c zone at Gesher Benot Ya’aqov (Rabinovich et al., 2008) indicate disarticulation by flexion, while transverse cutmarks in the same location, such as those observed at Lundby Mose (Leduc, 2010) indicate disarticulation performed while the leg is held in extension.

Figure 147 - Location of disarticulation cutmarks when the elbow is held in flexion or in extension

Figure 147 - Location of disarticulation cutmarks when the elbow is held in flexion or in extension

CAD: M.-C. Soulier

84The hide can be exploited to various ends (bedding and blankets, clothing, shelter, …) and the intended ends influence the approaches taken to skinning (ex: Manning, 1944; Binford, 1978; Grønnow et al., 1983). More accurate characterizations of the traces produced during this activity are much needed because, in spite of their informative potential, they are rarely discussed in any detail in the references currently available. More important still, it has been impossible to distinguish between cutmarks produced by skinning and those produced by the removal of tendons. Thus, at numerous sites, there has been little – or no – discussion of skinning activities. The study conducted by L. Niven at Vogelherd is an excellent example. Even though the illustrations of the cutmarks (Niven, 2006: 142-144) clearly show three steps in the skinning process – transverse cutmarks related to the circular starting incision, longitudinal cutmarks resulting from the longitudinal incisions, and oblique cutmarks produced during the detachment of the skin – she concludes that no cutmarks indicate this activity at the site. As a matter of fact, none of the cutmarks present have been described in the available literature. The consideration of the location of transverse cutmarks and longitudinal cutmarks is also of foundational importance because these data provide insight into the intentions underlying the removal of the hide. The longitudinal incision connects the circular incision to the incision made for evisceration. If this incision is made on the lateral surface of the limb, it is impossible to obtain a large hide in a single piece (Costamagno, 2012). At Les Abeilles, the longitudinal incision was made on the medial surface of the limb, which is compatible with the removal of large hides (Soulier, 2013). In contrast, the data from Cuzol‑de-Vers (Castel, 2003), Saint-Germain-la-Rivière (Costamagno, 2012), Isturitz (Soulier, 2013), Vogelherd (Niven, 2006) show that the longitudinal incisions were made on the lateral surface. The presence of glancing cutmarks under the circular incision indicates a related removal of the skin of the lower limbs (Soulier, Costamagno, 2017), bits of skin that are particularly strong, water-resistant, and insulating and, as a result, prized by the modern and ethnohistorical groups in cold climates (Seeman, 1933; Binford, 1981; Russel, 1995; Stefansson, Palsson, 2001; Abe, 2005). In reporting cutmarks, it is therefore particularly important to be as accurate as possible in reconstructing fragmented metapodials, and to omit cutmarks on fragments that could not be re-situated anatomically.

85The use of tendons is widely documented in ethnographic contexts, for such purposes as: making cordage, ligatures, or glue (Ekblaw, 1928; Malaurie, 1989; Russel, 1995; Gowdy, 1998), consumption as a source of protein (Stefansson, 1913; Costamagno, David, 2009). Prior to the current study, only one approach to tendon removal had been examined experimentally: the extraction of the tendon present at the distal end of the radius (Vigne, 2005). The process for extracting the thick tendons of the lower limbs remained undocumented and, as a result, is little mentioned in archaeozoological studies, though it has been suggested (Castel, 1999; Costamagno, 1999; Morel, Müller, 1997; Müller, 2013). According to the data from the T&H study sites, numerous cutmarks observed on metapodials from several sites can be attributed to the extraction of tendons and/or the interosseous medius (for example, Valensi, 1991: 819; Morel, Müller, 1997: 63; Cho, 1998: 293, 298; Niven, 2006: 144; Street, Turner, 2013: 103). W. Müller (2013) emphasizes that the tendons on the anterior surfaces of horse metapodials can be removed without leaving cutmarks on the bones because they lie directly on the surface of the bone. On the posterior surface, the flexor digitorum tendons can also be extracted without contact with the bone, as the latter is protected by the interosseous medius muscle. Thus, the longitudinal cutmarks observed on the horse metapodials from Monruz (Müller, 2013: 131, fig. 125) and Isturitz (Soulier, 2013) correspond to the removal of the tendinous muscle. Under the T&H PCR, cutmarks were observed inside the central troughs of the metapodials because, in artio‑dactyla and especially in cervids, the tendons are deepy embedded in these grooves (figure 124). Their removal therefore requires contact between the tools and the bone that creates characteristic cutmarks that can vary considerably depending on the gesture(s) employed. If the cutting-edge of the tool is used transversely with a oblique grazing motion to progressively detach the tendon, the cutmarks produced will be transverse or slightly oblique and short, forming along either side of the central trough (Mcs-f and Mts-f, figure 123). If the cutting-‑edge of the tool is inserted into the central trough behind the tendon, longitudinal cutmarks will form along the interior of the trough (Mcs-fˊ illkuand Mts-fˊ). The first of these methods appears to have been widespread in the Palaeolithic, as evidenced in the available reports on cutmarks at Combe-Saunière and Cuzoul-de-Vers (Castel, 1999), St-Germain-la-Rivière and Rond-du-Barry (Costamagno, 1999), Bois-Ragot, Troubat, and Rhodes II (Chevallier, 2015), and La Quina aval (Soulier, 2013). Some faunal assemblages do demonstrate the concomitant use of both techniques (Pataud: Cho, 1998; Vogelherd: Niven, 2006; Roc-de-Combe: Soulier, 2013; Roc-de-Marsal: Castel et al., 2017).

86The cutmarks resulting from trial-and-error in locating the joints that were produced on the T&H experimental sets were due to the inexperience of the experimenters with butchery. Theoretically, the identification of such marks on Palaeolithic fauna might indicate the performance of butchery tasks by inexperienced individuals, perhaps in training. However, even compared to the lithic evidence related to apprenticeship (Pigeot, 1987; Klaric, 2003; Roux, Bril, 2005; Simonet, 2008), the signatures in faunal assemblages are weak, and can easily be lost amongst or confounded with cutmarks resulting from defleshing, once all sequences of the butchery process have been completed. The only such marks that were found to be unambiguous are those that formed on the vertebrae (figure 119).

4 - Analyzing the recurrence of butchery cutmarks: the contribution of Geographic Information Systems

S. Chong [†], S. Costamagno, M.-C. Soulier

N.B.: This contribution is a modified version of the article Sebastian Chong should have finalized if he had had enough time. He invested a lot into this work since his Master 2 in 2011 and until December 2016 when he was seriously ill. Unfortunately he could not see it to the end. Despite all, the results obtained by Sebastian are revealing of the undeniable contribution of GIS to the questions related to butchery. The members of the Research Program are dedicating this chapter to his memory.

87As cutmarks are epiphenomena (Lyman, 1992), it is especially important to know which activity (e.g. the removal of the anterior limb extensor tendon, the disarticulation of the elbow, the defleshing of a bone) or which gesture is likely to leave, frequently or not, butchery cutmarks on the bones. Indeed, knowing the degree of incidence of the various steps of the chaîne opératoire on a given skeletal element or a skeletal part is essential in order to offer a gradation scale in the recognition of the various types of activities. In addition to the digitization of the cutmarks as graphic plates on Adobe Illustrator®, we wished to test the use of other referencing tools. Indeed, as drawing softwares do not generate any data, the quantitative study of the cutmarks is limited by fastidious manual counting – difficult to manage when the corpus is significant – or by complex manipulations to be able to use them with georeferencing tools. With over 6 000 cutmarks recorded on the anterior and posterior limbs, the PCR experimental corpus required the use of a tool that would be capable of recording these data, managing them, and treating them with a high degree of reliability and automation. In this view, Geographic Information Systems (GIS) appeared as perfectly adapted to our needs. Indeed, by considering anatomical charts as map bases, the x and y coordinates were attributed to each cutmark to which all the information from butchery parameters were associated. The advantage of the GIS is to allow questioning any of these parameters in order to obtain both a visually interpretable graphics rendering and a quantitatively usable set of data (Chong, 2011).

A - Study material

88All the cutmarks present on the bones of the anterior and posterior limbs (except for the coxal bone) of the nine deer carcasses were integrated into the GIS.

B - Method of study

89The GIS used is Quantum GIS (QGIS), 2.4 version. Free and open source, including its extensions, it can be set up in the usual operating systems and even for Android. In this sense, it is more interesting than ArcGIS usually used in zooarchaeology (Marean et al., 2001; Abe et al., 2002; Hodgkins et al., 2016) without performing less for the required use. The use of the GIS is done according to three main steps: importing the available data, inputting new data, then using them.

a - Cartographic backgrounds and georeferencing of data

90The cartographic backgrounds are reusing the anatomical charts used with Adobe Illustrator® as well as the anatomic parts retained for the qualitative study of the cutmarks (figure 148). In order to visualize them in QGIS, the cutmarks are associated to a projection, that is to say a coordinate system in space that can correspond, in other conditions of use than ours, to latitudes and longitudes. As the anatomic charts do not represent a surface of the terrestrial globe, a “non-terrestrial” projection was used. Chosen arbitrarily, it is limited to the dimensions of an A4 sheet. It corresponds in concrete terms to an orthonormal point of reference with the origin located at the bottom and on the left of the sheet. Every object present on the chart is thus given x and y coordinates.

Figure 148 - Anatomical portions considered for the scapula and long bones

Figure 148 - Anatomical portions considered for the scapula and long bones

CAD: M.-C. Soulier

91The plans of cutmarks done with Adobe Illustrator® are used to create analogous cutmarks in QGIS. Then they have the advantage of being coordinated on the map. After choosing the non-terrestrial projection mode, all the graphic objects present on screen are automatically coordinated. Then, QGIS allocates each cutmark to a line with a different identifier in a table. There are as many lines as cutmarks.

b - Data base

92Parallel to QGIS, an Excel data base was created. It is completed by simple cutting and pasting from QGIS. For the cutmarks, the two basic entries are the identifier and the coordinates of both extremities, the cutmarks being assimilated to straight lines. The other fields indicate the date of the butchery, the carcass code, the name of the experimenter, the type of tool used and the concerned animal.

93From the coordinates, some simple calculations can be made, such as the length of the cutmarks or their angle with regard to the elongation axis of the bone. For comparative purposes with the other available data, we have used the categories proposed by M.-C. Soulier and E. Morin (2016) and adapted the bone backgrounds to the same overall dimensions and orientations. As for orientation, three classes were retained: longitudinal cutmarks (0°-15° and 165°-180°), oblique cutmarks (15°-75° and 15°-165°), and transverse cutmarks (75°-105°) (figure 149). The length classes – long, intermediate, short – rely on the consideration of lengths related to the length of the bone (see Soulier, Morin, 2016 for discussion on this topic) in a logarithmic base 10. Short cutmarks correspond to values under -1.18, intermediate ones to values between -1.18 and 0.14 and long ones to values over 0.14.

Figure 149 - Categories applied to the orientation of the cutmarks

Figure 149 - Categories applied to the orientation of the cutmarks

Modified from Soulier, Morin, 2016

94In QGIS, the input can be manual, cutmark by cutmark or by group, by selecting several cutmarks with the same value. This selection is done from the cartographic background with the help of the mouse or from a dialog box by selecting criteria (selection of all the cutmarks of the same year for example). The input can also be automated by doing a “join”, a term that designates the action of overlapping two pieces of information. It allows, for example, associating the cutmarks found on the humerus with the information “humerus” (figure 150). Thus, in the data base, for each cutmark we have the anatomic element, the face of the bone and the part of the bone (figure 148). The join principle was also described by C. W. Marean et al. (2001) for the quantification of skeletal elements.

Figure 150 - Example of relationships applied to the data

Figure 150 - Example of relationships applied to the data

a: data related to the cutmark; b: data related to the section of the bone; c: join between the cutmark and the section of the bone

CAD: S. Chong

c - Query to consult the data base

95The query is the function that allows visualizing the cutmarks according to the chosen parameters that one wants to examine. The style of the cutmarks can be modified for a better visualization. The use of the GIS starts with the query. Based on the SQL language (Structured Query Language), the query is easily done through a dialog box by configuring conditions. For example, if one wants to visualize and manage the transverse cutmarks produced in 2010 by disarticulation, the expression to enter would be: “orientation”= “transverse” AND “year”=2010 AND “activity” = “disarticulation”.

d - Using the data base with space criteria

96Studying the location of the produced cutmarks is done with regard to a considered surface, which requires dividing up the skeleton according to the anatomical elements, face, portion, square grid (figure 151). The chosen division depends on the required precision and on the parameter that one wants to test.

97For particular cases such as the recurrence of cutmarks, it is necessary to extrapolate the cutmark (line) to a surface (polygon) with the help of a stamp that corresponds to the outline of the cutmark at a 1 cm distance. Thus the overlap of the polygons can be counted even if the cutmarks are not intersecting (figure 152).

C - Results

a - Skinning cutmarks

98Skinning cutmarks are only visible on parts devoid of flesh or tendinous masses. The metatarsal is the skeletal element presenting the strongest occurrence of this type of cutmark: cutmarks related to this butchery operation were noted in 16 cases out of 17 (table 23). Then comes the metacarpal (9/18) and phalanxes (4/8). In rare cases (3/18), tibias and tarsals can be affected.

99The bone/ tool contacts – and thus the legibility of butchery cutmarks – are greatly depending on the gestures carried out. Cutmarks connected to skin removal are the ones that are most frequently found on metapodials (13/18 for the metatarsal and 8/18 for the metacarpal). On the tibia and radio-ulna such cutmarks could potentially have been created but it is impossible to distinguish between them and defleshing cutmarks. Protected by ligament and tendinous masses, the short tarsal and carpal bones and the phalanxes do not bear any cutmarks related to these gestures (table 23).

Figure 151 - Placement of the cutmarks according to a: the anatomical element; b: the section; c: the grid overlay

Figure 151 - Placement of the cutmarks according to a: the anatomical element; b: the section; c: the grid overlay

CAD: S.Chong

Figure 152 - Example of the use of GIS in the analysis of the distribution of cutmarks from butchery in a given location

Figure 152 - Example of the use of GIS in the analysis of the distribution of cutmarks from butchery in a given location

a: cutmarks formed during three distinct butchery operations; b: creation of a buffer zone of 1 cm; c: zones of overlap; d: visualization of cutmark occurrences using isolines

CAD: S. Chong

100The longitudinal incision was carried out mostly on the medial face. Longitudinal cutmarks connected with this gesture are far from forming systematically since the metatarsal, which is the skeletal element with the strongest occurrence, is only affected by this type of cutmark in 10 cases out of 16, that is to say 62 % and the metacarpal in 6 cases out of 15, that is to say in 40 % of cases. This type of cutmark, although with significantly lower occurrences, was also observed on the distal portion of the tibia and the cubo‑navicular. In cases where this incision was made on another face (table 23), there are few cutmarks documenting these actions. For the anterior and posterior faces, the presence of the tendons can easily explain the absence of cutmark. On the other hand, on the lateral face of the metapodials, it should be noted that the only incision done on this face left characteristic cutmarks. In fact, cutmarks related to longitudinal incisions done on the lateral face are frequently observed in the archaeological context (Castel, 1999; Costamagno, 1999; Niven, 2006; Leduc, 2010; Soulier, 2013; Castel et al., 2017).

Table 23 - Occurrence of cutmarks from skinning on bones according to the gestures employed

Skeletal part

Radius

Tibia

Carpals

Tarsals

Metacarpal

Metatarsal

Phalanx 1/2

Number of skeletal part skinned

N

18

18

18

18

18

18

8

Skinning cutmarks

N with cutmarks

0

3

0

3

9

18

4

Skin removal cutmarks

N with cutmarks

-

-

-

-

8

13

-

Longitudinal incisions

medial face

N with cutmarks

0

3

0

3

6

10

0

N

15

16

15

16

15

16

4

lateral face

N with cutmarks

0

-

0

-

1

-

-

N

1

-

1

-

1

-

-

anterior face

N with cutmarks

0

0

0

0

0

0

0

N

1

1

1

1

1

2

1

posterior face

N with cutmarks

0

0

0

0

0

-

0

N

2

1

1

1

1

-

3

Circular incisions

tibia distal shaft

N with cutmarks

-

0

-

-

-

-

-

N

-

1

-

-

-

-

-

metapodial middle shaft

N with cutmarks

-

-

-

-

4

2

-

N

-

-

-

-

8

3

-

metapodial distal shaft

N with cutmarks

-

-

-

-

2

7

-

N

-

-

-

-

4

7

-

metapodial distal epiphysis

N with cutmarks

-

-

-

-

1

1

-

N

-

-

-

-

2

2

-

phalanx 1 or phalanx 2

N with cutmarks

-

-

-

-

-

-

3

N

-

-

-

-

-

-

5

near hoof

N with cutmarks

-

-

-

-

-

-

0

 

N

-

-

-

-

-

-

3

Abbreviations: N: number of times the gesture was performed, N with cutmarks: number of times the gesture produced a distinctive cutmark

101Cutmarks related to circular incisions of the skin are present in more than three-quarters of cases on metatarsals (10/12) (table 23). Less recurring on the phalanxes (3/5) and the metacarpals (7/14), they were never observed during the circular incision of the skin as closely as possible to the hooves.

102In addition to the gestures, the tool used for cutting seems to have an impact on the formation or not of the skinning cutmarks. Unretouched flakes are the only tools that have left traces connected to the removal of the skin in less than one in three cases. The other tools generated traces at least in one of two cases (table 24). For the longitudinal incisions done in medial or lateral faces, the observation is identical. On the other hand, in the case of circular incisions, unretouched flakes / points most often leave marks (9/16), followed by denticulates (5/9). Flake cleavers only left traces in 1 case out of 5 (table 24). For raw materials, it is more difficult to decide, flint may leave traces more frequently than quartzite (table 25).

Table 24 - Occurrence of cutmarks from skinning according to the tool used (remov. = removal, inc. = incisions, Long. = longitudinal, Circ. = circular)

 

Metacarpal

Metatarsal

Phalanx

N. side

 

Skin remov.

Long. inc.

Circ. ins.

Skin remov.

Long. inc.

Circ. ins.

Circ. ins.

Denticulate

4

3

3

5

3

2

6

Mousterian point

1

1

1

1

Unretouched point

1

1

2

1

1

Flake cleaver

1

1

3

2

1

3

Unretouched flake

1

1

1

3

3

3

3

7

Table 25 - Occurrence of cutmarks from skinning according to the raw material used (remov. = removal, inc. = incisions, Long. = longitudinal, Circ. = circular)

 

Metcarpal

Metatarsal

Phalanx

N side

 

Skin remov.

Long. inc.

Circ. ins.

Skin remov.

Long. inc.

Circ. ins.

Circ. ins.

Ophite

1

1

1

1

-

1

Quartzite

3

3

3

7

3

3

2

10

Schist

1

1

1

Flint

4

3

3

4

2

5

-

6

b - Disarticulation cutmarks

103With the exception of the pelvic femoral joint, which was systematically disarticulated, limb bones were segmented only on four half-carcasses. The metapodials and first phalanxes were separated on eight different limbs. Given the small size of the sample, it is therefore particularly difficult to discuss the frequency of appearance of the cutmarks and their degree of recurrence. There is no evidence to document shoulder disarticulation (table 26). On the other hand, the disarticulation of the elbow is systematically recorded on the humerus and the radius: the lateral face of the humerus systematically bears cutmarks (code Hd-h) whereas, on the radius, the cutmarks are less recurrent (table 26). At the wrist, the carpals systematically bear disarticulation cutmarks, which is not the case with the radius. Some traces found on the carpals are not located at the expected locations and show failed attempts at disarticulation. At the knee, the only evidence of disarticulation observed is found on the distal end of the femur at various locations (table 26). For the ankle, traces were exclusively recorded on tarsals (table 26). The calcaneus and talus bones systematically have disarticulation cutmarks at sometimes recurrent locations (Cal-a, Cal-e, Tal-a). Finally, in six out of eight cases, cutmarks document the disarticulation between the metapodials and the first phalanxes: they can be present on both bones or only on one (table 26).

Table 26 - Occurrence of cutmarks from disarticulation according to their location (see Annex 4 for cutmark codes)

Articulation

Skeletal element

Nb disarticulated

Nb with cutmarks

Occurrence according to location

Shoulder

 

 

 

 

 

Scapula

4

0

 

 

Humerus

4

0

 

Elbow

 

 

 

 

 

Humerus

4

4

Hd-a: 2; Hd-d: 4; Hd-f: 1; Hd-g: 2; Hd-h: 4

 

Radius

4

4

Rp-a: 2; Rp-e: 2; Rp-f: 1

 

Ulna

4

2

Up-c: 1

Wrist

 

 

 

 

 

Radius

2

1

Rd-d: 1; Ud-a: 1; Ud-g: 1

 

Carpals first row

4

4

Pyr-a: 3; Pyr-c: 2; Pyr-a: 3; lun-a: 1; lun-c: 1; lun-b: 2;
sca-a: 4; sca-c: 1; sca-b: 1

 

Carpals second row

2

2

unc-a: 1; unc-b: 1; ctt-a: 2; ctt-b: 1

Knee

 

 

 

 

 

Femur

4

3

Fd-a: 1; Fd-c: 1; Fd-d: 1; Fd-e: 1; Fd-g: 2

 

Tibia

4

0

 

Ankle

 

 

 

 

 

Tibia

4

0

 

 

Malleolus

4

2

 

 

Calcaneus

4

4

cal-a: 4; cal-b: 1; cal-c: 1; cal-e: 4; cal-g: 2; cal-i: 2

 

Talus

4

4

tal-a: 4; tal-b: 1; Tal-c: 3; Tal-e: 2

 

Cubonavicular

4

2

cbn-a: 2

 

Large cuneiform

4

1

gcf-a: 1

Metapodial / phalanx

 

 

 

 

Metapodial

8

5

Mcs-e: 1; Mcd-a: 1; Mcd-b: 1; Mtd-a: 2; Mtd-b: 2; Mtd-c: 1

 

Phalanx 1

8

5

ph1-a: 1; ph1-b: 3; ph1-c: 3

Disarticulation Phalanx 1 / phalanx 2

 

 

 

 

Phalanx

2

1

ph1-d: 1; ph2-d: 1

c - Tendon removal cutmarks

104On the metacarpals, the extraction of the extensor and flexor tendons left cutmarks in 6 out of 8 cases and 7 out of 9 cases (table 27). On the metatarsals, cutmarks related to the removal of the extensor and flexor tendons were respectively noted in 7 cases out of 10 and 7 cases out of 9 (table 27). It is interesting to note that unretouched edges are the only used tools that have, on some carcasses, left no trace. For phalanxes, it is difficult to assess the degree of occurrence of these cutmarks because only four half-carcasses have undergone this step. Of the four, two have cutmarks (table 27).

105On the anterior face of the metacarpals, there is no marked recurrence (table 27). The cutmarks are nevertheless much more frequent in the proximal half of the diaphysis near the groove (5/6). On the metatarsals, the cutmarks are very frequent and abundant at the level of the portion 2 on both sides of the furrow. They can also develop lower on the diaphysis but, depending on the carcasses, they are present at different heights. On the posterior face, the extraction of the tendons mostly leaves short cutmarks on the edges or inside the furrow whether on the metacarpals (5/6) or the metatarsals (7/7). When the tendon removal was carried out using longitudinal gestures, the longitudinal cutmarks (Mcs-fˊMts-fˊ) developing inside the furrow can be associated with these transverse cutmarks.

Table 27 - Occurrence of cutmarks from removal of tendons (see Annex 4 for cutmark codes)

Skeletal element

Tendons

Removal nb

Nb with cutmarks

Occurrence according to location

Metacarpal

 

 

 

 

 

Anterior

8

6

Mcp-b: 1; Mcp-c: 3; Mcs-e: 2; Mcs-e’: 1

 

Posterior

9

7

Mcp-e: 1; Mcs-f: 5; Mcs-f’: 1; Mcd-b’: 1; Mcd-f: 4

Metatarsal

 

 

 

 

 

Anterior

10

7

Mtp-c: 4; Mts-e: 5

 

Posterior

9

7

Mtp-d: 3; Mts-f: 7; Mts-f’: 3; Mtd-b: 1; Mtd-b’: 1; Mtd-f: 4

Phalanx 1

 

 

 

 

 

Posterior

4

2

ph1-f: 1; ph1-g: 1; ph1-h: 1

d - Defleshing cutmarks

106The results concern the 14 half-carcasses of red deer that were defleshed. Bones with no defleshing cutmarks (carpal bones, metapodials and phalanges), axial skeletal bones, coxal bone and cutmarks from rough defleshing of disarticulated carcasses were not considered.

Recurrence of cutmarks

107All the fleshy bones have defleshing cutmarks but their distribution is not homogeneous on the surface of the bones (figure 153). Some areas are systematically striated; this is the case of portion 2 in the lateral face and of portion 3 in the medial face of the scapula, but also of portion 3 of the posterior face of the tibia. Other areas, on the opposite, never have cutmarks: portions 5 and 6 of the radius in the posterior face and portion 6 of the tibia in the lateral face. Of the 96 defined zones, only 11 (17.7 % of them) show recurrences above 70 % (from 14 to 11 occurrences of cutmarks per portion) while more than one third (34.4 %) %) have cutmarks in less than one cases out of two. This heterogeneity in the occurrence of butchery cutmarks on the surface of bones is even more significant when working at a degree of resolution of about one square centimeter. Indeed, at this scale, a majority of bone surfaces (57 %) appear free of cutmarks (figure 154). The visualization by isolines of the defleshing cutmarks gives an image rather close to that perceived taking into account the bone portions (figure 155). However, it is possible to perceive well-defined zones with strong recurrence of cutmarks.

108Overall, areas with a high recurrence (> 11 recurrences) are also areas of high recurrence for all experimenters (figures 156-159). On the other hand, some areas of average recurrence (between 7 and 9) never record cutmarks for some experimenters (figures 160-162).

109Of the three half-carcasses that were defleshed with a flake cleaver, the one cut by V. Mourre showed cutmarks above the coronoid fossa of the humerus and the area near the femoral foramen, whereas no cutmarks was recorded in these areas for half-carcasses butchered by M. Deschamps (figure 159). With the exception of V. Mourre, the other butchers have butchered two half-carcasses at the most and it is therefore difficult to mention possible ways of working according to the butchers. It is interesting to note that for butcheries with denticulates, those cut by V. Mourre very frequently show cutmarks at the two aforementioned locations.

Number of cutmarks

110On average, 339 cutmarks were recorded on each half-carcass, but the very high standard deviation (152.9) shows that this number varies greatly from one experiment to another (table 28). The left side of red deer 6 defleshed with unretouched schist flakes by C. Thiébaut is the one showing the less cutmarks (n=91) and the left side of red deer 2 skinned by A. Coudenneau with flint denticulates the one with the most (n=589).

Table 28 - Number of cutmarks from defleshing recorded on meaty bones listed by anatomical side of the carcass

Experiments

Side

Number of defleshing cutmarks

Red deer 1

R

426

Red deer 1

L

385

Red deer 2

R

524

Red deer 2

L

589

Red deer 3

R

424

Red deer 3

L

312

Red deer 4

R

545

Experiments

Side

Number of defleshing cutmarks

Red deer 4

L

353

Red deer 5

R

244

Red deer 5

L

144

Red deer 6

R

169

Red deer 6

L

91

Red deer 8

R

278

Red deer 8

L

264

111Like for the cutmarks recurrence, there is a strong heterogeneity according to the considered bone portions (figure 163). The scapula, whatever the face, carries a large number of butchery cutmarks: more than 241 in portion 2 of the medial face and between 161 and 240 in portion 2 of the lateral face (table 29). For the long bones, only three portions (portions 1 and 2 of the lateral face of the radio-ulna, portion 4 of the posterior face of the tibia) bear between 161 and 240 cutmarks and three others between 81 and 160 cutmarks (portion 3 of the radio-ulna medial face, portion 3 of the tibia medial face, portion 3 of the tibia posterior face). All other portions bear less than 80 cutmarks. The marking visible on the 5 most affected portions represent 27 % of the total cutmarks. At the resolution of a square centimeter, it is possible to distinguish more precisely intensely striated areas (figure 154). The scapula is striated relatively evenly, although the neck is slightly more intensely striated than the rest. The other elements are characterized by large disparities in cutmarks concentration. With the exception of the femoral neck and tibial plateau, the proximal and distal extremities are almost cutmark free. The two areas with more than 32 cutmarks per square centimeter are located around the interosseous space of the radius in the lateral face, and on the distal diaphysis of the tibia in posterior face. These two account for 6 % of the total number of defleshing cutmarks but only for 0.2 % of the bone surface (table 30). Their location coincides with areas with very high recurrence numbered 8 and 9 (figure 164), to which can be added the neck of the humerus in its posterior face (area 6, figure 164) although this one was less intensely striated (between 16 and 31 cutmarks per square centimeter). In the other areas with very high recurrence (areas 1 to 5 and 7, figure 164), the intensity of the cutmarks is lower (between 8 and 15 cutmarks per square centimeter figure 154). These nine recurrent areas concentrate 12 % of the defleshing cutmarks for only 0.3 % of the bone surface. If we take into account the areas where cutmarks are present in at least one out of two cases (figure 165), 27 areas emerge that gather 37.4 % of the cutmarks.

Figure 153 - Distribution of cutmarks produced during defleshing, by section and surface

Figure 153 - Distribution of cutmarks produced during defleshing, by section and surface

CAD: S. Chong

Figure 154 - Number of cutmarks by cm² produced by defleshing

Figure 154 - Number of cutmarks by cm² produced by defleshing

CAD: S. Chong

Figure 155 - Distribution of cutmarks produced by defleshing. Visualisation with isolines

Figure 155 - Distribution of cutmarks produced by defleshing. Visualisation with isolines

CAD: S. Chong

Figure 156 - Distribution of cutmarks from defleshing produced according to experimenter: Vincent Mourre (VM)

Figure 156 - Distribution of cutmarks from defleshing produced according to experimenter: Vincent Mourre (VM)

CAD: S. Chong

Figure 157 - Distribution of cutmarks from defleshing produced according to experimenter: Céline Thiébaut (CT)

Figure 157 - Distribution of cutmarks from defleshing produced according to experimenter: Céline Thiébaut (CT)

CAD: S. Chong

Figure 158 - Distribution of cutmarks from defleshing produced according to experimenter: Aude Coudenneau / Sandrine Costamagno (AC/SC)

Figure 158 - Distribution of cutmarks from defleshing produced according to experimenter: Aude Coudenneau / Sandrine Costamagno (AC/SC)

CAD: S. Chong

Figure 159 - Distribution of cutmarks from defleshing produced according to experimenter: Marianne Deschamps (MD)

Figure 159 - Distribution of cutmarks from defleshing produced according to experimenter: Marianne Deschamps (MD)

CAD: S. Chong

Figure 160 - Distribution of cutmarks from defleshing according to the tool used: denticulate

Figure 160 - Distribution of cutmarks from defleshing according to the tool used: denticulate

CAD: S. Chong

Figure 161 - Distribution of cutmarks from defleshing according to the tool used: unmodified flake

Figure 161 - Distribution of cutmarks from defleshing according to the tool used: unmodified flake

CAD: S. Chong

Figure 162 - Distribution of cutmarks from defleshing according to the tool used: flake cleaver

Figure 162 - Distribution of cutmarks from defleshing according to the tool used: flake cleaver

CAD: S. Chong

Figure 163 - Number of cutmarks produced during defleshing, by section and surface

Figure 163 - Number of cutmarks produced during defleshing, by section and surface

CAD: S. Chong

Table 29 - Count of cutmarks produced during defleshing by skeletal element, surface, side and section

Element

Portion

Medial

Anterior

Lateral

Posterior

left

right

left

right

left

right

left

right

4

15

13

0

0

6

15

0

0

 Scapula

3

30

53

0

0

14

11

0

0

 

2

138

180

46

41

70

147

68

71

 

1

7

11

8

4

11

4

0

1

1

2

0

3

4

4

2

5

7

 

2

6

17

6

4

9

16

31

48

 Humerus

3

2

7

6

2

37

14

8

12

 

4

14

16

4

15

4

16

6

7

 

5

2

27

12

19

36

25

6

7

 

6

8

18

15

13

19

36

16

47

1

21

25

8

14

78

82

1

4

 

2

16

15

32

28

108

78

6

11

 Radius-ulna

3

35

59

4

7

6

21

4

4

 

4

18

16

23

19

5

25

2

0

 

5

1

14

5

6

4

14

0

0

 

6

0

0

6

2

0

0

0

0

1

7

18

3

0

0

0

15

1

 

2

25

21

19

16

19

29

10

2

 Femur

3

19

20

14

17

12

24

4

14

 

4

21

24

5

14

5

38

13

16

 

5

19

4

14

11

7

3

3

0

 

6

7

2

2

7

0

1

2

0

1

7

7

7

0

12

7

27

14

 

2

5

18

0

0

27

27

5

21

 Tibia

3

26

118

29

14

20

17

54

66

 

4

28

40

10

30

38

31

105

132

 

5

3

7

18

7

8

17

17

13

 

6

0

3

9

2

0

0

9

1

Table 30 - Percentage of cutmarks by size in cm² and relative corresponding bone surface

Number of cutmarks per cm²

Represented surface

Number of cutmarks per class

0

57 %

0 %

1

15 %

10 %

2-3

14 %

24 %

4-7

9.4 %

32 %

8-15

3.2 %

21 %

16-31

0.59 %

7 %

32-66

0.20 %

6 %

100 %

100 %

Figure 164 - Zones for which the number of cutmarks produced during defleshing is equal to or greater than 10 (zones of very high frequency)

Figure 164 - Zones for which the number of cutmarks produced during defleshing is equal to or greater than 10 (zones of very high frequency)

CAD: S. Chong

Figure 165 - Zones for which the number of cutmarks produced during defleshing is equal to or greater than 7 (zones of high frequency)

Figure 165 - Zones for which the number of cutmarks produced during defleshing is equal to or greater than 7 (zones of high frequency)

CAD: S. Chong

Cutmarks orientation

112On the long bones, longitudinal cutmarks are relatively infrequent since on average they represent less than 5 % of the cutmarks. With the exception of left tibia of deer 4 defleshed by V. Mourre with denticulates – whose percentage of longitudinal cutmarks is around 25 % – the frequency of this type of cutmarks varies between 0 and 13 % (figure 166). Whatever the long bone, this frequency is highly variable and, from the data available, it is difficult to identify the factor responsible for such disparities. The radio-ulna most frequently presents longitudinal cutmarks (figure 166c). Located mainly on the ulna, they are related to the particular morphology of this bone. The scapula differs clearly from the long bones since the longitudinal cutmarks represent on average 40.7 % of the cutmarks. On some scapulae, this frequency varies between 20 and 30 % while on others it is around 50 to 60 %: neither the experimenter nor the tool seems to play a role. For example, on scapulas scraped by V. Mourre using denticulates, the percentage of longitudinal cutmarks varies from simple (about 20 %) to triple (nearly 65 %) (figure 166a).

113On all bones, oblique cutmarks are the most numerous (table 31). On scapulae, they correspond on average to half of the cutmarks (figure 166a) but, in some cases, they are less abundant than the longitudinal cutmarks (red deer 1 right side, red deer 4 left side, red deer 8 left side). Depending on the long bones, the percentage of oblique cutmarks varies between 60 (tibia) and 75% (femur) but large disparities are perceptible according to the butcheries. Indeed, in some cases the transverse cutmarks are more numerous than the oblique ones (red deer 2 right humerus, red deer 6 left humerus, red deer 6 right and left radio-ulna, red deer 5 right tibia, red deer 6 right tibia, red deer 8 left tibia) and in others they are present in similar proportions (red deer 6 right humerus, red deer 2 right radio-ulna, red deer 1 left femur, red deer 4 left femur, red deer 8 left femur, red deer 1 left tibia, red deer 2 left tibia). Long bones defleshed with unretouched flakes show transverse cutmarks more frequently (7 out of 12 cases). In comparison, a single skeletal element defleshed with a flake cleaver has this characteristic (table 32). The type of tool used could therefore induce different defleshing gestures independently of the experimenter (table 33).

Table 31 - Number of longitudinal, oblique and transverse cutmarks by element and by side

Table 31 - Number of longitudinal, oblique and transverse cutmarks by element and by side

Table 32 - Number of cases in which transverse defleshing cutmarks (Nb trans.) are more frequent than or similar in number to oblique cutmarks, relative to the total number of butchery cutmarks (Nb butchery), according to the tool used

 

Nb trans.

Nb butchery

Denticulate

4

24

Mousterian point

2

4

Unmodified flake

7

12

Flake cleaver

1

12

Table 33 - Number of cases in which the transverse defleshing cutmarks (Nb trans.) are more frequent than or similar in number to the oblique cutmarks, relative to the total number of butchery cutmarks (Nb butchery), according to the experimenter

 

Nb trans.

Nb butchery

C. Thiébaut

2

8

V. Mourre

6

20

A. Coudenneau

1

4

A. Val

3

4

S. Costamagno

2

4

Figure 166 - Relative frequency of cutmarks per side, according to orientation

Figure 166 - Relative frequency of cutmarks per side, according to orientation

a: scapula; b: humerus; c: radius and ulna; d: femur; e: tibia

Cutmark length

114Regarding the length of the cutmarks, differences are always noticeable between the long bones and the scapula, which stands out with a larger proportion of elongated cutmarks, averaging 22.6 % (table 34, figure 167). These cutmarks are very few on the long bones and, with the exception of two bones (red deer 1 right femur; red deer 6 left tibia), their frequency does not exceed 5 %. No short cutmarks are present on the scapulae. These cutmarks are sometimes absent on the long bones, in particular on the humerus (11 cases out of 14) and the femur (10 cases out of 14). In 9 cases out of 14, the radio-ulna presents this type of cutmarks, which is also found in greater proportions than the other long bones (7.4 % on average against less than 2 %). It can be noted that on this bone, defleshing done with unretouched flakes has more frequently left this type of cutmarks (figure 167c). On the other bones, frequencies greater than 5 % are characteristic of bones defleshed with flakes with the exception of the left tibia of red deer 8 defleshed with denticulates (figure 167). The bone but also, to a lesser extent, the tool could play a role in the length of the cutmarks.

Table 34 - Number of long, intermediate and short cutmarks by skeletal element and side

Table 34 - Number of long, intermediate and short cutmarks by skeletal element and side

Figure 167 - Relative frequency of cutmarks per side, according to length

Figure 167 - Relative frequency of cutmarks per side, according to length

a: scapula; b: humerus; c: radius and ulna; d: femur; e: tibia

D - Discussion

a - Visibility of the various butchery activities: intrinsic bone factors

115For a cutmark to form, the cutting edge of the tool must come into contact with the bone. Muscle attachments, the abundance of fleshy or ligamentous masses, the lability of the articulations are factors that can influence the degree of occurrence of cutmarks in connection with a particular butchery operation. If, as one might expect, the loose articulations like the shoulder did not cause the creation of any disarticulation cutmarks, other much more rigid ones carry only few traces. This is the case, for example, of the distal extremities of the tibia and radius which, in our experiments, never bear disarticulation cutmarks unlike the adjacent tarsal or carpal areas. In contrast, the disarticulation of the elbow systematically leaves cutmarks on the humerus and radio-ulna. It appears that the different articulations do not have the same potential for recording the butchery cutmarks. Nevertheless, because of the reduced number of disarticulated carcasses during the experiments of the PCR, it is difficult to specify the causes of this disparity. Amongst the factors intrinsic to animal anatomy, the thickness of the ligaments could be a path to explore.

116The skinning gestures carried out at the level of fleshy portions are obviously inaccessible to archeozoologists. Despite the absence of meat on the bones of the legs extremities, the experiments carried out in the context of the PCR show that cutmarks are not forming systematically. The longitudinal traces in connection with the skinning are dependent on the face on which it was started, not only by the presence of tendons preventing any contact of the tool with the bone but also by the morphology of the face. The lateral and medial faces of the metatarsal, wider than those of the metacarpal and phalanxes, are indeed more frequently impacted by these longitudinal cutmarks. We can thus expect that, according to the species, these gestures do not have the same recording potential (difference between the bovids and the cervids for example). In our experiments, the circular starting incision was recorded in more than 2/3 of the cases on the metapodials and the first phalanxes whereas at the level of the second phalanxes (close to the hooves), the skinning never left traces because of the thickness of the ligaments. Thus, in layer IV at Combe‑Saunière, the rarity of the cutmarks linked to the circular initial incision on the metapodials and the phalanxes of reindeer (Castel, 1999) indicates a skinning of the hide as close as possible to the hooves, testifying to the care given to its removal because detaching the hide is much more problematic at the level of the phalanxes than at the metapodials.

117In our experiments, the removal of the tendons of the lower legs has left characteristic traces in nearly 3/4 of the cases. If for cervids this activity has a good chance of being visible due to the configuration of the metapodials, for other taxa, such as bovids or equines, the recording potential of these gestures is questionable, the tendons not being inserted in furrows or grooves. Further experiments would therefore be necessary.

118Concerning defleshing cutmarks, the muscular masses present on the bones have been proposed as a factor potentially able to limit the impact of the tools on the bones and thus the formation of butchery cutmarks (Shipman, Rose, 1983a; Binford, 1984; Gaudzinski et al., 2005; Yravedra et al., 2010). In our experiments, the number of cutmarks varies greatly from one butchery session to another: sometimes the fleshiest bones are more intensely striated, sometimes it is the less fleshy bones (table 32). Our results also show that bone surfaces are not impacted homogeneously. Some areas, which correspond to a large part of the bone surfaces, are completely devoid of cutmarks while others are very frequently marked. Areas with a high incidence of cutmarks always correspond to muscle insertion areas (table 35), but not all muscular insertions give rise to significant cutmarks occurrences. Due to such heterogeneity, the frequency of butchery cutmarks is largely dependent on the skeletal portions collected. As several authors have already suggested (Castel, 1991, 1999; Bartram, 1993a; Otárola‑Castillo, 2010; Leduc, 2010; Costamagno, 2012), the frequencies being based on the number of remains with cutmarks relative to the total number of remains should no longer be used for comparisons between skeletal elements, taxa or even sites. In order to evaluate on a site whether the defleshing was systematic or not, we recommend calculating the MNE of each skeletal element taking into account areas with very high (figure 164) or high recurrences (figure 165) and then seeing if these fragments bear cutmarks or not. On the other hand, the unit (CNC) introduced by Y. Abe et al. (2002) to overcome differential conservation problems that may affect MNE‑based cutmark frequencies must be abandoned because the assumption that missing areas would have cutmarks in proportions identical to the preserved zones is not validated by our results.

Table 35 - Muscle attachment areas in relationship to the zones of heavy cutmark occurrence on meaty bones

No. of the area with recurrences ≥ 7

Skeletal element

Face

Muscle insertions

1

scapula

lateral

infraspinatus m., supraspinatus m.

2

scapula

lateral

infraspinatus m.

3

scapula

lateral

deep plane of fibers of the infraspinatus m.

4

scapula

posterior

infraspinatus m.

5

scapula

posterior

infraspinatus m.

6

scapula

anterior

infraspinatus m.

7

scapula

anterior

infraspinatus m.

8

scapula

medial

subscapularis m.

9

humerus

medial

coracobrachialis m.

10

humerus

lateral

brachialis m.

11

humerus

lateral

extensor carpi radialis m.

12

humerus

posterior

brachialis m.

13

humerus

posterior

anconeus m.

14

femur

medial

vastus intermedius m., vastus medialis m.

15

femur

medial

vastus intermedius m.

16

femur

anterior

vastus intermedius m.

17

femur

anterior

vastus intermedius m.

18

femur

anterior

deep gluteal m.

19

femur

lateral

abductor thigh m.

20

radius

anterior

biceps brachii

21

ulna

medial

no insertion

22

radius

anterior

lateral digit extensor m.

23

radius

lateral

lateral digit extensor m.

24

tibia

lateral

tibialis cranial m.

25

tibia

lateral

digit lateral flexor m.

26

tibia

posterior

popliteus m., digit medial flexor m.

27

tibia

posterior

digit lateral flexor m.

In italics, zones of very high occurrence

b - Preliminary remarks on extrinsic bone factors that may affect the shape and frequency of defleshing cutmarks

119The meticulousness of the defleshing, the butcher’s expertise, the state of the carcass, the used tool or the culinary practices are all factors that have been put forward to explain the variability of the relative frequency of the butchery cutmarks on the bones (Guilday, 1962; Potts, 1982; Binford, 1988; Gifford-Gonzalez, 1989a; Delpech, Villa, 1993; Dominguez-Rodrigo, 2002; Costamagno, David 2009; Soulier, Morin, 2016).

120The experiments conducted in the context of the PCR allow discussing the impact of certain factors on the number, the orientation and the length of the cutmarks. The red deer carcasses were all butchered the day after the death of the animal and were defleshed uncooked. The abundance of defleshing cutmarks noted on the bones confirms that defleshing on uncooked meat produces a large number of cutmarks on the opposite to defleshing after boiling up the fleshy portions (Abe, 2005; Costamagno, David, 2009). The variability in the number of cutmarks according to the experiments shows that factors other than the cooking of the carcass affect the formation of the defleshing cutmarks. Because of the large number of experimenters (6) who performed the butchering, of the variability of the tools (4) and of the raw materials used (4), it is nevertheless extremely difficult to determine the respective share of these factors. The raw material could play a role since the two half-carcasses with the lowest number of cutmarks were done using schist and ophite tools. For example, M. Deschamps, who conducted a first butchery with quartzite flake cleavers, and a second one with ophite flake cleavers, left 545 and 144 cutmarks respectively on the fleshy bones (table 29). On the other hand, between quartzite and flint, there does not seem to be any difference. For the same butcher, the number of cutmarks varies from one experiment to another and this does not seem attributable to the used tool. Thus, V. Mourre who carried out four butcheries with quartzite denticulates left between 264 and 524 cutmarks. As the shreds of flesh remaining on the bones were not noted at the end of each experiment, it is impossible to know if these variations are due to the degree of thoroughness of the defleshing or to the expertise of the butcher, this number decreasing with the number of butchered carcasses.

121With respect to the orientation of butchery cutmarks, our results are in line with the hypotheses formulated by M.-C. Soulier and E. Morin (2016) since longitudinal cutmarks are infrequent in our experiments, as expected for defleshing intended for immediate consumption, except on the scapula because of the conformation of this element (see above). Our data are complementing the available actualist reference collections since they allow to document the variability of the orientation and the length of the butchery cutmarks on uncooked carcasses for immediate consumption with lithic tools. Longitudinal cutmarks are infrequent compared to cutmarks obtained during the removal of fillets for drying (Nilssen, 2000). Nevertheless, with respect to Y. Abe (2005)’s results obtained on boiled carcasses, there appears to be some overlap since, in some of our experiments, this percentage may be less than 3 % (figure 168). As expected, a very small percentage of longitudinal cutmarks is therefore not necessarily attributable to defleshing on cooked bone (Soulier, Costamagno, 2017). Moreover, the percentage of longitudinal cutmarks during the same butchery can vary significantly depending on the bones. In isolation, there may be some evidence of fillet removal due to a relatively high percentage of longitudinal cutmarks. For the interpretation of archaeofauna, it is therefore important to reason with all the fleshy bones of the same species as long as the factors responsible for these variations are not apprehended (Soulier, Morin, 2016). At the same time, although oblique cutmarks are often the most frequent cutmarks in our experiments, transverse cutmarks can also be particularly numerous in certain cases, sometimes exceeding oblique cutmarks, which differ from the actualist observations made by P. J. Nilssen (2000) and Y. Abe (2005). This disparity could be related to the types of tools used for butchery (see above) but additional experiments would be necessary to confirm this hypothesis. It is noteworthy, however, that the archaeological material also shows great variability (Soulier, Morin 2016: 50, figure 10). In the light of our results, the orientation of the cutmarks seems, in the current state of the data, a relevant criterion to identify the modes of preparation of the meat and its eventual cooking before its removal.

122Concerning the length of the cutmarks, our results are comforting those of M.-C. Soulier and E. Morin (2016), with an overabundance of intermediate cutmarks compared to elongated cutmarks. Unlike butcheries carried out with metal knives that do not create short cutmarks (Soulier, Morin, 2016), this type of traces was observed in 12 cases out of 14. This percentage shows a great variability. Here the bone, but also probably the tool, could play a role (Soulier, Costamagno, 2017).

Figure 168 - Comparison of the orientation and length of cutmarks from defleshing according to different available references

Figure 168 - Comparison of the orientation and length of cutmarks from defleshing according to different available references

5 - Analysis of the microscopic morphology of butchery cutmarks: connecting cutmarks to used tools

A. Val, S. Costamagno, M.-C. Soulier

123Our approach of microscopic analysis of the cutmarks and, more precisely, of the impact of certain features of stone tools used during the butchery (tool type and raw material) on the microscopic morphology of the cutmarks falls within the continuity of the research of M. Domínguez‑Rodrigo et al. (Domínguez‑Rodrigo et al., 2009; de Juana et al., 2010; Yravedra et al., 2010). We have chosen to apply the same analytical method as the one developed by these authors. This method combines the recording of different variables, simple to observe and quantify using an optical microscope, with statistical processing to determine the type of tool that produced the cutmarks. The choice to use this method, easily reproducible and applicable in theory to the study of archaeological material, makes it possible to compare our results with already existing data on the morphology of butchery marks produced by unretouched and retouched flakes and by bifaces (Domínguez‑Rodrigo et al., 2009; de Juana et al., 2010; Yravedra et al., 2010).

124Studying the influence of the tool used in butchery on cutmarks morphology was done in two distinct stages. In order to estimate the feasibility and the interest of such a study on one hand and to test the relevance of the retained observation criteria on the other hand, a first test-experiment was carried out, during which cutmarks were produced in a standardized way on modern bones. This experiment also allowed us to measure the advantages and disadvantages of two different microscopic observation methods: optical microscope and scanning electron microscope (SEM). The second phase consisted in analyzing cutmarks produced on the carcasses during the butchery sessions.

A - Test-experiment

125For this test-experiment, eight different tools, knapped by C. Thiébaut, V. Mourre and D. Colonge, were used: an unretouched flint flake an unretouched quartzite flake a flint side scraper, a quartzite side scraper, a flint denticulate, a quartzite denticulate, a flint biface and a quartzite biface. Cutmarks were produced on previously defleshed femur and humerus of adult sheep. In order to take into account only the “tool” variable, we have, as far as possible, applied the same force each time in the gesture done by a same and single person (A. Val). The angle formed between the active part of the tool and the surface of the bone was 90° in all cases. A first phase of observation was conducted using an optical microscope (Olympus SZX 16) with a precision of about one micron (magnification between 20 and 50 times). In order to test the possible advantages of a more precise microscopic method of analysis, we also made casts of the marks (use of Coltène Président and resin), carried out observations using a SEM, with a precision of about one nanometer.

126The results of this test-experiment proved sufficiently conclusive in terms of variations of the internal morphology of the cutmarks in relation to the type of blank and the raw material to encourage us to extend the application of the method using the carcasses from the PCR and to attempt applying it to the archaeological record. Moreover, this test-experiment allowed eliminating some observation criteria and demonstrated that the optical microscope is sufficiently precise to characterize the different variables recorded during the cutmarks analysis. From the point of view of the performances offered by both types of microscope, the SEM allows a greater degree of precision than the optical microscope and thus a more detailed image. In our view, the main advantage of the SEM is to present a strong illustrative value, in the sense that, in a phase of characterization of the various types of traces, it becomes possible to obtain excellent images clearly illustrating the subject and to describe precisely the various features of the marks. However, the degree of precision offered by the SEM is not essential and does not provide additional information. The use of this type of microscope can be quite expensive and time consuming. Therefore, for the next phases of the study, we chose to use only an optical microscope.

B - Analysis of butchery cutmarks on the experimental equipment

a - Considered variables

127The influence of both following variables on the morphology of the butchery traces was considered during this work: the type of tool used during the butchery (unretouched flake and unretouched point, flake cleaver, denticulate and Mousterian point) and the raw material of the tool (flint, quartzite, schist and ophite).

b - Material

128The experimental material studied consists of the carcasses exploited during the PCR butchery sessions between 2007 and 2011, that is to say eight carcasses, including one ewe (Ovis aries), six red deers (Cervus elaphus) and one bison (Bison bonasus). Disarticulated red deers have been excluded from our corpus to work only on the marks related to defleshing. In order to limit the number of factors considered, we focused our analysis on the long bones, humerii, radii, femora and tibiae, a total of 64 analyzed elements and 383 cutmarks identified and described (table 36). The choice of long bones compared to other elements of the skeleton (girdle bones, ribs, vertebrae, short bones and cranial bones) is motivated by their general abundance in archaeological sites. Ungulate long bones, and especially diaphyses, are dense and more resilient to differential preservation than other parts of the skeleton (Lyman, 198; Kreutzer, 1992; Lam et al., 1999). In addition, these bones are associated with meaty parts and have a high probability of being brought back from the slaughter site to the butchery site and/or base camp and potentially bear a significant number of cutmarks (e.g. see Binford, 1978, 1981; Abe, 2005). Thus, long bones constitute choice elements when constituting current comparison reference material directly transferable to the analysis of archaeological remains.

Table 36 - Composition of the study sample

Skeletal element

Material

NISP

Traces

Humerus

16

85

Radius

16

57

Femur

16

95

Tibia

16

146

Total

64

383

c - Method

Microscopic analysis

129All the experimental and archaeological material was studied in the TRACES laboratory, using an optical microscope with a 20 to 45 times magnification. Each of the cutmarks was numbered and recorded on anatomical charts on the Inkscape software.

130The criteria used during the microscopic analysis of the cutmarks on experimental and archaeological material (table 37) are directly inspired by those developed by M. Domínguez‑Rodrigo et al. (Domínguez‑Rodrigo et al., 2009; de Juana et al., 2010; Yravedra et al., 2010). These criteria, easy to observe with an optical microscope, have the advantage of being applicable to numerically large series, whether experimental or archaeological. Two variables considered at the beginning of the analysis were excluded following the test-experiment. These variables, inspired by C. Fritz (1999), were as follows: consistency or not of the cutmark depth and presence / absence of a micro-rim on one or both edges of the cutmark. We abandoned the criteria of consistency of cutmark depth, because it is too difficult to measure accurately (quantified) and extremely variable. Micro-rims, which are not very diagnostic, are generally not preserved on archaeological material, as some post-depositional processes tend to eliminate them; recording this criterion is therefore not relevant when studying fossil remains.

Table 37 - Criteria developed through microscopic analysis of the cutmarks on experimental and archaeological material

1

Trajectory of the groove (straight, curvy or sinuous)

2

Presence / absence of a“barb effect”

3

Shape of the groove (V or \_/)

4

Symmetry of the groove in cross-section (symmetrical / asymmetrical)

5

Presence / absence of a“shoulder effect”

6

Flaking on the edge of the groove (a: present on one edge; b: present on both edges; c: absent)

7

Extent of the flaking on the edge of the groove (a: >1/3 of the trajectory; b: < 1/3)

8

Presence / absence of micro-striations

9

Trajectory of the micro-striations (continuous or discontinuous)

10

Shape of the micro-striations (straight or irregular: curvy or sinuous)

11

Location of the micro-striations (a: on one side; b: at the bottom of the groove ; c: on both sides)

12

Presence / absence of multiple marks

13

Number of multiple marks

14

Presence / absence of fork-shaped marks

15

Number of fork-shaped marks

16

Presence / absence of intersecting marks

131For the definitions of the “barb effect” and “shoulder effect”, we are referring to those proposed by P. Shipman and J. Rose (1983a). The “barb effect” corresponds to the presence of a “barb” or group of micro-cutmarks forming at the beginning and/or at the end of the main cutmark (figure 169). The “shoulder effect” is associated with the existence of a secondary cutmark – on one or both edges of the main cutmark – parallel to the first one (figure 169). Most often, the “shoulder effect” is accompanied by micro-cutmarks located between the main cutmark and the secondary cutmark. The scarring corresponds to the “flakingcategory described by M. Domínguez‑Rodrigo et al. (2009: 2646) as a “series of exfoliation of the shoulder edge”. In other words, it is a bone surface removal, between the main cutmark and the “shoulder”. The scarring can be located along one or both edges of the groove and can develop all along the cutmark or only on part of it (criterion 7). The “fork-shaped marks” and “multiple marks” are described by S. de Juana et al. (2010). A “fork-shaped markis a trace that separates into two, like a fork (figure 169). The “multiple marks” category refers to a group of cutmarks produced during a single contact between the tool and the bone (figure 169). They correspond to “a set of multiple marks produced during a single stroke that are not in contact with each other” (de Juana et al., 2010: 1843). It has sometimes been difficult to estimate the number of bone / tool contacts corresponding to multiple marks; other criteria were taken into account in order to distinguish cutmarks generated by several gestures from multiple marks produced by a single gesture. We therefore define a multiple mark as a set of cutmarks less than 0.1 mm apart and with a similar overall direction (less than 45° difference). Marks perpendicular between them were thus considered as resulting from distinct bone / tool contacts. Finally, we added a category called “intersecting marks” to include the special cases of multiple marks that intersect and overlap (with an angle below 45°) (figure 169).

Microscopic analysis

132We have followed the same multi-varied statistical approach as that developed by M. Domínguez‑Rodrigo et al. (2009; used by de Juana et al., 2010; Yravedra et al., 2010) to allow us direct comparisons between our results and those presented in these research works. In order to take into account all the variables at the same time, a Categorical Principal Component Analysis (CATPCA) was performed with the SPSS software using the same discretization method used by M. Domínguez-Rodrigo et al. (that is by normalizing the main variable). To facilitate the interpretation of the statistical results, the final scores were obtained by Varimax rotation with a Kaiser normalization of the categorical principal component analysis. Of the sixteen variables considered (table 37), we retained the two numerical variables (13 and 15; table 37) and thirteen of the fourteen categorical variables. Criterion 7 “extension of the scarring on the edges of the cutmark” was excluded since it does not vary. Thus, when a scar is present, it is always on less than a third of the length of the groove. A binary logistic regression analysis was then carried out to distinguish the cutmarks produced by unretouched tools from those produced by retouched tools. The proportions for each variable were compared between these two groups by calculating proportions and confidence intervals rectified according to the Wald method (Agresti, Coull, 1998). The results of this statistical analysis conducted by E. Discamps were published in 2017 (Val et al., 2017).

Figure 169 - Different characteristics and categories of traces considered during microscopic analysis

Figure 169 - Different characteristics and categories of traces considered during microscopic analysis

The multiple traces presented in the photograph were produced in a single instance of bone / tool contact, during phase 1 of the project, the test experiment

Photographs: A. Val ; DAO : M. Coutureau

C - Results

a - Influence of the raw material

133Five different raw materials were chosen for the production of the tool blanks used during the butchery sessions: flint, quartzite, fine quartzite, ophite and schist. Whatever the raw material considered (fine like flint or grained like the others), this criterion does not seem to have any influence on the internal morphology of the butchery cutmark. Thus, a comparison between the cutmarks produced by the two denticulates (one in flint and one in quartzite) shows great similarities between the two groups of cutmarks for each of the 16 variables considered (p>0.05 in all cases; see tables 38-41).

b - Unretouched/ retouched blanks difference

Table 38 - Morphological characteristics of cutmarks according to the type of flake or tool used

Tool

Raw material

1. Trajectory

2. Barb effect

3. Form

4. Groove symmetry

5. Shoulder effect

straight = 57.7

present = 3.8

V = 84.6

symmetrical = 30.8

present = 47.2

Unretouched flake

quartzite

curvy = 23.1

absent = 96.2

\_/ = 15.4

asymmetrical = 69.2

absent = 52.8

 

sinuous = 26.9

straight = 57.2

present = 13

V = 82.6

symmetrical = 26.1

present = 69.6

Unretouched flake

fine quartzite

curvy = 21.4

absent = 87

\_/ = 17.4

asymmetrical = 73.9

absent = 30.4

 

sinuous = 21.4

s

straight = 60

present = 0

V = 93.3

symmetrical = 46.7

present = 53.3

Unretouched flake

chist

curvy = 20

absent = 100

\_/ = 6.7

asymmetrical = 53.3

absent = 46.7

 

sinuous = 20

straight = 43

present = 3.3

V = 93.3

symmetrical = 20

present = 50

Unretouched point

flint

curvy = 9

absent = 96.7

\_/ = 6.7

asymmetrical =80

absent = 50

 

sinuous = 48

t

straight = 72.7

present = 9.1

V = 86.4

symmetrical = 18.2

present = 90.9

Mousterian poin

flint

curvy = 4.5

absent = 90.9

\_/ = 13.6

asymmetrical = 81.8

absent = 9.1

 

sinuous = 22.8

straight = 62.9

present = 9.7

V = 80.6

symmetrical = 24.2

present = 69.4

Flake cleaver

quartzite

curvy = 14.5

absent = 90.3

\_/ = 19.4

asymmetrical = 75.8

absent = 30.6

 

sinuous = 22.6

straight = 73.7

present = 10.5

V = 63.2

symmetrical = 26.3

present = 78.9

Flake cleaver

ophite

curvy = 5.3

absent = 89.5

\_/ = 36.8

asymmetrical = 73.7

absent = 21.1

 

sinuous = 21

straight = 40.5

present = 14.3

V = 85.7

symmetrical = 38.1

present = 72.6

Denticulate

flint

curvy = 25

absent = 85.7

\_/ = 14.3

asymmetrical = 61.9

absent = 27.4

 

sinuous = 34.5

straight = 25

present = 12.5

V = 91.7

symmetrical = 29.2

present = 56.9

Denticulate

quartzite

curvy = 30.6

absent = 87.5

\_/ = 8.3

asymmetrical = 70.8

absent = 43.1

 

sinuous = 44.4

Data provided in percentages for the observation criteria 1 to 5

Table 39 - Morphological characteristics of cutmarks according to the type of flake or tool used

Table 39 - Morphological characteristics of cutmarks according to the type of flake or tool used

Data provided in percentages for the observation criteria 6 to 11 (cont. = continuous, discont. = discontinuous, str. = striations)

134In order to avoid the multiplication of the factors that can influence the morphology of the marks, we have only taken into account the long bones of cervids to test the influence of the type of blank used. Some differences seemed to emerge before statistical data processing (table 38): traces produced by unretouched blanks (flakes and points) differed from a quantitative point of view (number of cutmarks, number of multiple marks and average number of cutmarks composing the multiple marks; table 42) from those produced by retouched tools (denticulates, Mousterian points and flake cleavers), but also from a qualitative point of view by the presence or absence of “parasitic” marks (“barb effect”, “fork-shaped marks”, micro-cutmarks and “intersecting marks”; table 43). Indeed, the three types of processed tools (biface, Mousterian point and denticulate) produced not only more cutmarks than the unretouched cutting edges, both in absolute numbers and taking into account all secondary cutmarks, but they also produced multiple marks composed of a larger number of cutmarks than in the case of unretouched flakes and point. The traces associated with the retouched cutting edges are characterized by a greater frequency of “barb effect”, micro-cutmarks, “fork-shaped marks” and “intersecting marks”.

Table 40 - Morphological characteristics of cutmarks according to the type of flake or tool used

Tool

Raw material

12. Multiples marks

13. Nb multiples marks

14. Fork-shaped marks

15. Nb fork-shaped marks

16. Intersecting marks

Unretouched flake

quartzite

present = 53.8

between 2 and 4

present = 3.8

1

present = 3.8

absent = 46.2

absent = 96.2

absent = 96.2

Unretouched flake

fine quartzite

present = 43.5

between 1 and 3

present = 8.7

1

present = 4.3

absent = 56.5

absent = 91.3

absent = 95.7

Unretouched flake

schist

present = 60

between 2 and 6

present = 13.3

1

present = 0

absent = 40

absent = 86.7

absent = 100

Unretouched point

flint

present = 43.3

between 2 and 10

present = 16.7

1

present = 0

absent = 56.7

absent = 83.3

absent = 100

Mousterian point

flint

present = 59.1

between 2 and 12

present = 18.2

1

present = 9.1

absent = 40.9

absent = 81.8

absent = 90.9

Flake cleaver

quartzite

present = 43.5

between 2 and 12

present = 17.7

1

present = 11.3

absent = 56.5

absent = 82.3

absent = 88.7

Flake cleaver

ophite

present = 57.9

between 2 and 9

present = 5.3

1

present = 10.5

absent = 42.1

absent = 94.7

absent = 89.5

Denticulate

flint

present = 40.5

between 2 and 11

present = 21.4

between 1 and 2

present = 4.8

absent = 59.5

absent = 78.6

absent = 95.2

Denticulate

quartzite

present = 43.1

between 2 and 8

present = 9.7

between 1 and 2

present = 6.9

absent = 56.9

absent = 90.3

absent = 93.1

Data provided in percentages for the observation criteria 12 to 16

Table 41 - Abundance of cutmarks according to the type of flake or tool used in butchery on three taxa represented in the experimental sample (stags and does, Cervus elaphus; ewe, Ovis aries and bison, Bison bonasus)

Tool

Raw material

Taxon

Nb long bones defl

Nb marks

Nb average marks by bone

Nb total marks*

Nb average total marks by bone*

Unretouched flake

quartzite

C. elaphus

4

26

6.5

49

12.25

Unretouched flake

fine quartzite

C. elaphus

4

23

5.75

38

9.5

Unretouched flake

schist

C. elaphus

4

15

3.75

40

10

Total flakes

12

64

5.3

118

9.8

Unretouched point

flint

C. elaphus

4

30

7.5

73

18.25

Mousterian point

flint

C. elaphus

4

22

5.5

79

19.75

Total points

8

52

6.5

152

19

Flake cleaver

quartzite

C. elaphus

8

62

7.75

123

15.4

Flake cleaver

ophite

C. elaphus

4

19

4.75

44

11

Total flake cleavers

12

81

6.75

167

13.9

Denticulate

flint

C. elaphus

8

84

10.5

181

22.6

Denticulate

quartzite

C. elaphus

8

72

9

156

19.5

Total denticulates

16

156

9.75

337

21.1

Total Cervids

48

353

7.4

783

16.3

Unretouched flake

quartzite

O. aries

8

13

1.6

21

2.6

Various**

flint and quartzite

B. bonasus

8

17

2.1

18

22.5

* including all of the secondary traces associated with a primary cutmark in cases of multiple traces
** several categories of tool were used for the butchery of the bison: unmodified flakes, denticulates, flake cleavers, Mousterian points and bifaces

135However, none of these apparent differences statistically significant (Val et al., 2017). In other words, none of the 16 variables considered, nor any combination of them, makes it possible to distinguish in a statistically significant way the cutmarks produced by unretouched blanks from those produced by tools. The total number of cutmarks within the multiple marks associated with the unretouched blanks does not differ statistically from that associated with the retouched cutting edges (X2 =9.98, df =11; p>0.05). Similarly, although intersecting marks are more frequent in the case of multiple marks produced by retouched tools (17 %) than by unretouched blanks (4 %), being extremely rare in the second case (n=2) however, the difference is not statistically significant (X2 =3.54; p>0.05). Fork-shaped marks are present for 16 % of the marks made by denticulates, 18 % by Mousterian points, 15 % by flake cleavers and only 11 % by unretouched blanks (table 43), but, again, these differences are not statistically significant (X2 =1.50; p>0.05). Similarly, although the trajectory of the cutmarks varies, it is straight for 52 % of the marks produced by unretouched blanks (sinuous: 31 %; curved: 17 %) and for 45 % of the marks produced by tools (sinuous: 32 %; curved: 22 %), these variations are not validated statistically (X2 =1.71; p>0.05) (Val et al., 2017).

c - Differences between tools (Mousterian point, cleavers and denticulates)

136No variable or combination of variables is diagnostic to distinguish cutmarks made by the cleavers from those made by the Mousterian point. On the other hand, the marks associated with the denticulates have certain features that are observed neither in the case of the cutmarks associated with the unretouched flakes, nor in the case of those associated with the points or the flake cleavers. Firstly, from a quantitative point of view, the denticulates produced the greatest number of cutmarks, either in gross value or including all secondary cutmarks, in the case of multiple marks (table 42). In addition, cutmarks produced by denticulates show more frequently a barb‑effect than those produced by other unretouched or retouched blanks (table 41) and this difference is statistically significant (X2 =4.17; p<0.05). Micro‑cutmarks are also more often observed on cutmarks made by denticulates than on those associated with flakes, points and flake cleavers (table 43), but these differences are not statistically supported (p>0.05 in all cases).

Table 42 - Quantitative characteristics of cutmarks observed on cervid bones, according to the tool used

Tool type

Nb average marks by bone

Nb average total marks by bone*

Nb max. striations in multiple marks

Nb average striations in multiple marks

Unretouched blanks

6.3

12.8

12

3.3

Mousterian points

5.5

19.75

12

5.4

Denticulates

9.8

21.1

11

3.7

* including all of the secondary cutmarks associated with a primary cutmark in cases of multiple traces

Table 43 - Morphological characteristics of traces influenced by the type of tool employed during butchery (data reflects the percentage of cutmarks presenting each of the mentioned criteria)

Tool type

Fork-shaped marks

Barb effect

Micro-striations

Flaking

Intersecting marks*

Unretouched blanks

10.6 % (10/94)

5.3 % (5/94)

14 % (13/93)

2.1 % (2/94)

4.3 % (2/46)

Mousterian points

18.2 % (4/22)

9.1 % (2/22)

22.7 % (5/22)

4.5 % (1/22)

15.4 % (2/13)

Flake cleavers

14.8 % (12/81)

9.9 % (8/81)

23.5 % (19/81)

3.7 % (3/81)

23.7 % (9/38)

Denticulates

16.0 % (25/156)

13.5 % (21/156)

34.6 % (54/156)

1.3 % (2/156)

13 % (9/69)

* percentage of “intersecting marks” among the multiple traces

137Finally, two particular types in the organization of the cutmarks composing multiple marks made by denticulates were recorded. “Type 1” is a group of cutmarks including a long main cutmark, most often with a straight or slightly sinuous trajectory, and several short secondary cutmarks parallel to each other, of similar length and situated at one extremity of the main cutmark, following the same direction. The trajectory of these short cutmarks varies and may be straight, sinuous or curved (figure 170). “Type 2” is a group of marks consisting of two main cutmarks, usually curved and surrounding, distally or proximally, two or three secondary shorter and also curved cutmarks (figure 170).

Figure 170 - Two groupings of multiple traces characteristic of the cutmarks produced by denticulates

Figure 170 - Two groupings of multiple traces characteristic of the cutmarks produced by denticulates

Photographs: A. Val; CAD: M. Coutureau

D - Discussion

a - Tool raw material influence and cutmarks morphology: comparison with existing data

138Various research works that have used microscopic analysis techniques performing better than a simple optical microscope (in particular the SEM) have succeeded in highlighting a direct relationship between the raw material of the tools used during cutting and the internal morphology of the cutmarks. However, all of these studies are dealing with raw materials of completely different nature: stone and metal (Greenfield, 1999, 2000), stone and bamboo (West, Louys, 2007), stone, bamboo and shell (Choi, Driwantoro, 2007). It should also be noted that these studies are based on the observation of a relatively limited number of cutmarks and that the purely descriptive results given are not supported by statistical tests. Concerning the possible differences between the cutmarks produced by various lithic tools materials (for example obsidian, flint chert, quartz and quartzite), our results confirm the majority of the existing remarks on this subject in the literature, namely the extreme difficulty, even the impossibility, of identifying the raw material of the lithic tools used during the butchery from the morphology of the marks, even with the help of a SEM (see e.g. Olsen, 1988; Greenfield, 2006; de Juana et al., 2010).

139Whether or not real differences exist in the microscopic morphology of the cutmarks produced by tools knapped in various lithic raw materials, these are impossible to demonstrate with the methods of analysis used up to now, including with the multi-varied and systematic approach we followed (Val et al., 2017).

b - Lithic blank influence on the cutmarks morphology: comparisons with the results of M. Domínguez-Rodrigo et al. (2009) and S. de Juana et al. (2010)

140S. de Juana et al. (2010) recognize statistically discriminating three variables between the cutmarks produced by unretouched flakes, retouched flakes and bifaces: scars (presence/ absence and their extension), frequency of the multiples marks and frequency of fork-shaped marks. Our results differ considerably (table 44). In particular, whereas in the study conducted by S. de Juana et al. the scarring of the cutmark edges is frequent and this for all types of cutmarks, it is observed only in eight cases out of 353 or only 2 % in our sample (table 43). Conversely, the multiple marks recorded on half of the cutmarks generated by the unretouched blanks in our study are totally absent from the cutmarks associated with the unretouched flakes in the sample of S. de Juana et al. Finally, in the work of these authors, the frequency of the fork-shaped marks is more important for the unretouched flakes than for the retouched flakes. It is also higher in the retouched flakes than for bifaces. In our experimental sample, this increase, while visible, is not as pronounced and is not statistically significant (table 44).

Table 44 - Quantitative data for the different variables considered diagnostic by S. de Juana et al. 2010 for identifying the cutmarks produced by unmodified flakes, retouched flakes and bifaces compared with the results of our study

Tool type

Flaking (present)

Fork-shaped marks (present)

Nb fork-shaped marks

Multiples marks (present)

Nb multiple marks

Biface*

74.1 %

83.5 %

between 2 and 9

93 %

1-10

Retouched flake*

54/105 (51.4 %)

44/105 (41.9 %)

between 1 and 2

64/105 (61 %)

1-3

Unretouched flake*

36/246 (14.6 %)

3/246 (1 %)

0

0/246 (0 %)

0

Unretouched point

2/94 (2 %)

10/94 (11 %)

1

46/94 (49 %)

1-10

Mousterian point

1/22 (5 %)

4/22 (18 %)

1

13/22 (59 %)

1-12

Flake cleaver

3/81 (4 %)

12/81 (15 %)

1

38/81 (47 %)

1-12

Denticulate

2/156 (1 %)

25/156 (16 %)

between 1 and 2

69/156 (44 %)

1-11

* data adapted from M. Domínguez-Rodrigo et al., 2009; S. de Juana et al., 2010

141In the current state of the data, it is difficult to identify with certainty the factors that may explain these differences. The lithic blanks considered in our study were all knapped in fine‑grained or slightly coarser-grained rocks (flint fine quartzite, quartzite, ophite and schist). The raw material of the unretouched and retouched flakes used for the study of S. de Juana et al. (same experimental sample as that described and analyzed by M. Domínguez-Rodrigo et al., 2009) is not specified; the bifaces are in flint and fine and coarse quartzite. Nevertheless, although S. de Juana et al. do not mention the possible role played by the raw material on the morphology of the cutmarks, our results suggest that it is non-existent (see Part I, chapter 3.5.C.a).

142The size of our corpus (n=94 cutmarks for the unretouched blanks, n=259 for the retouched cutting edges) is comparable to that of the corpus studied by S. de Juana et al. (n=246 for unretouched flakes, n=105 for retouched flakes, n=212 for bifaces) and therefore hardly explains the significant differences observed in the results. In addition, all the cutmarks, with the exception of those generated by bifaces, were produced during experimental butchery sessions on animal carcasses (five goats and a young cow in the case of M. Domínguez‑Rodrigo et al., 2009; a sample reused by S. de Juana et al.; six deers in our study). This implies that cutmarks were created without any control or looking for systematization neither of the applied force, nor of the angle between the tool and the bone, or of the direction of the cutting motions. The bifaces were used to produce cutmarks on bones already defleshed and cleaned. However, as a control measure, these same tools were also used to deflesh two whole deer limbs and no difference was recorded between the two groups of resulting cutmarks (de Juana et al., 2010).

143The only differences between the two studies that could contribute to explain the variations observed between our results and those proposed by S. de Juana et al. are related to the size of the animals on which the butchery was carried out and to the types of marks analyzed. Thus, the size difference between goat and cervids bones could have influenced the morphology of the cutmarks although the nature of the relationship between the size of the animal and the morphology and frequency of the cutmarks remains very poorly understood (see Part I, chapter 3.5.D.c “Influence of the Taxon?”). Finally, while our study focuses on cutmarks present only on long bones and associated exclusively with defleshing activities, those analyzed by S. de Juana et al. (2010) are located both on long bones and on flat bones (ribs) and potentially associated, for some of them, with activities other than the removal of muscle masses, for example hide removal, scraping, tendon removal or disarticulation. Further studies are needed to explore the potential role of these factors (animal size, bone category considered and butchery gesture) on the internal morphology of the cutmarks.

c - Preliminary remarks on the possible role of other variables

Influence of the butcher?

144Testing the influence of the level of physical strength and expertise of the person carrying out the butchery on the number and internal morphology of the cutmarks was not part of the objectives of the theme. Nevertheless, all the butchers were adults with more or less the same level of expertise, in order to minimize the influence of this factor. Moreover, in view of the results obtained following the observation of the cutmarks produced on the experimental sample, it seems that the potential differences in physical strength did not affect the cutmarks from neither a qualitative nor quantitative point of view (table 45).

145A comparison between the number of cutmarks recorded on the long bones of half-carcasses prepared using unretouched flakes of slightly coarse-grained raw materials (quartzite, figure quartzite and schist) by three different persons (V. Mourre, C. Thiébaut and A. Val) shows no difference in terms of neither morphology (tables 38-40) nor cutmarks number (table 45). The example of flint denticulates confirms that the physical strength of the butcher has a limited impact on the abundance and the morphology of the cutmarks, since the cutmarks observed on the long bones of the two half-carcasses prepared with these tools have the same morphological features (tables 38-40) but differ quantitatively in a significant way (from single to double, table 42) although they were produced by two women, with equivalent physical strength (C. Thiébaut and A. Coudenneau). Finally, the best example of the lack of a direct link between the butcher and the number and morphology of cutmarks is that of the two half-carcasses prepared by the same butcher (C. Thiébaut).

Table 45 - Abundance of cutmarks according to the person performing the butchery. These data only include the long bones of deer half-carcasses that were butchered by a single person in one session (i.e. the half‑carcass prepared by A. Coudenneau and S. Costamagno with a Mousterian point is therefore excluded)

Butcher

Sex

Tools

Nb bones

Nb marks

Average nb by bone

Nb total marks*

Average Nb total by bone*

VM

quartzite denticulate

8

72

9

156

19.5

quartzite flake cleaver

4

39

9.75

74

18.5

quartzite flake

4

26

6.5

49

12.25

CT

schist flake

4

15

37.5

40

10

flint denticulate

4

50

12.5

120

30

AC

flint denticulate

4

34

8.5

61

15.25

AV

fine quartzite flake

4

23

5.75

38

9.5

MD

quartzite flake cleaver

4

23

5.75

49

12.25

ophite flake cleaver

4

19

4.75

44

11

M-PC

flint point

4

30

7.5

73

18.25

Abbreviations: VM = Vincent Mourre; CT = Céline Thiébaut; AC = Aude Coudenneau; AV = Aurore Val; MD = Marianne Deschamps; M-PC = Marie-Pierre Coumont.
* including all of the secondary cutmarks associated with a primary cutmark in cases of multiple traces

146This experimenter, using in the first case schist flakes and in the second flint denticulates, produced respectively the smallest and the largest number of marks of the experimental sample (both in absolute numbers and taking into account secondary marks: table 45).

Influence of the taxon?

147Red deers make up most of the experimental sample, while animals belonging to other species used for butchery sessions are represented by only two carcasses: an adult female bison (Bison bonasus) and an adult ewe (Ovis aries). It is therefore premature at this time to make definite conclusions about the influence of the taxon on the quantitative and qualitative characteristics of butchery cutmarks. However, some preliminary remarks are worth mentioning.

148No difference in the internal morphology of the cutmarks according to the considered taxon was noted. However, significant quantitative differences are to be emphasized. For butchering the bison carcass, unretouched and retouched flakes (denticulates, bifaces and Mousterian points) in flint and quartzite were used. Only 17 butchery cutmarks were recorded on the long bones (including a single multiple mark composed of a main cutmark and an auxiliary one; table 41). For comparison, even the unretouched flakes in coarse-grained material that, in the experimental sample made of cervid bones, left the least cutmarks, produced an average of 5.3 traces per bone (9.8 including all secondary marks), compared with an average of 2.1 (2.25 with secondary marks) for flakes and tools used for the bison (table 41). Moreover, the marks observed are all superficial and small. The human factor can be eliminated since the various people who carried out the butchery on the bison (V. Mourre, C. Thiébaut, M. Deschamps, É. Claud, S. Costamagno and A. Val) also participated in the butchery of half‑carcasses of cervids. In addition, this factor seems to have no influence on the abundance of cutmarks. The size of the animal could be involved. The relationship, positive or negative, between the size of the animal and the abundance of cutmarks on the carcass is still poorly understood; opinions on the question differ and experimental data sometimes contradict each other (see, for example, Crader, 1983; Lyman, 1992, 2005; Domínguez-Rodrigo, Barba, 2005; Pobiner, Braun, 2005; Potter, 2005; Bello et al., 2009; Domínguez-Rodrigo, Yravedra, 2009 and discussion in Part II, chapter 3.1).

149For some very large animals, and in particular proboscideans, several authors have suggested that the importance of muscle masses, tendons and ligaments, contributes to limit the frequency of bone / tool contacts and thus the potential number of cutmarks (Villa, 1990; Fosse, 1998; Gaudzinski et al., 2005; Villa et al., 2005; Mussi, Villa, 2008; Yravedra et al., 2010, 2012). The possibility of totally preparing an elephant carcass with very few cutmarks has also been confirmed through ethnographic observations (Crader, 1983; Haynes, Krasinski, 2010). A similar hypothesis, based on the large size of the animal, could be put forward to explain the small number of cutmarks observed on the bison carcass. On the other hand, such a hypothesis is no longer valid in the case of a small caprine. As for the bison, the cutting up of the ewe carcass, for which pseudo‑Levallois points and unretouched quartzite flakes were used, produced only a very small number of cutmarks (only 15 traces recorded on long bones; table 41). For the moment, we do not have a satisfactory hypothesis to offer to explain such a significant difference between the number of marks observed on the ewe bones compared to those of the cervids, except a possible difference at the level of muscular masses, more or less developed depending on whether we are dealing with wild (here, deer) or domesticated (sheep) species.

6 – Summary

S. Costamagno, M.-C. Soulier, A. Val

150The experiments carried out as part of the PCR research project on butchering carcasses using Mousterian tools have allowed us to produce a butchery cutmark reference collection, which can greatly contribute to update our interpretative frameworks.

151In terms of butchery gestures and exploitation of the related resources, the protocol used allowed us to relate the cutmarks to a particular activity, or even a specific gesture. As observed by P. J. Nilssen (2000), it appears that some cutmarks close to the articular surfaces, traditionally considered as disarticulation marks following the work of L. R. Binford (1981), are in fact defleshing marks. We were able to show that, depending on the location and the direction of the disarticulation marks, it was possible to tell whether certain joints were held bent or extended. Although difficult to identify on the long bones, faltering cutmarks potentially produced by inexperienced butchery could be observed on the vertebral bodies. We identified different types of skinning marks, which, according to their location, allowed us to establish the ways in which the skin was removed (either as a single piece or by separating the skin of the trunk from that of the legs). Thus, the presence of longitudinal cutmarks on the lateral surface of the metapodials, documented at several Palaeolithic sites, appears to reflect the separate removal of the skin of the limbs from that of the trunk. Cutmarks related to tendon removal, considered for the first time in an experimental context, were also particularly informative, as they allowed us to identify, according to their direction, the different methods used. In the European Palaeolithic context, it appears that tendon removal using tools held in a transverse manner to the axis of the bones was the most common technique. Thanks to the PCR reference collection, it is now possible to interpret the butchery gestures made with lithic tools much more precisely. Our experiments show that the type of tool had little influence on the location or direction of the butchery marks. As these experiments were based on flake tools, further experiments with blade tools are nonetheless required. The new cutmark coding proposed, which takes these advances into account, is a quick way of identifying the different gestures and activities carried out on a given archeofauna. By developing this type of approach on a large number of Palaeolithic bone assemblages, it may then be possible to start discussing any potential cultural norms governing the butchery of the exploited species.

152Quantitative studies carried out using GIS have allowed us to improve our knowledge on the factors that can influence the formation of butchery marks. Whatever the butchery operations considered, the question of whether or not butchery gestures are recorded mainly depends on the bone on which the operation takes place. It appears that, depending on the bone and the species, a given activity has a greater or lesser probability of being detected. This variability can also be observed at the level of the bone itself: for example, the longitudinal incision of the skin on the posterior surface will never leave cutmarks because of the presence of the flexor tendon. On the fleshy long bones, the presence of defleshing cutmarks depends on the muscle insertions. As a result, significant portions of long bone surfaces are completely devoid of cutmarks. This heterogeneity in the recording of cutmarks calls into question the quantitative units usually used to document the relative abundance of butchery marks (% of pieces presenting cutmarks). Based on the areas with a high potential for documenting cutmarks, it now seems possible to quantify the number of occurrences of a given operation. Due to the number of variables involved in our experimental protocol, it is difficult to discuss the extrinsic factors that may affect the shape and frequency of the butchery marks. Some hypotheses – which need to be completed by further experiments performed by a single experimenter – can nevertheless be proposed. Some raw materials may affect the number of cutmarks; however, the type of tool does not seem to have an impact, and, in this case; it is generally the butcher and/ or the degree of thoroughness with which the bones are defleshed that could be evoked to explain any disparities observed. Regarding the orientation and the length of the defleshing marks, the PCR reference collection completes the available modern reference collections (Nilssen, 2000; Abe, 2005). It thus allows us to highlight a characteristic signature of butchery carried out on raw meat with lithic tools, continuing the discussion initiated by M.-C. Soulier and E. Morin (2016). It is clear that further experiments using carcasses in different conditions and / or cooking could provide crucial information on these matters.

153Regarding the morphology of the butchery marks in relation to the tool used, distinguishing between the cutmarks produced by different lithic raw materials appears impossible based on the method of analysis used by the research project, consistent with previously obtained results (Olsen, 1988; Greenfield, 2006; de Juana et al., 2010). With respect to the distinction between unretouched/ retouched edges, no differences could be observed based on the criteria used, which somewhat contradicts conclusions drawn by S. de Juana et al. (2010). It also appeared relatively difficult to apply this multi-variate method to the fossil record, given the wide diversity of lithic tools that could have been used during butchery activities in Palaeolithic contexts. Nonetheless, both quantitative and qualitative criteria seem to distinguish the cutmarks made by denticulates from those made by other retouched tools. However, more work is needed to confirm these observations as it is not currently possible to exclude the fact that other tools such as endscrapers and side scrapers, which were also potentially used during butchery, could produce patterns of cutmarks similar to those produced by denticulates. Although the data is limited because of the number of experimenters and tools used, our results suggest that neither the sex of the butcher nor their physical strength have any influence on the morphology of the cutmarks. This morphology is not dependent on the size of the butchered animal either. To take these questions further, it would be worth conducting a new series of butchery experiments, based on the same taxon and using the same type of tool, varying only the person carrying out the butchery (e.g., male / female, child / adult, experienced butcher / beginner). The potential influence of animal-related factors other than size would also need to be considered and tested, as some authors have already suggested (see for example Binford, 1984; Dewbury, Russell, 2007). These factors notably include the age of the animal, the stage of decomposition and the condition of the carcass (fresh versus frozen), which could, particularly in the case of frozen carcasses or carcasses in the stage of rigor mortis, have an impact on the strength required during butchery and thus potentially increase the amount of bone / tool contact. Finally, the use of more powerful analytical tools would need to be tested to provide accurate quantitative data as well as roughness measurements – using a laser profilometer Alicona 3D Infinite Focus or laser interferometer – to see whether it is possible to advance in these matters relating to cutmarks morphology.

A. Val, S. Costamagno, M.-C. Soulier

Haut de page

Table des illustrations

Titre Figure 100 - Forms used for recording the gestures used and the instances of tool contact with bone during the experiments
Légende a: form for skinning activities; b: form for the neck; c: form for the ribcage; d: form for the lumbar vertebrae and the sacrum; e: form for the forelimb; f: form for the hind limb
Crédits CAD: M.-P. Coumont
URL http://journals.openedition.org/palethnologie/docannexe/image/4089/img-1.jpg
Fichier image/jpeg, 168k
Titre Figure 101 - Example of the form used to record instances of tool contact with bone, the duration of the butchery activities, and comments from the experimenter during the experiment
Crédits CAD: M.-P. Coumont
URL http://journals.openedition.org/palethnologie/docannexe/image/4089/img-2.jpg
Fichier image/jpeg, 335k
Titre Figure 102 - Skinning of red deer 1
Légende a: circular incision on the metapodial and start of the longitudinal incision on the skin of the medial surface (denticulate); b: incision on the medial surface of the limb (Mousterian point); c: detachment of the hide from the ribcage (denticulate); d: detachment of the skin from the tibia (Mousterian point)
Crédits Photographs: PCR Des Traces et des Hommes
URL http://journals.openedition.org/palethnologie/docannexe/image/4089/img-3.jpg
Fichier image/jpeg, 313k
Titre Figure 103 - Form for recording the gestures used in the skinning of red deer 1
Légende a: left side with the location of the incision of the skin and the instances of contact with bone (Mousterian points); b: right side (denticulates). The bones indicated in grey made contact with tools, without precise contact location
Crédits CAD: M.-P. Coumont
URL http://journals.openedition.org/palethnologie/docannexe/image/4089/img-4.jpg
Fichier image/jpeg, 100k
Titre Figure 104 - Form for recording defleshing and tendon removal on the forelimbs of red deer 1
Légende a: right side (denticulates); b: left side (Mousterian points)
Crédits CAD: M.-P. Coumont
URL http://journals.openedition.org/palethnologie/docannexe/image/4089/img-5.jpg
Fichier image/jpeg, 74k
Titre Figure 105 - a: defleshing of the left radius and ulna of red deer 1; b: disarticulation of the coxofemoral joint of the left hind limb of red deer 1
Crédits Photographs: PCR Des Traces et des Hommes
URL http://journals.openedition.org/palethnologie/docannexe/image/4089/img-6.jpg
Fichier image/jpeg, 149k
Titre Figure 106 - Form for recording the defleshing and tendon removal on the hind limbs of red deer 1
Légende a: right side (denticulates); b: left die (Mousterian points)
Crédits CAD: M.-P. Coumont
URL http://journals.openedition.org/palethnologie/docannexe/image/4089/img-7.jpg
Fichier image/jpeg, 91k
Titre Figure 107 - Gestures related to the disarticulation of the phalanges of a red deer 1
Crédits CAD: M.-P. Coumont
URL http://journals.openedition.org/palethnologie/docannexe/image/4089/img-8.jpg
Fichier image/jpeg, 43k
Titre Figure 108 - Gestures related to the defleshing and disarticulation of the post-cranial axial skeleton
Légende a: right side; b: left side
Crédits CAD: S. Costamagno
URL http://journals.openedition.org/palethnologie/docannexe/image/4089/img-9.jpg
Fichier image/jpeg, 104k
Titre Figure 109 - Defleshing and segmentation of the axial skeleton of red deer 1
Légende a: defleshing of the neck; b: removal of the sirloin; c: defleshing of the ribs; d: disarticulation of the ribs by percussion; e: incision between the ribs; f: detachment of the ribs by flexion and sawing
Crédits Photographs: PCR Des Traces et des Hommes
URL http://journals.openedition.org/palethnologie/docannexe/image/4089/img-10.jpg
Fichier image/jpeg, 557k
Titre Figure 110 - Skinning of red deer 7
Légende a: incision of the skin on the internal surface of the right hind limb; b: circular incision of the skin at the shaft of the right metatarsal; c: incision of the skin on the sternum
Crédits Photographs: PCR Des Traces et des Hommes
URL http://journals.openedition.org/palethnologie/docannexe/image/4089/img-11.jpg
Fichier image/jpeg, 259k
Titre Figure 111 - Defleshing and disarticulation of the left forelimb of red deer 7
Légende a: removal of the limb; b: cutting the tendons and ligaments for removal of the muscles; c: disarticulation of the scapula and the humerus; d: disarticulation of the elbow; e: disarticulation of the “wrist”; f: disarticulation of the radius and the first row of carpal bones
Crédits Photographs: PCR Des Traces et des Hommes
URL http://journals.openedition.org/palethnologie/docannexe/image/4089/img-12.jpg
Fichier image/jpeg, 398k
Titre Figure 113 - Defleshing and disarticulation of the left forelimb of red deer 7
Légende a-b: disarticulation of the shoulder; c-e: disarticulation of the elbow; f: disarticulation of the radius from the first row of right carpal bones
Crédits Photographs: PCR Des Traces et des Hommes
URL http://journals.openedition.org/palethnologie/docannexe/image/4089/img-13.jpg
Fichier image/jpeg, 285k
Titre Figure 114 - Defleshing and disarticulation of the right hind limb of red deer 7
Légende a: disarticulation of the coxofemoral joint of the right limb; b: disarticulation of the right knee; d: detachment of the muscles inserted into the tibia; c, e: disarticulation of the tarsal and metatarsal
Crédits Photographs: PCR Des Traces et des Hommes
URL http://journals.openedition.org/palethnologie/docannexe/image/4089/img-14.jpg
Fichier image/jpeg, 314k
Titre Figure 115 - State of the bones after the defleshing of red deer 7
Légende a: left forelimb; b: left hind limb; c: right forelimb; d: right hind limb
Crédits Photographs: PCR Des Traces et des Hommes
URL http://journals.openedition.org/palethnologie/docannexe/image/4089/img-15.jpg
Fichier image/jpeg, 218k
Titre Figure 116 - Cutmarks related to skinning of a: the metacarpal; b: the metatarsal; c: the tibia; d: the phalanges; e: the vestigial phalanges; f: pyramidal, lunate and pisiform; g: the calcaneus and the cubonavicular bones
Légende Abbreviations: M = medial, A = anterior, L = lateral, P = posterior, Ab = abaxiale
Crédits CAD: M.-C. Soulier and M. Coutureau
URL http://journals.openedition.org/palethnologie/docannexe/image/4089/img-16.jpg
Fichier image/jpeg, 114k
Titre Figure 117 - Cutmarks from skinning on the metatarsal
Crédits Photographs: M.-C. Soulier
URL http://journals.openedition.org/palethnologie/docannexe/image/4089/img-17.jpg
Fichier image/jpeg, 59k
Titre Figure 118 - Cutmarks from butchery on the mandible
Légende a: cutmarks related to skinning; b: cutmarks related to defleshing; c: cutmarks related to disarticulation. Abbreviations:V = vestibular, L = lingual
Crédits CAD: M.-C. Soulier and M. Coutureau
URL http://journals.openedition.org/palethnologie/docannexe/image/4089/img-18.jpg
Fichier image/jpeg, 78k
Titre Figure 119 - Cutmarks related to defleshing on a: the axis; b: the other cervical vertebrae; c: the sacrum; d: the thoracic vertebrae; e: the lumbar vertebrae
Légende Abbreviations: Cr = cranial, D = dorsal, L = lateral, V = ventral
Crédits CAD: M.-C. Soulier and M. Coutureau
URL http://journals.openedition.org/palethnologie/docannexe/image/4089/img-19.jpg
Fichier image/jpeg, 174k
Titre Figure 120 - Cutmarks related to defleshing on a: the scapula; b: the calcaneus; c: the humerus; d: the radioulnar/ ulna; e: the femur / the patella; f: the tibia / the fibula
Légende Abbreviations: M = medial, A = anterior, L = lateral, P = posterior
Crédits CAD: M.-C. Soulier and M. Coutureau).
URL http://journals.openedition.org/palethnologie/docannexe/image/4089/img-20.jpg
Fichier image/jpeg, 277k
Titre Figure 121 - Cutmarks related to the disarticulation of the limbs on a: the humerus; b: the radioulnar/ ulna; c: the femur; d: the fibula, the sesamoids, the malleolus bone; e: the metacarpal; f: the metatarsal; g: the carpal bones; h: the tarsal bones; i: the first phalanges; j: the second phalanges and the vestigial phalanges. The cutmarks from testing for the location of the joint are in blue
Légende Abbreviations: M = medial, A = anterior, L = lateral, P = posterior, Ab = abaxial
Crédits CAD: M.-C. Soulier and M. Coutureau
URL http://journals.openedition.org/palethnologie/docannexe/image/4089/img-21.jpg
Fichier image/jpeg, 260k
Titre Figure 122 - Cutmarks related to the segmentation of the vertebral column on a: the atlas; b: the axis; c: the other cervical vertebrae; d: the thoracic vertebrae; e: the lumbar vertebrae. The cutmarks from testing for the location of the joint are in blue
Légende Abbreviations: Cr = cranial, Ca = caudal, D = dorsal, M = lateral, V = ventral
Crédits CAD: M.-C. Soulier and M. Coutureau).
URL http://journals.openedition.org/palethnologie/docannexe/image/4089/img-22.jpg
Fichier image/jpeg, 172k
Titre Figure 123 - Cutmarks related to the extraction of the tendons on a: the metacarpal; b: the metatarsal; c: the first phalanges
Légende Abbreviations: M = medial, A = anterior, L = lateral, P = posterior, Ab = abaxial
Crédits CAD: M.-C. Soulier and M. Coutureau).
URL http://journals.openedition.org/palethnologie/docannexe/image/4089/img-23.jpg
Fichier image/jpeg, 149k
Titre Figure 124 - Removal of the tendons
Légende a: transverse gesture for removal of the tendon; b: longitudinal gesture for removal of the tendon; c: cutmarks characteristic of transverse gestures used for tendon detachment; d: cutmarks characteristic of longitudinal gestures for tendon detachment
Crédits CAD: M.-C. Soulier
URL http://journals.openedition.org/palethnologie/docannexe/image/4089/img-24.jpg
Fichier image/jpeg, 184k
Titre Figure 125 - Illustration of the gestures relative to some cutmarks attributed by A.B. Galán and M. Dominguez‑Rodrigo (2013) to cutmarks from disarticulation
Légende The double arrows indicate the gestures associated with these cutmarks
Crédits CAD: M.-C. Soulier
URL http://journals.openedition.org/palethnologie/docannexe/image/4089/img-25.jpg
Fichier image/jpeg, 50k
Titre Figure 126 - Location of the cutmarks related to codes S-12 and Hp-3 (white lines) interpreted by L.R. Binford (1981) as cutmarks from disarticulation
Légende The dotted line indicates the position of the joint and the arrows indicate the gestures associated with these cutmarks
Crédits Photograph and CAD: M.-C. Soulier
URL http://journals.openedition.org/palethnologie/docannexe/image/4089/img-26.jpg
Fichier image/jpeg, 78k
Titre Figure 127 - Red deer limbs showing the parts of the bones in direct contact with the skin (dotted areas)
Crédits Photographs and CAD: M.-C. Soulier
URL http://journals.openedition.org/palethnologie/docannexe/image/4089/img-27.jpg
Fichier image/jpeg, 67k
Titre Figure 128 - Illustration of the new codes for cutmarks related to butchery on the mandible
Légende The dark grey zones correspond to the zones where only a single activity was documented; the very light grey zones correspond to zones where cutmarks could not be attributed to a particular activitya Abbreviations: dir. = direction of the cutmark; T = transverse; O = oblique; L = longitudinal. c Abbreviations: Vest. = vestibular; Ling. = lingual. d The activities are taken from Binford (1981), Nilssen (2000) and the T&H experimental reference (see Annex 4 for details). Abbreviations: DP = skinning; DS = disarticulation; DC = defleshing. Caution, the axis considered is different between the vertical and the horizontal branch..
URL http://journals.openedition.org/palethnologie/docannexe/image/4089/img-28.jpg
Fichier image/jpeg, 56k
Titre Figure 129 - Illustration of the new codes related to cutmarks from butchery on the atlas
Légende The dark grey zones correspond to the zones where only a single activity was documented; in the light grey zones, the orientation of the cutmark allowed identification of the corresponding activity, the very light grey zones correspond to ambiguous zones where cutmarks cannot be attributed to a particular activitya Abbreviations: dir. = direction of the cutmark; T = transverse; O = oblique; L = longitudinal. c Abbreviations: Dors. = dorsal; Vent. = ventral; Cra. = cranial; Cau. = caudal. d The activities are taken from Binford (1981), Nilssen (2000) and from the T&H experimental reference (see Annex 4 for details). Abbreviations: EV = evisceration; DS = disarticulation; DC = defleshing, tât. = testing for joint
URL http://journals.openedition.org/palethnologie/docannexe/image/4089/img-29.jpg
Fichier image/jpeg, 52k
Titre Figure 130 - Illustration of the new codes related to cutmarks from butchery on the axis
Légende The dark grey zones correspond to the zones where only a single activity was documented; in the light grey zones, the orientation of the cutmark allowed identification of the corresponding activity, the very light grey zones correspond to ambiguous zones where cutmarks cannot be attributed to a particular activitya Abbreviations: dir.= direction of the cutmark;T = transverse; ST = sub-transverse; O = oblique; L = longitudinal. c Abbreviations: Med. = medial; Lat. = lateral; Dors. = dorsal; Vent.= ventral. d The activities are taken from Binford (1981), Nilssen (2000) and the T&H experimental reference (see Annex 4 for details). Abbreviations: EV = evisceration; DS = disarticulation; DC = defleshing; tât. = testing for join
URL http://journals.openedition.org/palethnologie/docannexe/image/4089/img-30.jpg
Fichier image/jpeg, 74k
Titre Figure 131 - Illustration of the new codes related to cutmarks from butchery on the other cervical vertebrae
Légende The dark grey zones correspond to the zones where only a single activity was documented; in the light grey zones, the orientation of the cutmark allowed identification of the corresponding activitya Abbreviations: dir. = direction of the cutmark; T = transverse; ST = sub-transverse; O = oblique; L = longitudinal. c Abbreviations: Med. = medial; Lat. = lateral; Dors. = dorsal; Vent. = ventral; Cra. = cranial. d The activities are taken from Binford (1981), Nilssen (2000) and the T&H experimental reference (see Annex 4 for details). Abbreviations: EV = evisceration; DS = disarticulation; DC = defleshing; tât. = testing for join
URL http://journals.openedition.org/palethnologie/docannexe/image/4089/img-31.jpg
Fichier image/jpeg, 89k
Titre Figure 132 - Illustration of the new codes related to cutmarks from butchery on the thoracic vertebrae
Légende The dark grey zones correspond to the zones where only a single activity was documented; in the light grey zones, the orientation of the cutmark allowed identification of the corresponding activity, the very light grey zones correspond to ambiguous zones where cutmarks cannot be attributed to a particular activitya Abbreviations: dir. = direction of the cutmark; T = transverse; O = oblique; L = longitudinal. c Abbreviations: Med. = medial; Lat. =lateral; Dors. = dorsal; Vent. = ventral; Cra. = cranial; Cau. = caudal. d The activities are taken from Binford (1981), Nilssen (2000) and the T&H experimental reference (see Annex 4 for details). Abbreviations: EV = evisceration; DS = disarticulation; DC = defleshing
URL http://journals.openedition.org/palethnologie/docannexe/image/4089/img-32.jpg
Fichier image/jpeg, 83k
Titre Figure 133 - Illustration of the new codes related to cutmarks from butchery on the lumbar vertebrae
Légende The dark grey zones correspond to the zones where only a single activity was documented; in the light grey zones, the orientation of the cutmark allowed identification of the corresponding activitya Abbreviations: dir. = direction of the cutmark; T = transverse; O = oblique; L = longitudinal. c Abbreviations: Med. = medial; Lat. = lateral; Dors. = dorsal; Vent. = ventral; Cra. = cranial; Cau. = caudal. d The activities are taken from Binford (1981), Nilssen (2000) and the T&H experimental reference (see Annex 4 for details). Abbreviations: EV = evisceration; DS = disarticulation; DC = defleshing
URL http://journals.openedition.org/palethnologie/docannexe/image/4089/img-33.jpg
Fichier image/jpeg, 92k
Titre Figure 134 - Illustration of the new codes related to cutmarks from butchery on the sacrum
Légende The dark grey zones correspond to the zones where only a single activity was documenteda Abbreviations: dir. = direction of the cutmark; T = transverse; O = oblique; L = longitudinal. c Abbreviations: Med. = medial; Lat. = lateral; Dors. = dorsal; Vent. = ventral; Cra. = cranial. d The activities are taken from Binford (1981), Nilssen (2000) and the T&H experimental reference (see Annex 4 for details). Abbreviations: EV = evisceration; DS = disarticulation; DC = defleshing
URL http://journals.openedition.org/palethnologie/docannexe/image/4089/img-34.jpg
Fichier image/jpeg, 67k
Titre Figure 135 - Illustration of the new codes related to cutmarks from butchery on the humerus
Légende The dark grey zones correspond to the zones where only a single activity was documented; in the light grey zones, the orientation of the cutmark allowed identification of the corresponding activity, the very light grey zones correspond to ambiguous zones where cutmarks cannot be attributed to a particular activitya Abbreviations: dir.= direction of the cutmark;T = transverse; O = oblique; L = longitudinal; ST = sub-transverse; SL = sub-longitudinal. b The long bones were subdivided into six sections. c Abbreviations: M = medial; A = anterior; P = posterior; L = lateral. d The activities are taken from Binford (1981), Nilssen (2000), Vigne (2005), Costamagno, David (2009) and the T&H experimental reference (see Annex 4 for details). Abbreviations: DP = skinning; DS = disarticulation; DC = defleshing
URL http://journals.openedition.org/palethnologie/docannexe/image/4089/img-35.jpg
Fichier image/jpeg, 108k
Titre Figure 136 - Illustration of the new codes related to cutmarks from butchery on the radius
Légende The dark grey zones correspond to the zones where only a single activity was documented; in the light grey zones, the orientation of the cutmark allowed identification of the corresponding activity, the very light grey zones correspond to ambiguous zones where cutmarks cannot be attributed to a particular activitya Abbreviations: dir.= direction of the cutmark;T.= transverse; O.= oblique; L = longitudinal; ST = sub-transverse; SL = sub-longitudinal. b The long bones were subdivided into six sections. c Abbreviations: M = medial; A = anterior; P = posterior; L = lateral. d The activities are taken from Binford (1981), Bez (1995), Nilssen (2000), Costamagno, David (2009) and the T&H experimental reference (see Annex 4 for details). Abbreviations: DP = skinning; DS = disarticulation; DC = defleshing; TN = removal of the tendons. XX+XX indicates that both activities are documented; XX(+XX?) indicates that the first activity mentioned is attested but that the second, in parentheses, is uncertain; XX?/XX? indicates that the protocol used does not allow discrimination between both activities
URL http://journals.openedition.org/palethnologie/docannexe/image/4089/img-36.jpg
Fichier image/jpeg, 85k
Titre Figure 137 - Illustration of the new codes related to cutmarks from butchery on the ulna
Légende The dark grey zones correspond to the zones where only a single activity was documented; in the light grey zones, the orientation of the cutmark allowed identification of the corresponding activity, the very light grey zones correspond to ambiguous zones where cutmarks cannot be attributed to a particular activity a Abbreviations: dir.= direction of the cutmark;T.= transverse; O.= oblique; L = longitudinal; ST = sub-transverse; SL = sub-longitudinal. b The long bones were subdivided into six sections. c Abbreviations: M = medial; A = anterior; P = posterior; L = lateral; Palm. = palmar. d The activities are taken from Binford (1981), Bez (1995), Nilssen (2000), Costamagno, David (2009) and the T&H experimental reference (see Annex 4 for details). Abbreviations: DP = skinning; DS = disarticulation; DC = defleshing
URL http://journals.openedition.org/palethnologie/docannexe/image/4089/img-37.jpg
Fichier image/jpeg, 68k
Titre Figure 138 - Illustration of the new codes related to cutmarks from butchery on the carpal bones
Légende The dark grey zones correspond to the zones where only a single activity was documented; in the light grey zones, the orientation of the cutmark allowed identification of the corresponding activitya Abbreviations: dir. = direction of the cutmark; T = transverse; O = oblique; L = longitudinal. c Abbreviations: M = medial; A = anterior; P = posterior; L = lateral; Palm. = palmar. d The activities are taken from Binford (1981), Bez (1995), Nilssen (2000), Costamagno, David (2009) and the T&H experimental reference (see annex 4 for details). Abbreviations: DP = skinning; DS = disarticulation; DC = defleshing; TN = removal of the tendons, ANT = anterior; POST = posterior; RAD = radius; CAR1 = proximal row of carpals; CAR2 = distal row of carpals
URL http://journals.openedition.org/palethnologie/docannexe/image/4089/img-38.jpg
Fichier image/jpeg, 96k
Titre Figure 139 - Illustration of the new codes related to cutmarks from butchery on the metacarpal
Légende The dark grey zones correspond to the zones where only a single activity was documented; in the light grey zones, the orientation of the cutmark allowed identification of the corresponding activity, the very light grey zones correspond to ambiguous zones where cutmarks cannot be attributed to a particular activitya Abbreviations: dir. = direction of the cutmark; T = transverse; O = oblique; L = longitudinal. b The long bones were subdivided into six sections. c Abbreviations: M = medial; A = anterior; P = posterior; L = lateral; Palm. = palmar. d The activities are taken from Binford (1981), Bez (1995), Nilssen (2000), Costamagno, David (2009) and the T&H experimental reference (see Annex 4 for details). Abbreviations: DP = skinning; DS = disarticulation; DC = defleshing; TN = removal of the tendons, ANT = anterior; POST = posterior
URL http://journals.openedition.org/palethnologie/docannexe/image/4089/img-39.jpg
Fichier image/jpeg, 87k
Titre Figure 140 - Illustration of the new codes related to cutmarks from butchery on the scapula
Légende The dark grey zones correspond to the zones where only a single activity was documented; in the light grey zones, the orientation of the cutmark allowed identification of the corresponding activity ().a Abbreviations: dir. = direction of the cutmark; T = transverse; O = oblique; L = longitudinal; ST = sub-transverse. c Abbreviations: M.= medial; A = anterior; P = posterior; L.= lateral. d The activities are taken from Binford (1981), Bez (1995), Nillsen (2000), Costamagno, David (2009) and the T&H experimental reference (see Annex 4 for details). Abbreviations: DS = disarticulation; DC = defleshing
URL http://journals.openedition.org/palethnologie/docannexe/image/4089/img-40.jpg
Fichier image/jpeg, 45k
Titre Figure 141 - Illustration of the new codes related to cutmarks from butchery of the femur
Légende The dark grey zones correspond to the zones where only a single activity was documented; in the light grey zones, the orientation of the cutmark allowed identification of the corresponding activity, the very light grey zones correspond to ambiguous zones where cutmarks cannot be attributed to a particular activitya Abbreviations: dir. = direction of the cutmark; T = transverse; O = oblique; L = longitudinal. b The long bones were subdivided into six sections. c Abbreviations: M = medial; A = anterior; P = posterior; L = lateral; Palm. = palmar. d The activities are taken from Binford (1981), Nillsen (2000), Vigne (2005), Costamagno, David (2009) and the T&H experimental reference (see annex 4 for details). Abbreviations: DP = skinning; DS = disarticulation; DC = defleshing
URL http://journals.openedition.org/palethnologie/docannexe/image/4089/img-41.jpg
Fichier image/jpeg, 107k
Titre Figure 142 - Illustration of the new codes related to cutmarks from butchery of the tibia
Légende The dark grey zones correspond to the zones where only a single activity was documented; in the light grey zones, the orientation of the cutmark allowed identification of the corresponding activity, the very light grey zones correspond to ambiguous zones where cutmarks cannot be attributed to a particular activitya Abbreviations: dir. = direction of the cutmark; T = transverse; O = oblique; L = longitudinal. b The long bones were subdivided into six sections. c Abbreviations: M = medial; A = anterior; P = posterior; L = lateral; Palm. = palmar; Cra. = cranial. d The activities are taken from Binford (1981), Nilssen (2000), Vigne (2005), Costamagno, David (2009) and the T&H experimental reference (see Annex 4 for details). Abbreviations: DP = skinning; DS = disarticulation; DC = defleshing
URL http://journals.openedition.org/palethnologie/docannexe/image/4089/img-42.jpg
Fichier image/jpeg, 113k
Titre Figure 143 - Illustration of the new codes related to cutmarks from butchery of the fibula, patella and malleolus
Légende The dark grey zones correspond to the zones where only a single activity was documenteda Abbreviations: dir. = direction of the cutmark;T = transverse; O = oblique. b Abbreviations: Prox. = proximal. c Abbreviations: A = anterior; L = lateral. d The activities are taken from Binford (1981), Nilssen (2000), Vigne (2005), Costamagno, David (2009) and the T&H experimental reference (see annex 4 for details). Abbreviations: DS = disarticulation; DC = defleshing
URL http://journals.openedition.org/palethnologie/docannexe/image/4089/img-43.jpg
Fichier image/jpeg, 23k
Titre Figure 144 - Illustration of the new codes related to cutmarks from butchery on the tarsal bones
Légende The dark grey zones correspond to the zones where only a single activity was documenteda Abbreviations: dir. = direction of the cutmark; T = transverse; O = oblique; L = longitudinal. c Abbreviations: M = medial; A = anterior; P = posterior; L = lateral. d The activities are taken from Binford (1981), Bez (1995), Nilssen (2000), Costamagno, David (2009) and the T&H experimental reference (see Annex 4 for details). Abbreviations: DP = skinning; DS = disarticulation; TN = removal of the tendons, ANT = anterior; POST = posterior; TIB = tibia; TAR1 = proximal tow of tarsals; TAR2 = distal row of tarsals; MTM = metatarsal; MET = metapodial; PH1 = first phalanx; PH2 = second phalanx; exten. = extension; flex. = flexion. XX+XX indicates that both activities are documented; XX(+XX?) indicates that the first activity mentioned is attested but that the second, in parentheses, is uncertain
URL http://journals.openedition.org/palethnologie/docannexe/image/4089/img-44.jpg
Fichier image/jpeg, 95k
Titre Figure 145 - Illustration of the new codes related to cutmarks from butchery of the metatarsal
Légende The dark grey zones correspond to the zones where only a single activity was documented; in the light grey zones, the orientation of the cutmark allowed identification of the corresponding activitya Abbreviations: dir.= direction of the cutmark;T = transverse; O = oblique; L = longitudinal; ST = sub-transverse. b The long bones were subdivided into six sections. c Abbreviations: M = medial; A = anterior; P = posterior; L = lateral; Palm.= palmar. d The activities are taken from Binford (1981), Bez (1995), Nilssen (2000), Costamagno, David (2009) and the T&H experimental reference (see Annex 4 for details). Abbreviations: DP = skinning; DS = disarticulation; TN = removal of the tendons, ANT = anterior; POST = posterior; TIB = tibia; TAR1 = proximal row of tarsals; TAR2 = distal row of tarsals; MTM = metatarsal; MET = metapodial; PH1 = first phalanx; PH2 = second phalanx; exten. = extension; flex. = flexion
URL http://journals.openedition.org/palethnologie/docannexe/image/4089/img-45.jpg
Fichier image/jpeg, 97k
Titre Figure 146 - Illustration of the new codes related to cutmarks from butchery on the phalanges and sesamoids
Légende The dark grey zones correspond to the zones where only a single activity was documented; in the light grey zones, the orientation of the cutmark allowed identification of the corresponding activity, the very light grey zones correspond to ambiguous zones where cutmarks cannot be attributed to a particular activitya Abbreviations: dir.= direction of the cutmark;T = transverse; O = oblique; L = longitudinal; ST = sub-transverse, SL = sub-longitudinal. c Abbreviations: M = medial; A = anterior; P = posterior; L = lateral; Cra. = cranial; Ab = abaxial. d The activities are taken from Binford (1981), Bez (1995), Nilssen (2000), Costamagno, David (2009) and the T&H experimental reference (see Annex 4 for details). Abbreviations: DP = skinning; DS = disarticulation; TN = removal of the tendons, ANT = anterior; POST = posterior; TIB = tibia; TAR1 = proximal row of tarsals; TAR2 = distal row of tarsals; MTM = metatarsal; MET = metapodial; PH1 = first phalanx; PH2 = second phalanx
URL http://journals.openedition.org/palethnologie/docannexe/image/4089/img-46.jpg
Fichier image/jpeg, 97k
Titre Figure 147 - Location of disarticulation cutmarks when the elbow is held in flexion or in extension
Crédits CAD: M.-C. Soulier
URL http://journals.openedition.org/palethnologie/docannexe/image/4089/img-47.jpg
Fichier image/jpeg, 21k
Titre Figure 148 - Anatomical portions considered for the scapula and long bones
Crédits CAD: M.-C. Soulier
URL http://journals.openedition.org/palethnologie/docannexe/image/4089/img-48.jpg
Fichier image/jpeg, 65k
Titre Figure 149 - Categories applied to the orientation of the cutmarks
Crédits Modified from Soulier, Morin, 2016
URL http://journals.openedition.org/palethnologie/docannexe/image/4089/img-49.jpg
Fichier image/jpeg, 10k
Titre Figure 150 - Example of relationships applied to the data
Légende a: data related to the cutmark; b: data related to the section of the bone; c: join between the cutmark and the section of the bone
Crédits CAD: S. Chong
URL http://journals.openedition.org/palethnologie/docannexe/image/4089/img-50.jpg
Fichier image/jpeg, 30k
Titre Figure 151 - Placement of the cutmarks according to a: the anatomical element; b: the section; c: the grid overlay
Crédits CAD: S.Chong
URL http://journals.openedition.org/palethnologie/docannexe/image/4089/img-51.jpg
Fichier image/jpeg, 33k
Titre Figure 152 - Example of the use of GIS in the analysis of the distribution of cutmarks from butchery in a given location
Légende a: cutmarks formed during three distinct butchery operations; b: creation of a buffer zone of 1 cm; c: zones of overlap; d: visualization of cutmark occurrences using isolines
Crédits CAD: S. Chong
URL http://journals.openedition.org/palethnologie/docannexe/image/4089/img-52.jpg
Fichier image/jpeg, 64k
Titre Figure 153 - Distribution of cutmarks produced during defleshing, by section and surface
Crédits CAD: S. Chong
URL http://journals.openedition.org/palethnologie/docannexe/image/4089/img-53.jpg
Fichier image/jpeg, 92k
Titre Figure 154 - Number of cutmarks by cm² produced by defleshing
Crédits CAD: S. Chong
URL http://journals.openedition.org/palethnologie/docannexe/image/4089/img-54.jpg
Fichier image/jpeg, 86k
Titre Figure 155 - Distribution of cutmarks produced by defleshing. Visualisation with isolines
Crédits CAD: S. Chong
URL http://journals.openedition.org/palethnologie/docannexe/image/4089/img-55.jpg
Fichier image/jpeg, 84k
Titre Figure 156 - Distribution of cutmarks from defleshing produced according to experimenter: Vincent Mourre (VM)
Crédits CAD: S. Chong
URL http://journals.openedition.org/palethnologie/docannexe/image/4089/img-56.jpg
Fichier image/jpeg, 80k
Titre Figure 157 - Distribution of cutmarks from defleshing produced according to experimenter: Céline Thiébaut (CT)
Crédits CAD: S. Chong
URL http://journals.openedition.org/palethnologie/docannexe/image/4089/img-57.jpg
Fichier image/jpeg, 71k
Titre Figure 158 - Distribution of cutmarks from defleshing produced according to experimenter: Aude Coudenneau / Sandrine Costamagno (AC/SC)
Crédits CAD: S. Chong
URL http://journals.openedition.org/palethnologie/docannexe/image/4089/img-58.jpg
Fichier image/jpeg, 75k
Titre Figure 159 - Distribution of cutmarks from defleshing produced according to experimenter: Marianne Deschamps (MD)
Crédits CAD: S. Chong
URL http://journals.openedition.org/palethnologie/docannexe/image/4089/img-59.jpg
Fichier image/jpeg, 72k
Titre Figure 160 - Distribution of cutmarks from defleshing according to the tool used: denticulate
Crédits CAD: S. Chong
URL http://journals.openedition.org/palethnologie/docannexe/image/4089/img-60.jpg
Fichier image/jpeg, 80k
Titre Figure 161 - Distribution of cutmarks from defleshing according to the tool used: unmodified flake
Crédits CAD: S. Chong
URL http://journals.openedition.org/palethnologie/docannexe/image/4089/img-61.jpg
Fichier image/jpeg, 75k
Titre Figure 162 - Distribution of cutmarks from defleshing according to the tool used: flake cleaver
Crédits CAD: S. Chong
URL http://journals.openedition.org/palethnologie/docannexe/image/4089/img-62.jpg
Fichier image/jpeg, 78k
Titre Figure 163 - Number of cutmarks produced during defleshing, by section and surface
Crédits CAD: S. Chong
URL http://journals.openedition.org/palethnologie/docannexe/image/4089/img-63.jpg
Fichier image/jpeg, 77k
Titre Figure 164 - Zones for which the number of cutmarks produced during defleshing is equal to or greater than 10 (zones of very high frequency)
Crédits CAD: S. Chong
URL http://journals.openedition.org/palethnologie/docannexe/image/4089/img-64.jpg
Fichier image/jpeg, 34k
Titre Figure 165 - Zones for which the number of cutmarks produced during defleshing is equal to or greater than 7 (zones of high frequency)
Crédits CAD: S. Chong
URL http://journals.openedition.org/palethnologie/docannexe/image/4089/img-65.jpg
Fichier image/jpeg, 39k
Titre Table 31 - Number of longitudinal, oblique and transverse cutmarks by element and by side
URL http://journals.openedition.org/palethnologie/docannexe/image/4089/img-66.jpg
Fichier image/jpeg, 112k
Titre Figure 166 - Relative frequency of cutmarks per side, according to orientation
Légende a: scapula; b: humerus; c: radius and ulna; d: femur; e: tibia
URL http://journals.openedition.org/palethnologie/docannexe/image/4089/img-67.jpg
Fichier image/jpeg, 157k
Titre Table 34 - Number of long, intermediate and short cutmarks by skeletal element and side
URL http://journals.openedition.org/palethnologie/docannexe/image/4089/img-68.jpg
Fichier image/jpeg, 102k
Titre Figure 167 - Relative frequency of cutmarks per side, according to length
Légende a: scapula; b: humerus; c: radius and ulna; d: femur; e: tibia
URL http://journals.openedition.org/palethnologie/docannexe/image/4089/img-69.jpg
Fichier image/jpeg, 148k
Titre Figure 168 - Comparison of the orientation and length of cutmarks from defleshing according to different available references
URL http://journals.openedition.org/palethnologie/docannexe/image/4089/img-70.jpg
Fichier image/jpeg, 68k
Titre Figure 169 - Different characteristics and categories of traces considered during microscopic analysis
Légende The multiple traces presented in the photograph were produced in a single instance of bone / tool contact, during phase 1 of the project, the test experiment
Crédits Photographs: A. Val ; DAO : M. Coutureau
URL http://journals.openedition.org/palethnologie/docannexe/image/4089/img-71.jpg
Fichier image/jpeg, 219k
Titre Table 39 - Morphological characteristics of cutmarks according to the type of flake or tool used
Légende Data provided in percentages for the observation criteria 6 to 11 (cont. = continuous, discont. = discontinuous, str. = striations)
URL http://journals.openedition.org/palethnologie/docannexe/image/4089/img-72.jpg
Fichier image/jpeg, 124k
Titre Figure 170 - Two groupings of multiple traces characteristic of the cutmarks produced by denticulates
Crédits Photographs: A. Val; CAD: M. Coutureau
URL http://journals.openedition.org/palethnologie/docannexe/image/4089/img-73.jpg
Fichier image/jpeg, 54k
Haut de page

Pour citer cet article

Référence électronique

Sandrine Costamagno, Marie-Cécile Soulier, Aurore Val et Sébastian Chong, « The reference collection of cutmarks »Palethnologie [En ligne], 10 | 2019, mis en ligne le 01 novembre 2019, consulté le 18 mars 2024. URL : http://journals.openedition.org/palethnologie/4089 ; DOI : https://doi.org/10.4000/palethnologie.4089

Haut de page

Auteurs

Sandrine Costamagno

University Toulouse Jean Jaurès / CNRS, UMR 5608 – Traces
costamag[at]univ-tlse2.fr

Articles du même auteur

Marie-Cécile Soulier

University Toulouse Jean Jaurès / CNRS, UMR 5608 – Traces
mariecsoulier[at]gmail.com

Articles du même auteur

Aurore Val

Abteilung für Ältere Urgeschichte und Quartärökologie Department – Universität Tübingen / Evolutionary Studies Institute / University of the Witwatersrand
aurore_val[at]yahoo.com

Articles du même auteur

Sébastian Chong

Haut de page

Droits d’auteur

CC-BY-NC-ND-4.0

Le texte seul est utilisable sous licence CC BY-NC-ND 4.0. Les autres éléments (illustrations, fichiers annexes importés) sont « Tous droits réservés », sauf mention contraire.

Haut de page
Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search