Navigation – Plan du site

AccueilNuméros10PARTIE I - Le référentiel expérim...Stone tool reference collection

PARTIE I - Le référentiel expérimental

Stone tool reference collection

Émilie Claud, Céline Thiébaut, Aude Coudenneau, Marianne Deschamps, Vincent Mourre, Michel Brenet, Maria Gema Chacón‑Navarro, David Colonge, Cristina Lemorini, Serge Maury, Christian Servelle et Flavia Venditti
Cet article est une traduction de :
Le référentiel des outils lithiques [fr]

Plan

Haut de page

Texte intégral

1 – Introduction

C.Thiébaut., É. Claud, A. Coudenneau, M. Deschamps, V. Mourre

1In order to address the questions raised in the introduction (see also Part I, chapter 1.1), we focused part of our research on identifying how different tools types considered typical of one of Bordes’ facies or a particular techno-complex (Delagnes et al., 2007) functioned.

2Bifaces are commonly described as multi-functional tools (see Part I, chapter 2.10) despite the relatively limited number of available use-wear analyses of what often concerns small samples. This being the case, it was necessary to test this hypothesis with new techno-functional data. Similarly, Middle Palaeolithic notched tools are generally associated with woodworking and have not been the subject of any systematic experimentation, with the available functional analyses concerning only a limited number of artefacts (see Part I, chapter 2.7). Apart from rare exceptions, Mousterian flake cleavers, typical of the Vasconian, are known uniquely from the Franco-Cantabrian region. While it could be assumed that the robust active edge of these tools would serve for heavy-duty activities, to date no experimental work has confirmed this hypothesis and very little use-wear analysis has been carried out (see Part I, chapter 2.9). Our goal was therefore to document the potential uses of these tools in order to evaluate the oft-suggested cultural influences underlying the manufacture of this tool type. Finally, we also focused on better understanding the function of Middle Palaeolithic points whose morphological and technological variability could potentially reflect equally diverse functions (see Part I, chapter 2.8), particularly the use of points to arm hunting weapons and its wider implications for Middle Palaeolithic subsistence strategies.

3In terms of other tool types (e.g. side scrapers, unmodified flakes previously available reference collections built by several participants (Lemorini, 2000; Coudenneau, 2004; Claud, 2008) served as a basis for interpreting the archaeological material. For this reason, experiments specifically designed to investigate the use or analysis of these artefact types was unnecessary.

4By better understanding the function of retouched Mousterian tools we hope to produce a clearer appreciation of the reality of Middle Palaeolithic facies, technological groups, or techno-complexes defined and used by a majority of prehistorians.

5A second, more transversal, aspect of this research concerns the relationship between tool function and raw materials. The use of so-called “mediocre” raw materials, such as quartz and quartzite, is relatively frequent during the Middle Palaeolithic, with technological analyses now systematically including these materials. Certain of these studies have shown that debitage methods were specifically adapted to these materials, leading to the production of blanks with different characteristic forms (Jaubert, Mourre, 1996; Thiébaut et al., 2009b). A widely held view is that quartz and quartzites were used only as substitutes for flint. Moreover, use-wear analyses in Middle Palaeolithic contexts have most often focused uniquely on flint with other raw materials included only recently (Marquez-Mora, Preysler 2002; Borel, 2008; Cristiani et al., 2009; Derndarsky, 2006; Plisson, 2008; Gibaja et al., 2009; Clemente-Conte et al., 2012; Lazuén, 2012a, 2012b; Lelouvier et al., 2012; Cura et al., 2014; Daffara et al., 2014; Venditti, 2014; Lemorini in Jaubert et al., in prep.; Annex 1). In order to test this hypothesis and evaluate to what extent blank type is connected to function, it was necessary to build experimental reference collections and develop a use-wear approach adapted specifically to these types of raw materials (see Part I, chapter 2.6).

2 - Use-wear analysis of stone tools: history and methodology

É. Claud

A - The birth of the discipline

6In the 1930s, the Russian prehistorian S. A. Semenov founded a new discipline, use-wear analysis, whose goal was to study traces of use on archaeological objects. This pioneering archaeologist was interested in traces with both “active” (use‑related) and “passive” (connected to manufacture but which could also result from prehension, transport or handling) origins and demonstrated that objects made from a variety of raw materials (e.g. flint, obsidian, bone, wood, shell, metal) could bear diagnostic traces of use. He developed the first genuine methodology for directly addressing the function of prehistoric tools using a stereomicroscope (low magnification) and microscope (high magnification) to evaluate striations, rounding, scarring or polish on the active areas of tools. By devising the first systematic experimental tests and thanks to the use of high-powered microscopy, he was able to compare traces on archaeological tools with those on experimental replicas and thus deduce the movement and type of material worked (Semenov, 1964, 2005). Alongside S. A Semenov, other Russian researchers helped develop the discipline from the middle of the 20th century onwards: Korobkova, Shchelinskij, Filippov and Skakun (see Levitt, 1979; Plisson, 1988a, 1988b; Skakun, 1992; Korobkova, 1993; Shchelinskij, 1993). Semenov’s doctoral dissertation, Prehistoric Technology, written in Russian, was only translated into English and made available to non-Russian speakers in 1964. The translation of his dissertation was an awakening for prehistorians interested in functional analyses. Between 1964 and 1980, following the occasional attempt to incorporate use-wear traces (Keller, 1966; Wilmsen, 1968; Kantman, 1970a, 1971, Lenoir, 1971; Nance, 1971; Rosenfeld, 1971), multiple methods of functional analysis inspired by the work of Semenov were developed, tested and applied. These new methods integrated a larger body of experimental work and the use of low- and high-powered microscopy.

7From 1974, Tringham processing, also inspired by the work of Semenov, proposed a low power approach using a stereomicroscope (or binocular loupe) at low magnification (10 to 80 ×) to identify the formation of microscopic scarring or microflaking edge‑rounding and striations, and deduce the direction of movement and hardness of the material work (Tringham et al., 1974). Other researchers, including Odell (1975, 1981), Lawrence (1979), Cotterell and Kamminga (1979), Odell and Odell-Vereecken (1980), Roy (1982), Shea (1987, 1988) and Akoshima (1987), further tested and developed this approach.

8At the same time, Keeley developed a high-powered approach that employed a metallographic microscope and 100 to 400 × magnification with a reflected light source (Keeley, 1974, 1976, 1980). This approach allowed micro-traces (micro-polishes and micro-striations) to be observed, which also provided evidence of the direction of movement and the type of material worked (e.g. bone, wood, meat, skins). Several blind-tests would confirm the reliability of the method (Keeley, Newcomer, 1977; Newcomer, Keeley, 1979; Gendel, Pirnay, 1982), leading a large community of researchers to adopt it, including Vaughan (1981a, 1981b), Anderson-Gerfaud (1981), Plisson (1982a, 1982b, 1985, 1986), Moss (1983a; Moss, Newcomer, 1982) and Beyries (Beyries, Boëda, 1983; Beyries, 1987a, 1988b, 1993b).

9Finally, in 1980, building on the high-power approach, Anderson‑Gerfaud introduced the “very high power approach” that utilised a scanning electron microscope (SEM) and magnifications up to 10 000 ×, to identify the materials worked (collagen, animal tissue, tendons, hairs, blood, plant fibres, pollen, grains, phytoliths) based on the morphology of micro-traces combined with the chemical composition of residues (sampled with a micro-probe).

B - Choice of observation methods

10The coexistence of the low- and high-power approaches and differences in time investment and the accuracy of interpretations, particularly in regards the materials worked (hardness versus nature), led to significant methodological debate and disagreements between practitioners of these two approaches. In fact, partisans of the low-power approach, argued, on one hand, that macroscopic edge-damage (i.e. scarring and rounding) was not always present and difficult to distinguish from natural, accidental or technologically induced traces (Keeley, 1976; Hayden, Kamminga, 1979; Moss, 1983b; Vaughan, 1981a; Mansur-Franchomme, 1986), and therefore should be discounted as the sole criterion for determining tool function. On the other hand, macro-traces were only considered diagnostic in specific cases, such as projectile damage or edge-rounding due to the scraping of abrasive materials (Anderson-Gerfaud et al., 1987). In turn, the notion that micro‑polishes were diagnostic of specific contact materials was equally questioned, notably due to the multiple factors that could lead to their formation and the morphological resemblance of polishes resulting from use on different materials (Odell, Odell‑Vereecken, 1980; Holley, Del Bene, 1981; Grace et al., 1985; Newcomer et al., 1986; Unger-Hamilton, 1989; Levi Sala, 1989; Rees et al., 1991).

11These debates appeared to be fuelled by problems of definition in part due to the fact the characteristics of particular traces vary between magnifications and observation techniques and are influenced by differences in observer experience, sample treatment (cleaning), the nature of reference collections and, finally, the preservation of archaeological material (d’Errico, 1985; Vaughan, 1985; Newcomer et al., 1986; Mansur‑Franchomme, 1986; Moss, 1983b, 1986; Plisson, 1988c; Plisson, van Gijn, 1989). In fact, since the 1990s, most use-wear research employs a combination of low and high (even very high) power approaches in order to integrate complementary information with different degrees of precision depending on the material worked (e.g. Philibert, 1994; Gassin, 1996; Longo et al., 1997; Geneste, Plisson, 1996; Texier et al., 1996; Fagnart, Plisson, 1997; Plisson, Beyries, 1998; Bosquet, Jardon-Giner, 1999; Lemorini, 2000; Plisson, 2000; Beugnier, Plisson, 2004; Caspar in Locht et al., 2002; Gibaja Bao, 2007; Plisson, 2007; Rots, 2010).

12In the framework of this collective research project, we combined two scales of magnification in a complementary manner. When preservation conditions permitted, objects were analysed using both low (stereomicroscope) and high magnification (metallographic microscope) in order to identify all relevant functional evidence. For quartzites (Part I, chapter 2.6), a metallographic microscope equipped with a DIC system (Differential Interferential Contrast, Plisson, 2000) was used to overcome problems due to reflections from crystals. In addition to the Coudoulous assemblage, several experimental quartz / quartzite points were analysed using a metallographic microscope equipped with a interferometric system (observations by F. Venditti and C. Lemorini, see Part I, chapter 2.8).

13For these experimental reference collections, we focused on documenting macroscopic edge- damage (i.e. macro-traces) for several reasons:

  • macro-traces are generally more robust than micro-traces meaning that they have a higher chance of being identified in relatively old industries that often suffer from problems of poor preservation (Beyries, 1990);

  • macro-traces have been shown to be highly dependent on tool morphology, notably the presence of retouch and edge angle, which is why we constructed specific reference collection for different tools (Tringham et al., 1974; Gonzáles Urquijo, Ibáñez Estévez, 1994; Trias, 2002). For example, prior to our work, the influence of retouch on the development of scarring had not been systematically explored;

  • damage on the edges of quartzite or ophite tools has rarely been documented (although see Cristiani et al., 2009; Lelouvier et al., 2012; Cura et al., 2014), and a recent study has shown macro-traces on certain quartzites to be unreliable (Gibaja et al., 2009). As the hardness and degree of cohesion between grains varies between types of quartzite and can condition edge wear (Leipus, Mansur in Gibaja et al., 2009), we chose to explore the interpretive potential of macro-trace development on the specific quartzites found in the archaeological assemblages studied as part of the PCR.

14Use-induced micro-traces were nevertheless described in detail for the experimental collection of quartzite flakes. Although micro-traces identified on experimental quartzite assemblages from the Iberian Peninsula are available in several previous publications (de Araújo Igreja, 2009; Gibaja et al., 2009; Clemente-Conte, Gibaja Bao, 2009; Aranda et al., 2014), we needed to assess whether the quartzites in the studied archaeological assemblages, notably those from the Aure Valley and the terraces of the Garonne River, lent themselves to use-wear analysis. As with macro-traces, it appears that certain quartzites preserve diagnostic micro-traces while others do not (Beyries, 1982; Plisson, 1986).

3 - Experimental protocoles

C.Thiébaut, É. Claud

15Different blanks were produced by experienced knappers (V. Mourre, S. Maury, D. Colonge, M. Brenet) who attempted to reproduce the morphological characteristics of the archaeological pieces as accurately as possible using the same debitage, retouch, and shaping methods.

  • 1 This resin is perfectly adapted to producing highly-accurate moulds of macro-traces. However, while (...)

16Each piece was photographed and imprints and moulds of the active zones were made before use in order to document their initial form and make comparisons following use. Impressions were taken with the silicone addition dental epoxy Provil Novo Light produced by Heraeus Kulzer (D’Errico, 1988; D’Errico et al., 2004; Ollé, 2003; Dubreuil, 2004). This impression medium is extremely precise, quick to prepare and, when combined with a Pascal Rosier reactive polyurethane resin, produces very fluid accurate (less than 0.4 % shrinkage) and resistant moulds1 that allow use-wear traces to be identified and distinguished from other alterations (small retouch negatives or pre-use modifications).

17During experimentation, a standardised recording form was used for each tool to note different types of information, the material worked and the type of activity performed (figure 21), with the different motions used, the materials that came into contact with the edge, the time the object was used in each stage and any potential constraints recorded in detail on the back. A specific form was used for experiments with propelled points (figure 22).

18Multiple criteria were used to describe the active zones of the experimental working edges both on the forms and in the database: location, length, cutting edge angle, morphology (in plan, profile, and section), and retouch characteristics (position, morphology, inclination, extent and distribution).

19Prior to all experimentation, a common terminology was developed in order that each experimenter and analyst used a precise and identical vocabulary (see Part I, chapter 2.4).

Figure 21 - Example of the data collection form for stone tools used during the experiments (excluding spear-points)

Figure 21 - Example of the data collection form for stone tools used during the experiments (excluding spear-points)

Figure 22 - Example of the data collection form for objects used as thrusting or projectile point

Figure 22 - Example of the data collection form for objects used as thrusting or projectile point

CAD: A. Coudenneau

4 - Use-wear and tool function: definitions and descriptive terminology

É. Claud, C.Thiébaut, A. Coudenneau

A - Functional terminology

  • 2 Certain authors use the terms ‘activity’ to designate the mode of use (van Gijn, 1989b) or action ( (...)

20Reconstructing how a tool was used is the first interpretive stage in any functional analysis and consists in determining the material worked (nature, hardness and degree of humidity as well as any potential additives), the movement employed, and prehension mode (hand‑held or hafted). A consideration of active edge morphology and use-wear traces can sometimes identify the types of activities2 carried out on a site (e.g. butchery, hide, wood or boneworking, collecting grasses) or even the precise step in the chaîne opératoire of a particular task (e.g. defleshing hides).

21Subsequently identifying the function of a tool is more difficult and presupposes an under-standing of all its potential purposes (Sigaut, 1991), which includes how it was used as well as its technical, symbolic and social functions (Faulkner in Hayden, 1979b: 61; Plisson, 1993).

B - Tool Function

a - The material worked

22While micro-polishes provide evidence for the precise nature (e.g. meat, skin, wood) of the material worked, macroscopic use-wear sheds light on its relative hardness, which can be determined by its deformation potential and the intensity of damage produced during use. Here we have included meat, skin (fresh as well as dry but still supple skins), grasses and loose earth in the category of soft materials, plant material and rigid, dry skins in the medium-hard category, and bone, dry deer antler, heat treated wood, ivory, stone, mineral pigments and shells in the hard material category. Intermediate categories, such as “soft to medium-hard”, can equally be used to integrate wear due to contact with materials of varying hardness. This would be the case in butchery, where edges come into contact with soft materials (flesh and skins), but often touch harder elements such as cartilage or bone, which produce heterogeneous damage typical of this activity.

b - Actions

23Type of action refers to the movement of a tool in relation to the material worked. Working in the context of Mousterian stone tool assemblages, we focused on actions performed with the straight or pointed edges of tools and relatively flat surfaces, excluding burins and trihedral edges (Gutierrez Saez, 1993).

  • 3 We prefer the terms continuous contact and percussive motions to fixed percussion (grinding) or per (...)

24Movements can first be characterised by the type of motion3 employed (figure 23), which conditions how the force is exerted (Gutierrez Saez, 1993):

  • continuous contact (pressure): prolonged force exerted during continuous contact between the tool and the material worked;

  • percussive motions:

    • direct percussion: instantaneous force applied during discontinuous contact between the tool and the contact material;

    • indirect percussion: instantaneous force exerted during continuous contact between the tool and the contact material, implying the use of an intermediate piece and a percussion tool.

Figure 23 - Terminology employed: gestures

Figure 23 - Terminology employed: gestures

CAD: C. Thiébaut

25The type of contact (figure 24) between the tool and the worked material can be linear (the edge), punctiform (point), or diffuse (surface).

  • linear contacts involve contact of the cutting edge on the worked material;

  • diffuse contacts are different by the application of the edge and a part of the ventral or the dorsal surface of the blank close to the edge on the worked material;

  • punctiform contacts involve a limited contact of the tip of the tool.

Figure 24 - Terminology employed: types of contact

Figure 24 - Terminology employed: types of contact

CAD : C. Thiébaut

26We distinguished several direction of movement in relation to the worked material:

  • longitudinal in the case of motions parallel to the edge;

  • transversal motions perpendicular to the edge;

  • and rotational motions.

27The user will employ multiple tool movements that combine the position of the tool on the material (for example, perpendicular or oblique in the case of a tangential cutting motion) and a non-linear, lever or rotational trajectory (figure 25).

Figure 25 - Terminology employed: movements

Figure 25 - Terminology employed: movements

CAD: C. Thiébaut

28The direction of use can correspond to several different actions, each varying according to orientation of the movement and the angles between the two surfaces of the tool and the contact worked (Gutierrez Saez, 1993). The leading edge angle is formed by the intersection of the worked surface with the leading edge of the tool, while the trailing edge angle describes the intersection between the adjacent surface and the material (Gassin, 1996; figure 26).

Figure 26 - Position of the face, the trailing face and the associated angles

Figure 26 - Position of the face, the trailing face and the associated angles

Modified from Pasquini, 2003-2004

29For example, when used in a longitudinal motion, the tool ‘cuts’ in the case of a unidirectional movement and ‘saws’ when moved in both directions. The working angle (perpendicular or oblique) can be used to distinguish a normal cutting movement from a tangential cutting movement, where only one surface is in close contact with the material. In the case of transverse, unidirectional actions (figure 27), it is possible to distinguish scraping with a positive rake angle (leading edge angle greater than 90°) versus a negative rake angle (leading edge angle less than 90°, Gassin, 1996). Scraping can also be bidirectional, combining both positive and negative rake motions. Perforation implies a unidirectional rotational movement, while piercing involves a bidirectional movement. Finally, we distinguished an oblique (chopping) from perpendicular motion (as with an adze) based on the angle between the working edge and the plane describing the orientation of the movement.

Figure 27 - Value of the leading angle according to the mode of action of an edge applied in continuous, unidirectional, transverse action (positive and negative rake angle)

Figure 27 - Value of the leading angle according to the mode of action of an edge applied in continuous, unidirectional, transverse action (positive and negative rake angle)

Modified from Gassin, Garidel, 1993

c - Prehension mode

30Prehension mode describes both the active lithic element and the entire structure of the tool: hand-held, wrapped or hafted. The latter can be considered “composite” tools, as they comprise a lithic component and a handle that serves as a lever arm.

31Multiple criteria have been advanced to describe wrapping and hafting arrangement, including the nature of the components, the presence / absence and type of bindings or adhesives, and the hafting arrangement based on the combinations proposed by D. Stordeur (1987, figure 28), to which we have added the term “pinch haft” (Anderson‑Gerfaud, Helmer, 1987; Lemorini, 2000) to designate a hafting arrangement where the active part (flint) is inserted within a longitudinally split (partially or completely) handle and fixed with bindings or a combination of bindings and adhesives.

Figure 28 - Existing combinations of the different hafting attributes

Figure 28 - Existing combinations of the different hafting attributes

Fem.: female; jux.: juxtaposed

Modified from Stordeur, 1987

C - Use-wear traces: terminology

a - Macro-traces

32Scarring refers to micro-flake negatives detached by the application of pressure or percussion on the edge of stone tools (Odell, 1975; Cotterell, Kamminga, 1979; Prost, 1989).

33We adopted the terminology proposed by the Ho Ho Classification and Nomenclature Committee Report following the conference in Burnaby, Canada (Hayden, 1979c) as well as terms found in the work of Odell and Odell‑Vereecken (1980), Tringham et al. (1974), Akoshima (1987) and Prost (1989). These terms, employed by a majority of use-wear analysts, include location, position, distribution, disposition, number, morphology, termination, extent, initiation, size, elongation and direction (table 11), to which we have added the number of superimposed scarring episodes.

34Only damage initiated from the pointed active end of a tool (e.g. the distal extremity of a point) with fracture waves that developed perpendicular to and removed the tool edge were designated “burinating removals”.

35Fractures, like scarring, develop by compression and were described using terminology primarily based on the work of Fisher et al. (1984) and O’Farell (2004, 2005) (table 12).

Table 11 - Terminology adopted for the description of scars

Table 11 - Terminology adopted for the description of scars

Illustration modified from Gonzáles Urquijo, Ibáñez Estévez 1994; Coudenneau 2004; Prost 1989; Lawrence 1979

Table 12 - Terminology adopted for the description of fractures

Table 12 - Terminology adopted for the description of fractures

Illustration of fracture removal A. Coudenneau, all others after O’Farrell 1996, modified, and Fischer et al. 1984

36Edge-rounding, although highly informative for certain types of materials, is not often defined in the literature. This term refers to a rounded relief connected to the minute lose of material and a significant reduction in the sharpness of the edge. While edge-rounding with flint generally takes a smooth, irregular aspect, two types of rounding have been identified on quartzites (figure 29):

  • a smooth aspect linked to the minute lose of material or the micro-rounding of crystals;

  • a coarse aspect due to the removal, crushing, or fragmentation of numerous small crystals.

Figure 29 - Types of micro-rounding observable on quartzite

Figure 29 - Types of micro-rounding observable on quartzite

a: scraping fresh lamb hide (hide defleshing); b: scraping of a fresh cow bone (removal of periosteum)

Photographs: É. Claud

37The position and intensity of edge-rounding was recorded during our experiments (table 13).

Table 13 - Terminology adopted for the description of rounding

Position

Intensity

Edge-rounding

- dorsal surface
- ventral surface
- bifacial
- bifacial but concentrated on the ventral surface
- bifacial but concentrated on the dorsal surface Intensity

- none
- weak (discontinuous and limited to the edge)
- average (continuous along the edge)
- heavy (continuous, along the edge, scarring with rounded ridges)

38The term crushing (Hayden, 1979c) designates small repeated losses of material linked to the application of pressure. This type of damage takes the form of concentrated, superimposed scars, whose characteristics (morphology, initiation, termination) cannot be described due to their small size and superimposition. For flint, ‘edge-blunting’ was used to designate a completely crushed and blunted edge with no cutting potential. This blunting has a rugged aspect and irregular topography compared to typical edge-rounding (see above).

b - Micro-traces

39Criteria used to describe micro-polishes on flint were borrowed from Plisson (1985) and those for micro-striations from the work of Mansur-Franchomme (1986). The terminology used to describe micro‑traces observed on quartzites was based on the work of J. F. Gibaja et al. (2009) and I. Clemente-Conte and J. F. Gibaja Bao (2009).

40Polishes refer to the optical effect of a glossy surface due to modifications of the micro-relief (Plisson, 1985). The structure, or coalescence, of the modified micro‑topography can be fluid (the polish follows the relief without affecting it), soft (peaks are slightly altered, with the polished surfaces exhibiting a grainy or smooth aspect) or hard (intense alteration of the peaks, sometimes flattening them and producing convex, flat, undulated or greasy surfaces). Hard polishes are associated with smooth textures while the texture of both fluid and soft polishes is more rugged. Polish density reflects the combination of coalescent and non-coalescent areas within the polish itself, and can be diffuse (greater non-coalescent compared to coalescent areas), intermediate (regular alternation between non-coalescent and coalescent areas) or dense (absent or residual non‑coalescent areas). Coalescent zones can be interrupted by ‘pits’ of various diameters, cracks, and striations. In addition to these features, micro-polishes can be described by their position, distribution, extent, orientation, contour and microtopography as well as the type and density of coalescence (table 14).

41Striations are defined as linear wear traces with multiple potential morphologies (Kamminga, 1979; Keeley, 1980; Mansur-Franchomme, 1986) and can be described according to their position, length, quantity or orientation (table 15).

42Several additional descriptive criteria were recorded for the quartzites. The surface of crystals embedded in the matrix is partially dissolved or eroded during use. In cases where these processes are continuous (or ‘continuous breakage’ Gibaja et al., 2009; Clemente-Conte, Gibaja Bao, 2009), the entire crystalline structure disappears, leading to the development of a continuous polish on this new surface and adjacent matrix. The characteristics of this erosion, the frequency of ‘continuous breakage’, the location of the polish and striations on the matrix and or crystals, as well as the types of scarring evident on the crystals were all taken into consideration when describing micro-traces on quartzites (table 16).

Table 14 - Terminology adopted for the description of micro-polishes

Table 14 - Terminology adopted for the description of micro-polishes

Photographs: É. Claud; illustration modified after Plisson 1985

Table 15 - Terminology for the description of striations

Table 15 - Terminology for the description of striations

Photographs: É. Claud

Table 16 - Supplementary terminologies for the micro-traces observed on quartzite

Table 16 - Supplementary terminologies for the micro-traces observed on quartzite

Photographs: É. Claud

D - Analytical protocoles

43Following the washing of artefacts, sometimes with soapy water, all pieces with persistent residues were soaked in oxygenated water (for animal residues) or acetone (plant residues). Usewear traces were documented macroscopically using a stereomicroscope (12 to 30 × magnification), with a portion of the reference collection examined microscopically, in most cases using a Leica DM2500M metallographic microscope (100 to 500 ×). For the experimental quartz‑quartzite points and the analysis of the Coudoulous assemblage (F. Venditti, C. Lemorini), a metallographic microscope equipped with an interferometric system was used. Three complementary scales of observation were used: fractures and macro-damage were observed with the naked eye, rounding and scarring with a stereomicroscope, and micro-polishes, striations, and micro‑rounding with a metallographic microscope.

44All data recorded for the tools, morphology of active zones, and activities performed were incorporated in an Excel spreadsheet. The active part of each piece was compared to the corresponding mould, with the description of the use-wear traces equally included in the database.

5 - Unmodified flint edges: current ideas in use-wear analysis

É. Claud, A. Coudenneau

45Thanks to considerable work in use-wear analyses, particularly the publication of reference collections that have primarily focused on describing how the unmodified edges of flint behaves, (see Tringham et al., 1974; Hayden, 1979c; Odell, Odell-Vereecken, 1980; Vaughan, 1981a, 1981b; Plisson, 1985; Prost, 1989; Gonzáles Urquijo, Ibáñez Estévez, 1994; Trias, 2002), combined with the individual experience of the use-wear analysts who participated in this research project (see for example Lemorini, 2000; Coudenneau, 2004, 2005; Claud, 2008), we now know how macro- and micro-traces develop on the unmodified edges of flint tools. Here we quickly review the characteristics of these traces, especially the factors influencing their development, before presenting the construction of the different reference collections. Two examples provide a general illustration of micro- and macro‑trace formation as a function of the hardness of the material worked and the types of actions employed (figures 30-31).

A - Macro-traces

46It should first be recalled that, while varying according to the mode of use, multiple criteria are necessary for describing edge-damage, as no single aspect is ever diagnostic.

47The location, distribution, disposition, and orientation of wear is highly correlated with the type of action: a longitudinal motion (cutting, sawing) will produce bifacial, discontinuous, isolated or aligned damage oriented obliquely, or sometimes perpendicularly to the edge whereas transverse motions (scraping) will result in generally unifacial (or asymmetrically organised on the two surfaces), continuous, aligned or superimposed damage oriented perpendicularly (figure 30). On the other hand, the size, quantity, and number of generations in the case of superimposed scars, are conditioned by the hardness of the material worked. All things being equal (notably the type of action), working soft materials, such as meat or skins, produces less scarring compared to medium-hard to hard materials such as bone. Percussion produces larger scars due to the generally increased force involved. Scar morphology is also linked to both the hardness of the contact material and the action used. Although certain trends can be observed, no single criteria is sufficiently diagnostic on its own:

Figure 30 - Overview of the macro-traces observed on flint pieces

Figure 30 - Overview of the macro-traces observed on flint pieces

Photographs: É. Claud and A. Coudenneau; CAD: É. Claud and M. Coutureau

  • working soft or medium-hard materials, especially in a longitudinal motion (although sometimes observable with transverse or percussive motions), generally produces half-moon shaped scars;

  • quandrangular scars are generally associated with scraping medium-hard materials;

  • trapezoidal scars are frequent with hard contact materials;

  • while semi-circular scars are common with all types of worked materials.

48Finally, termination type depends on multiple factors, including hardness, type of action as well as working angle, and is therefore a somewhat unreliable criterion. With that said, cutting soft materials (fresh skins, meat without touching the bone) tends to produce a limited quantity of scars with stepped terminations while longitudinal and percussive motions, notably on medium‑hard and hard materials will sometimes result in scars with transverse terminations, which are rare with transverse motions. Micro‑flake initiations are often bending, with only hard materials (organic, often mineral, or contact with stones embedded in sediments) occasionally producing cone initiations.

49The working of particularly abrasive materials (i.e. skins, soil, bark) results in the development of macroscopically rounded edges, which intensifies both with increased use time and the abundance of abrasive particles either in the material worked (earth, bark, skins) or introduced (ochre, cinders) during use.

50Edge-blunting due to crushing can be observed with the working of hard, naturally abrasive mineral materials, with the degree of rounding increasing with duration of use and the hardness of the contact material.

51Like scarring, the location and orientation of this type of damage depends upon tool movement.

B - Micro-traces

52The observation of different experimental polishes demonstrates that it is possible to determine the mode of use when multiple criteria are considered rather than coalescence alone, which is directly but not exclusively linked to the material worked (Vaughan, 1981a; Plisson, 1985; Mansur-Franchomme, 1986; Plisson, van Gijn, 1989).

53As with scarring, the position and orientation of polish are directly conditioned by the direction of movement (oblique or parallel micro-polishes organised symmetrically on both edge aspects in the case of longitudinal motions, perpendicular or oblique micro-polishes organised asymmetrically on the two edge aspects with transverse motions (figure 31) while the other properties are primarily influenced by the nature of the material worked:

  • extent and distribution vary according the hardness of the material: soft materials, such as fresh hides or meat, conform to and polish the entire micro-topography, while harder materials, like bone, come into limited contact with the tool edge, meaning that only the projecting portions of the edge are affected (figure 31);

  • coalescence, density, contour and brightness are all dependent on the nature and condition of the material worked, particularly humidity and the presence of abrasives;

  • the presence of pits or cracks within the polish is also linked to the nature of the material worked: the first are mainly observed in hideworking, the second with boneworking.

54Polishes reflect the types of use and assume a characteristic (i.e. diagnostic) aspect after only a few minutes of use prior to becoming what has been described as a generic weak polish (Vaughan, 1981a; Plisson, van Gijn, 1989). Beyond this stage, duration of use, like the degree of force exerted, affects only the extent and density of the polish.

Figure 31 - Overview of the micro-traces observed on flint pieces

Figure 31 - Overview of the micro-traces observed on flint pieces

Photographs: É. Claud and A. Coudenneau; CAD: É. Claud and M. Coutureau

55However, certain actions result in the development of very little micro-polish due to intense scarring and or very limited contact with the material, which is the case for piercing and direct and indirect percussion on medium-hard to hard material. The experimental pieces generally exhibited rare or weakly developed polishes that were seldom indicative of the material worked and often took the form of linear striations or brightspots on the highest points of the edge. More-over, the presence of flint chips embedded in the material often led to the development of shiny micro-polishes with a hard coalescence and a flat or slightly domed micro‑topography connected to the tool rubbing against micro-fragments detached during use.

56Striations observed on experimental pieces used in different ways are often associated with micro-polishes. Tool movement conditions the orientation and position of striations, while their morphology has been shown by Mansur-Franchomme (1986) to vary as the flint’s surface evolves during use (i.e. changes in the relative fluidity of the silica gel). The surface condition of flint artefacts varies according to the nature and humidity of the contact material as well as the duration of use. Working fresh hides and cervid antler, meat processing and digging have all been shown to induce rough striations that are rare and often difficult to detect, apart from digging, which produces frequent rough-bottomed striations associated with smooth-bottomed ones. Woodworking produces smooth or multiple additive striations while working dried skins and grasses produces, respectively, smooth or a combination of abraded striations and filled‑in striations. Finally, direct and indirect percussion on medium-hard and hard organic materials often induce multiple, hard striations with flat or convex topographies that are likely due to friction between the tool’s surface and micro-fragments detached during use.

57Moreover, striation depth varies with applied force while striation width is dependent on the size of the abrasive particles, whether derived from the immediate environment (dust, soils), the worked material, additives, or micro-fragments detached from the tool during use. These two criteria equally vary according to the nature of the worked material (Mansur-Franchomme, 1986).

58Both polishes and striations can be associated with micro‑rounding, which, in rare cases, can be uniquely present (scraping or cutting dried skin without an abrasive). Micro-rounding is consistently present on pieces with blunted edges, namely those used to work abrasive material (dried skins, bark, digging), while the formation of ‘domed’ areas associated with crushing generally do not correspond to microscopic edge-rounding, but rather a greater loss of material in form of superimposed micro-scarring. Micro-rounding can also be observed on pieces exhibiting no visible macroscopic rounding due to a short duration of use or the working of softer, less-abrasive materials (fresh skins, tendons).

59The intensity and position of the rounding varies as a function of the abrasiveness of the contact material and or the addition of an abrasive agent (e.g. ochre, cinders), as well as the duration of use, direction of movement, and the angle between the contact material and the tool.

6 - Influence of raw material on use-wear traces: the quartzite flakes reference collection

É. Claud, V. Mourre, C.Thiébaut, M. Deschamps, A. Coudenneau, D. Colonge, M.-G. Chacón-Navarro

A - Why such a reference collection?

60Since the emergence of Prehistory as a discipline, most researchers studying the lithic industries of Western Europe have tended to put aside or neglect those done at the expense of materials other than flint, especially quartz and quartzite. In a regional context in which large sedimentary basins rich in good quality flint dominate, widely used by Palaeolithic populations, prehistorians seem to consider that using flint would be a norm while the use of other so-called “substitute” materials would only be last resort. The embarrassment they feel when facing other materials can be seen notably in the multiple attempts to artificially group the mentioned industries under a common name: for example “non-flint stone tools” (MacRae, Moloney, 1988) or more recently “non-flint raw material”, clumsily translated in French (“matières premières lithiques alternatives”) (Sternke et al., 2009). The contradictory and erroneous expression “non-siliceous materials” is also regularly found to designate quartz and quartzite, nearly exclusively made of silica (Mourre, 1996, 1997). It has been proposed to name these materials by the expression “materials known as mediocre” to emphasize the abyss separating their perception by prehistorians and the major role they played, however, for the Palaeolithic human groups in Europe (Mourre, 2003b: 259).

61The difficulty to name materials other than flint is connected to a difficulty to study and interpret them. In France, “knapped stone technology” remains too often “knapped flint technology”, to such an extent that “flint is often used as synonymous of “lithic remain” by students, even by some researchers. If this factual situation is easily explained by the omnipresence of flint in regions where the discipline emerged and developed and where, as early as the end of the 19th century, the chronocultural sequences that we are still broadly using today were devised, it must not for all that be accepted as such. The low investment concerning the analysis of quartz and quartzite remains is not the privilege of technological studies: it is also found in the context of functional analyses that long remained restricted to flint blanks.

62It must be specified that the functional analysis of quartz or quartzite remains faces non-negligible methodological and technical difficulties The main difficulties are linked to the structure of the fracture surfaces and to the high reflectivity of the quartz crystals observed with high magnification Based on systematic experimental research, pioneering work in this field was carried out by K. Knutson (1988a, 1988b) and S. Sussman (1985, 1988a, 1988b). They were the first to carry out functional analyses on quartz assemblages by optical microscopy and scanning electron microscope (SEM). Although the SEM solves all the optical problems concerning quartz reflection, it is nonetheless a costly equipment. K. Knutson developed new less expensive observation techniques such as the analysis with metallographic microscope with interferometry in reflecting light (Nomarski system or DIC: Differential Interferential Contrast). This system limits surface reflection and amplifies depth differences, which allows having a very clear vision of the quartz topography, and thus easily recognizing use-wear traces (Plisson, 2000; de Araújo Igreja, 2009).

63The confocal system is another system that can be applied to a metallographic microscope and that gives the same results as the SEM. It is a laser scanning of sections of the studied sample. With the help of a software, a detailed image of the analyzed surface is reproduced. Although less costly than the SEM, it also requires a significant financial investment and it is not really easier to use for a use-wear study laboratory. The application of the confocal microscope to the use-wear traces analysis of quartz objects is recent. M. Derndasky, from the Institut für Ur‑und Frühgeschichte of Vienna (Austria) started using the confocal system with this purpose in 1999 (Derndasky, 2009; Derndasky, Ocklind, 2001).

64After a long neglect, functional studies of quartz and quartzite Palaeolithic industries have recently met a significant development (see especially Gibaja et al., 2002; Cristiani et al., 2009; Lombard, 2011; Clemente-Conte et al., 2014; Cura et al., 2014; Lemorini et al., 2014; Márquez et al., 2016; Ollé et al., 2016; de Lombera-Hermida, Rodríguez-Rellán, 2016).

65The archeological corpus studied in the context of the Research Program includes sites with industries on various raw materials: flint but also mostly quartz and quartzite (see especially Mauran, Coudoulous, Les Fieux layer K and Grotte du Noisetier, Abri Olha Olha, Gatzarria, Isturitz). It was absolutely impossible to propose a global functional interpretation of the studied assemblages while keeping aside an always significant part of the objects, truly “carrying sense”, whether or not it is majority. In the context of the experimental sessions, we constituted a reference collection with materials collected as close as possible to the sites, in order to characterize use-wear traces proper to the quartzite unretouched flakes, depending on the use mode of the object. Our goal was to determine if the types of quartzite used at the studied sites would record macro and microscopic traces that could serve as a basis for functional studies. Once this informative potential confirmed, the drawn reading keys should allow us a comparative study between the experimental material and the archeological material.

B - Used tools and activities carried out

66Various quartzite tools were used in the context of varied activities (see Part I, chapters 1.2 and 1.3). This chapter is concerned only with unretouched flakes, as experiments and use-wear traces on the other tool types are described in the following chapters (denticulates in Part I, chapter 2.7, points in chapter 2.8 and flake cleavers in chapters 2.9). Some experiments were done with unretouched schist flakes, but their restricted numbers does not allow as precise and complete results as for quartzite to be transferable to the archaeological register. Besides, we realized, after carrying out several experiments, that we had used two types of schist whose microscopic aspect differed, which complicated the observations and comparisons between objects.

6765 unretouched flakes were used, four with two active areas, which represent a total of 69 active areas. The flakes, for most flakes with cortical back and pseudo‑Levallois points, come from the debitage of blocks collected in the alluvium of the Neste or the Garonne Rivers. Following the observations done on coarse grained quartzite denticulates (absence of macro-traces, see Part I, chapter 2.7), we have, as early as 2011 and in order to constitute the reference collection, preferred using fine or medium grained quartzite, which corresponded, moreover, to the quartzite types mostly found on the sites of Mauran and Grotte du Noisetier.

68These unretouched blanks were used in the context of various activities belonging to the chaînes opératoires of acquiring and processing plant and animal materials (see Part I, chapters 1.2 and 1.3). Only uses with continuous contact (non-percussive) were carried out. Indeed, the traces connected to percussive gestures are documented through the use of quartzite or ophite flake cleavers in percussion, for acquiring wood and treating carcasses (Part I, chapter 2.9).

69Except for two small pseudo-Levallois points used for defleshing the bison and hafted in a wood pinch haft, all the flakes were held with bare hands. No macro-trace connected to prehension or hafting has been detected, and we did not look for these traces at a microscopic scale.

a - Acquiring and working wood

70Whether for the phase of acquisition or processing, 14 items were used mostly to work hazel wood (n=9), very common at Grotte du Noisetier (Théry-Parisot in Mourre et al., forthcoming). Hazel wood was mostly worked green in:

  • longitudinal actions (sawing), for acquiring branches (n=3), with durations of use varying between 20 and 41 minutes;

  • transverse actions (linear contact), to make hunting spears;

  • and rotational actions, to drill a hazel wood spear.

71A green acacia branch was also sawn, a green ivy branch was barked and a dried pine spear was shaped.

72The durations of use vary much, between 10 and 101 minutes.

73Scraping actions were globally carried out without difficulty, except for the dirtying of cutting edges that required the experimenters to interrupt regularly their activity to remove, even approximately, wood fibers adhering to the tool and reducing the sharpness of the edge. At the end of the experiment, one item used for 67 minutes to shape a dry pine spear showed a decrease in efficiency.

74On the other hand, sawing experiments were laborious and the goal could not be achieved because of a lack of efficiency of the flakes. As for drilling, the flake was efficient at first but had to be abandoned after 29 minutes as its morphology did not allow digging sufficiently deeper into the thickness of the spear.

b - Butchery working

75Various animal species were used for food purpose: sheep, deer, cow and bison. The various steps of the butchery chaîne opératoire, except for bone fracturing, were implemented, using 33 blanks.

76The cutting actions are the most numerous. In total, 14 blanks allowed skinning various carcasses (red deer and bison). Amongst them, eight were dedicated only to this activity. The others were used afterwards to deflesh disarticulate and remove tendons. The blanks used only to deflesh are four and those that were used only to disarticulate are eight (cervid and bison). Three blanks were used at the same time for defleshing, disarticulating and tendon removal. Except for three items used for a very short time (one and two minutes), the durations of use of the flakes vary between 14 and 109 minutes, with an average of 37 minutes.

77The hafting of pseudo-Levallois points in pinch hafts (straight handle of green wood simply split at one end) seemed to us to be a good alternative to their use with bare hands, due to the small dimensions of these items and to the rapidity of making such handles.

78Except for the need to frequently clean the cutting edges to keep them at their top sharpness – flesh residues quickly fill the hollows between the grains and stick to the cutting edge –, the flakes allowed us to achieve the desired objectives. However, a noticeable decrease in efficiency was noted at the end of the experiments for five items that were used for skinning (for 29, 43, 47, 57 and 62 minutes) and six items that were used for disarticulation (for 10, 17, 19, 29, 93 and 109 minutes).

79Several items were used in a transverse action (scraping) to prepare cow, sheep or horse bones fracturing to recover the marrow (8 flakes, used for between 10 and 20 minutes).

c - Hide working

80The blanks intended for hide working – they were 13 – were used at various steps of the chaîne opératoire, from the defleshing of fresh hide up to the cutting of dry hide. The defleshing activity, carried out mostly on fresh and re‑humidified bison hide, was carried out with nine blanks, seven having worked in tangential cutting and two in scraping, for a duration varying between 6 and 107 minutes. Transverse actions proved little adapted, as scraping only removed very little flesh from the grain, as was evidenced for flint flakes or side scrapers. As for flakes used in tangential cutting, they did allow reaching the defleshing objective, but the fact that their cutting edge got dirty and needed a regular cleaning somewhat slowed the operation.

81Two flakes were used to cut up strips of dried humidified red deer and bison hides, during 22 and 24 minutes, without any unacceptable difficulty being encountered, except that the thick and rigid hide of the bison made its cutting less easy.

82An item was used for about 40 minutes to shave a dry red deer hide with a transverse action. The cutting edge, relatively long and with convex delineation, was well adapted to this activity, and the work could be done efficiently as long as the hide was stretched properly.

83Finally, a last item was used to pierce a dry deer hide during 8 minutes. The point proved too large and the section little adapted (insufficiently resistant trihedral) to be really adapted to this activity, even if the piercing could be carried out.

d - Working other materials

84In order to complete the reference collection we have, more anecdotally, used quartzite blanks for horn working in longitudinal action (n=2, for 8 and 14 minutes), to dig soil (n=1, for 17 minutes) and for harvesting wild grasses (n=1, for 50 minutes). Dry bone was also worked, although for the time period considered this material is slightly anachronistic. These experiments were carried out with a view to comparison with traces resulting from woodworking (see Part I, chapter 1.3). Dry bone scraping was done with both active parts of the same blank during 8 and 10 minutes. A flake was used to incise dry bone (for 10 minutes), a second one to incise a dry bone (for 12 minutes). Finally, two more flakes were used to extract a round item in a shoulder bone following a punctiform contact with longitudinal action (2 minutes, abandoned after fracturing and replaced by a flake used for 51 minutes).

C - Description of the use-wear traces

85Highly detailed descriptions of macro- and microscopic traces according to the various tested modes of use are available as a table presented in Annex 2. We are presenting below an overview of the features of macro-traces, then of the micro-traces.

a - Low magnification observations: macro-traces

86Firstly, it is useful to characterize the state of the cutting edges before use (figure 32). With regards to the unretouched flint cutting edges, the unretouched quartzite cutting edges have a more irregular delineation, connected to the presence of numerous quartz crystals. This unevenness is all the more marked on medium grained quartzite that also show a more uneven surface topography (figure 32e-f). On the latter, the crystals are large in size and the matrix between the grains is rarer.

87Macro‑traces connected to use, visible with a stereomicroscope, are fractures, scars and rounding. Such traces have often been seen on the cutting edges, indicating at this first level clues about the mode of use carried out. Only three items, made of fine grained quartzite, do not show any macroscopic trace or traces too fine to be interpreted as having a functional origin in an archeological context. One of them cut red deer meat during 15 minutes, another skinned and disarticulated a red deer during 32 minutes (but the experimenter had prehension difficulties with this tool, considered as little efficient) and a last one scraped green ivy wood (stripping the bark) during ten minutes, the angle of the edge used being widely opened (70°). In the same way as on flint tools, the working of very soft and little abrasive materials – without bone contact in the case of butchery for example – as well as the use of open cutting angles thus reduced the formation of traces at this scale, which is all the more true if the durations of use are short.

88Let us begin with some general remarks on the observation of macro-traces on quartzite. Difficulties in reading the scars have sometimes been met:

  • identifying the precise morphology of the scars is more delicate to establish on quartzite cutting edges than on flint cutting edges, even more if the negative presents a feather termination (and therefore in continuity with the surface of the tool, without rupture). The reading is made difficult by the heterogeneous aspect of the surface except in the case of cortical surfaces and very fine grained quartzite. As an example, woodworking produces, on flint, scars with a bending initiation, which then develops on the surface in a particularly low angled manner, thus removing very little material (like a fracture “languette”) and that sometimes ends without rupture (feather termination). On quartzite, the developing on the surface of this type of scar is sometimes difficult to perceive. Only the semi-abrupt or abrupt feature of the initiation is then visible;

  • the characterization of the initiation of the scars (cone versus bending) is also more difficult to establish especially when those are small in dimensions. When the quartzite grain size is coarse, the initiation area can indeed correspond to the removal or the alteration of one or two grains, which can leave an irregular surface;

  • finally in a general manner, the more the quartzite grain size is fine the more the macroscopic traces are close to the ones visible on flint, and the more their reading is obvious.

Figure 32 - Unmodified edges before use, viewed at different levels of magnification

Figure 32 - Unmodified edges before use, viewed at different levels of magnification

a-b: flint; c-d: fine-grained quartzite (alluvium of the Neste River); e-f: medium-grained quartzite (alluvium of the Garonne River)

Photographs: É. Claud

89Except for the difficulties that do not always allow characterizing precisely all the morphological aspects of each scar like on flint, we were able to note that the quartzite cutting edges globally follow the same general rules of use-wear than those in flint, except for some specificities that we are presenting below.

90Like for flint, the motion done influences the position, the distribution and the orientation of the scars. The scars tend to be unifacial, continuous and oriented perpendicularly in the case of scraping for example, while they are rather bifacial, discontinuous and oblique in the case of a longitudinal action (cutting, sawing, figures 33‑42). During drilling, the scars are distributed on the point, on both converging cutting edges or even on the mesial ridge in the case of a trihedral section of the active area (figure 34e-f, 35g-i, 36g-i, for a comparison with the borer used to drill bone). The hardness of the materials also influences the development of the macro-traces, especially the dimension and the number of scars (figures 37-42). All parameters being equal (duration of use, cutting angle, …), the working of bone or horn (hard materials) produce more scars than that of fresh hide, meat (soft materials), wood or dry skin (medium-hard materials).

91Similarly, the material worked induces the development of scars with specific morphologies. Like for flint, in the context of butchery work, discontinuous, low angled (shallow), triangular, oblique scars, with bending initiations and step terminations were also observed on the quartzite cutting edges (figures 33f-h, 40-41). As for woodworking, it leaves quadrangular, semi-circular or trapezoidal scars whose initiation, often clearly bending, makes the edge semi-abrupt or abrupt (figure 35a-c,e). On the opposite, for flint the scars are clearly and systematically superimposed (several generations) on the cutting edge that worked on hard materials (especially by scraping but also by drilling or sawing), but in the case of quartzite the superimpositions are rare and more difficult to perceive (figure 36a-c).

92In reality, after the initiation of the scars, the cutting edge seems to wear more by crushing, which will eventually lead to the forming of a blunting, or of a rounding, the latter being more dull and showing a more irregular topography than the rounding classically documented on tools that worked on hide (figure 36d-e). A rounding was also seen on artefacts used for sawing wood (acacia and green hazel wood, figure 35f) while it is absent or very fine on the flint flakes used for woodworking (but present on flint denticulates used for sawing dry wood, see Part I, chapter 2.7). Rounding is also sometimes detected on the active parts that scraped and drilled wood (figure 35d,g-i). The latter is less intense and finer that that resulting from the working of harder materials such as bone.

93On the items used at length for wood scraping (between 20 and 101 minutes) the use-wear polish, well developed under the microscope, is sometimes already perceptible under stereomicroscope in association with the rounding (figure 35d). The distinction between traces connected to the working of a hard material such as bone and a medium-hard material such as wood is thus difficult to make using the number of generations of scars as superimpositions are rare; it has to concentrate mostly on the modifications seen on the cutting edge, that is to say its blunting or rounding: its intensity, its regularity and its possible association with a macro-polish especially in the context of a transverse action. Furthermore, whether the flakes are in flint or quartzite, one has to keep in mind that there is a continuity between the hardness of the worked materials, leading for example to a proximity of the macro-traces left by the working of dry and hard wood (such as acacia), of the hard animal materials such as a weak bone (e.g. the carcass of an immature) or of a soaked horn. This continuity must encourage us to be cautious in the interpretation of the traces that could be considered as “intermediary”. For example, the observation of continuous, unifacial, non-superimposed scars, associated with some crushing creating a slight blunting without macro‑polish will lead to an interpretation lesser in precision (scraping of medium-hard to hard material) than in the case of traces with well-marked features (superimposed scars with step terminations with an irregular strong rounding for bone versus aligned scars with feather termination with regular rounding and associated to a macro-polish for wood).

Figure 33 - Macro-traces observed on the tools that were used on soft materials such as fresh hide (skinning, hide defleshing), meat (defleshing) and soft to medium-hard materials (defleshing, removal of tendons)

Figure 33 - Macro-traces observed on the tools that were used on soft materials such as fresh hide (skinning, hide defleshing), meat (defleshing) and soft to medium-hard materials (defleshing, removal of tendons)

The white lines indicate zones of rounding

Photographs: É. Claud; CAD: C. Thiébaut and M. Coutureau

Figure 34 - Macro-traces resulting from scraping, cutting and perforating dry hide

Figure 34 - Macro-traces resulting from scraping, cutting and perforating dry hide

a, c and d: rounding; b, e, f: semi-circular scars with a flexion initiation. The white lines indicate zones of rounding

Photographs: É. Claud; CAD: C. Thiébaut and M. Coutureau

Figure 35 - Macro-traces related to scraping, sawing and perforation of wood

Figure 35 - Macro-traces related to scraping, sawing and perforation of wood

The white lines indicate zones of rounding

Photographs: É. Claud; CAD: C. Thiébaut and M. Coutureau

Figure 36 - Macro-traces observed on edges used to clean or work hard organic materials (bone, horn)

Figure 36 - Macro-traces observed on edges used to clean or work hard organic materials (bone, horn)

The white lines indicate zones of rounding / blunting; d, e, and i show them in detail

Photographs: É. Claud; CAD: C. Thiébaut and M. Coutureau)

Figure 37 - Schematic illustration of the macro-traces observed on an unmodified flake in quartzite used to scrape wood

Figure 37 - Schematic illustration of the macro-traces observed on an unmodified flake in quartzite used to scrape wood

Photograph and CAD: C. Thiébaut

94Finally, like on flint, hide working leaves little or no scarring but real easily readable rounding (figure 42), especially in the case of scraping: the rounding is intense, glossy and has a fine and regular surface similar to that observed on flint (figures 33d, 34a). Slight rounding can also form at the time of skinning and tendon removal (figure 33b), in parallel to numerous scars.

95Digging soil, to extract tubers for example, also produces very strong macro‑rounding that differ from that produced by hide working by their strong intensity (for a similar duration of use) and their more covering aspect (figure 43c-d).

96Thus the study of the reference collection shows that most utilizations produce traces that can possibly be used in the context of archaeological material study. Macro‑traces, at the minimum, will allow to spot the location of an active area and in the best of cases will be diagnostic of the motion and of the relative hardness of the worked material or even of the activity conducted (e.g. butchery scarring). Part of the activities implying contacts with soft materials (skinning and defleshing with little bone contact, scraping of green and soft wood), especially in the case of short durations of use and open cutting angles are likely not to be detected and recognized as such in the archaeological register, even in well preserved series.

Figure 38 - Schematic illustration of the macro-traces observed on an unmodified flake in quartzite used to skin a carcass

Figure 38 - Schematic illustration of the macro-traces observed on an unmodified flake in quartzite used to skin a carcass

Photograph and CAD: C. Thiébaut

97While cutmarks in the context of butchery, when accidental contacts with bones or cartilage has happened, are especially characteristic and distinguishable from the macro-traces produced by other activities, it seems that the identification of the different stages of the butchery, on the basis of the macro-traces only, is jeopardized. Indeed, the macro-traces produced by skinning, defleshing and disarticulation are extremely close (see Annex 2). The number and morphology of the scars vary according to the frequency of the contacts with the hard materials and the duration of use, independently from the precise stage of the butchery. Thus, a cutting edge used for skinning in an area where the bone is located just under the skin is susceptible to bear scars as marked as a cutting edge that was used for disarticulation. Only the presence of rounding could distinguish the flakes used for skinning, but its low intensity (on artefacts used for 13, 29, 47 and 62 minutes) and its non-systematic or extremely rare presence even on items that were used for quite a long time (for 14, 43, 57 and 76 minutes) make it a little informative criteria, especially in the case of studying altered archeological series that include a natural rounding present on all the edges and ridges.

Figure 39 - Schematic illustration of the macro-traces observed on an unmodified flake in quartzite used to deflesh a fresh hide

Figure 39 - Schematic illustration of the macro-traces observed on an unmodified flake in quartzite used to deflesh a fresh hide

Photograph and CAD: C. Thiébaut

Figure 40 - Schematic illustration of the macro-traces observed on an unmodified flake in quartzite used to disarticulate a carcass

Figure 40 - Schematic illustration of the macro-traces observed on an unmodified flake in quartzite used to disarticulate a carcass

Photograph and CAD: C. Thiébaut

Figure 41 - Schematic illustration of the macro-traces observed on an unmodified flake in quartzite used to disarticulate a carcass

Figure 41 - Schematic illustration of the macro-traces observed on an unmodified flake in quartzite used to disarticulate a carcass

Photograph and CAD: C. Thiébaut

Figure 42 - Schematic illustration of the macro-traces observed on an unmodified flake in quartzite used to scrape a dry hide

Figure 42 - Schematic illustration of the macro-traces observed on an unmodified flake in quartzite used to scrape a dry hide

Photograph and CAD: C. Thiébaut

Figure 43 - Macro-traces observed on edges used to cut grasses and dig soil

Figure 43 - Macro-traces observed on edges used to cut grasses and dig soil

The white lines indicate zones of rounding/ blunting

Photographs: É. Claud; CAD: C. Thiébaut and M. Coutureau

b - High magnification observations: micro-traces

98To describe the surfaces and cutting edges state of the quartzite flakes, we used the DIC system (see above) coupled with a metallographic microscope.

99Due to the presence of quartz crystals, the quartzite flakes cutting edges are characterized, before use, by a much more irregular topography (figure 44). The crystals are angular and sometimes show striations or deformations (ripples, folds) on their surface due to technological origin.

Figure 44 - Edges before use, viewed with a microscope (200 x magnification)

Figure 44 - Edges before use, viewed with a microscope (200 x magnification)

a: flint; b: fine-grained quartzite from the Neste River

Photographs: É. Claud

100Contrary to the reading of the macro-traces that had very rarely and partially been the subject of research and publications, the micro-traces on quartzite have already been described in the context of symposiums and publications by several colleagues by using the same observation method (Beyries, 1982; Plisson, 1986; de Araújo Igreja, 2009; Gibaja et al., 2009; Clemente-Conte, Gibaja Bao, 2009; Part I, chapter 2.2.B). On this point, we aimed to verify that the quartzites coming from the Aure Valley and the terraces of the Garonne River (fine-grained black ones and medium-grained brown ones) that are making up our reference collection were indeed adapted too this type of analysis, contrary to those anciently studied by our predecessors that bore no micro-traces (Beyries, 1982; Plisson, 1986).

101The potentialities of this raw material for microscopic studies having been confirmed, the study of the reference collection allowed us to draw determining criteria and to have comparison traces for the study of archeological tools.

102Like J. F. Gibaja et al. (2009), we observed the necessity to examine the crystals with a magnification above that normally used for flint study, that is to say 500 times. Indeed, this is about studying the use-wear that each crystal bears on its surfaces, ridges and edges, as from one crystal to another the type of traces are often distinct: for example a first one bears scars, a second one is eroded and a third one is striated and scarred. Thus the observation of all the traces is necessary for the identification of the mode of use of the tool.

103The use-wear features produced on flint, especially microscopic ones, are not transposable to quartzite, hence the interest of a specific reference collection. Indeed, one could expect that the matrix, even if it shows a coarser grain size than flint, would show a similar use-wear for a same mode of use. But in the case of butchery, the matrix does not bear any polish, while the latter are frequent on flint. I. Clemente-Conte and J. F. Gibaja Bao (2009) have also noted on quartzite from the north of the Iberian Peninsula and on rhyolites from Argentina (Tierra de Fuego) that the matrix wears very little, only recording traces after a very long duration of use. They explain this difference with regard to flint by much varied factors such as the mineralogical composition of the matrix, its degree of compression and by the irregularity of the micro-topography. Indeed, about the last factor, it seems logical that the main use-wear is located at the level of the crystals that are protruding on the surface and that the matrix, located between them, is submitted to less contacts. Consequently, contrary to macro‑traces, the micro‑traces are not more difficult to read on medium grained quartzite than on fine grained quartzite. In fact the matrix gets little wear and in both cases, it is the crystals that record in the most rapid and readable manner the use-wear traces. Thus it is more easy, at a microscopic scale, to study a cutting edge showing numerous and large crystals and little matrix than the opposite.

104Despite these differences, a certain number of use-wear rules common to flint and quartzite can be retained, like the fact that the harder the material is, the more the wear will be marginal and discontinuous (located on the exposed areas) or again that the action mode (transversal / longitudinal) influences the position (unifacial / bifacial) and the distribution of the traces (continuous / discontinuous). Furthermore, the morphology of the micro‑polish on the quartzite, when they are present, is similar to that of the polish described on flint, and thus constitutes an important clue to determine the nature of the material worked. The possible striations or deformations within the polish are providing, like in the case of flint, information on the kinematics of the action.

105Our detailed microwear observations below, are echoing those done by our colleagues on the quartzite of the Iberian Peninsula (see Gibaja et al., 2009; Clemente-Conte, Gibaja Bao, 2009).

106Amongst the various use-wear types found on quartzite, erosion has often been documented. This type of trace, called corrosion by J. F. Gibaja et al. (2009) was defined as the loss, disappearance, or dissolution of part of the crystals’ surface, can take various aspects that principally depend on the worked material. Erosion can be observed inside the surface of the crystal or be localized in the periphery or on the edge of the crystal, the preceding authors having proposed the terms of isolated or continuous erosion to qualify these two distribution types. They are not really in opposition as they are often seen on one or several crystals in the same active area, but it is frequent that a type of distribution is more frequent. On the whole of the reference collection, we have more often observed, at the scale of a crystal, a continuous erosion (e.g. figure 45a), or even both erosion types in association (e.g. figure 46g) than only an isolated erosion. Continuous erosion can develop into continuous breakage, defined by J. F. Gibaja et al. (2009) as the disappearance of the crystal from its periphery, the erosion thus uncovering the matrix on which the micro-polish develops. This phenomenon occurs in the case of the working of soft abrasive materials such as hide (figure 46c-d), but we have also more rarely observed it in the case of woodworking(figure 47a,e) and even more rarely in the case of items used for soil digging. The dimension and the regularity of the pitting are two other criteria retained to describe the erosion observed. The pitting appears as finer and more circular in the case of the working of soft materials such as meat and hide, erosion leaving a grained surface with a rather regular topography (figures 45a,c-e, 46). On the opposite, hard materials such as bone and horn produce a coarser pitting with variable morphology and depth, leading to a more irregular micro-topography (figure 48b,f). For medium-hard materials such as wood, it seems that the situation is intermediary. Some items used for scraping show a fine and regular erosion (figure 47b,d,f) while on some others the erosion is coarser (figure 47c,h-i). It is notably the case of wood drilling that produced an irregular and deep pitting in a highly localized area corresponding to the extremity of the active area, which could in this case result from the presence of quartzite microflakes at the bottom of the perforation (figure 47h-i). As for the items used for disarticulating, implying contacts with soft materials such as meat and hard ones such as bone, they show a variable pitting, sometimes on the same crystal (figure 45f). In the case of wild grasses cutting, erosion appeared little frequent and little intense (figure 49d), while the item used for soil digging is marked essentially by erosion, which is much advanced and for which the pitting is average in size and rather irregular (figure 49e). Finally, the distribution of the erosion on a crystal, especially in the case of continuous erosion, can inform us on the mode of action carried out, as it starts by the most exposed edge to motion (close to the tip of the edge in the case of oblique scraping, or perpendicular to the edge in the case of cutting).

Figure 45 - Micro-traces resulting from cutting activities related to butchery

Figure 45 - Micro-traces resulting from cutting activities related to butchery

a: continuous breakage and striations; b, g: scarring; c: rough, fluid or soft polish on a zone of erosion; d-f: continuous erosion; h: fine and superficial oblique striations on a crystal

Photographs: É. Claud

Figure 46 - Micro-traces observed on tools used to work fresh or dry hides

Figure 46 - Micro-traces observed on tools used to work fresh or dry hides

a: continuous erosion; b, f and h: rounding and polish with a rough, fluid or soft coalescence on the matrix; c: continuous breakage: the rounding and the polish (with craters) occur mostly in a line devoid of crystals; d: crystal in the process of dissipation by continuous erosion; e, g: start of continuous erosion and short, isolated striations in g

Photographs: É. Claud

Figure 47 - Micro-traces related to woodworking

Figure 47 - Micro-traces related to woodworking

a: micro-polish; b: fine, regular erosion, rounding and striations on the matrix, c-d: striations on crystals, isolated and continuous erosion in c and isolated erosion with micro-polish on the edge in d; e: continuous erosion, striations and micro-polish; f, g and i: erosion of varying intensity and striations on the crystals; h: large and deep pitting (extremity of the active zone

Photographs: É. Claud

107The crystals can also show striations, often associated to erosion, which will bring information about the motion of the tool. They are abrasive striations that tear some of the crystal’s surface and leave a linear mark. They are absent to rare, and rather fine and shallow in the case of soft materials working (skin processing, butchery, figure 45h), while they are more frequent, deep and large in the case of hard materials working such as bone, although their presence is not systematic (figure 48d-e,g). The working of these materials, especially by scraping, sometimes generates numerous, adjacent striations that transform the crystals into sticks. They are sometimes associated to striations that develop directly on the matrix. All the striations are organized into groups within which they are sub parallel. They are generally parallel or oblique to the cutting edge in the case of a longitudinal action and perpendicular in the case of scraping, but we have often noted the presence of parasitic striations or groups of striations, intersecting with the others (figure 48d-e). In the case of the medium-hard materials (wood), the frequency of the striations is variable although always lower than that described in the context hard animal material working. They can be fine and shallow but also sometimes on the same item rather deep and large, irregular in contour (figure 47c-d,f-g). In the context of grass cutting, the striations are rather rare, short and of medium width. They are localized only on the crystals, often associated to a beginning of erosion (figure 49d).

108Micro-rounding of the cutting edges have been detected on hide-working tools, on some items that worked wood by scraping or drilling and on the item used for soil digging. The intensity of the rounding connected to the working of the hide correspond to that observed on flint for equivalent modes of use, that is to say that when scraping, especially hide, the rounding is intense (figure 46c-d,h). The rounding it very intense and on top of all very covering in the case of the item that was used to dig soil. If macro‑rounding has been observed on items used for working hard animal material and sawing wood, we did not detect rounding at a microscopic scale on these objects. The edges of the crystals can also sometimes be rounded or blunted at the same time as eroded, this is the case of hide working and butchery. In the case of the working of wood, of bone and of the cutting of grasses, the presence of hard coalescence polish and of domed or draped topography (see below) can also round the edges of the crystals (figures 47d, 49b). The position of the rounding on the cutting edge or on the edges of a crystal can also provide information on the mode of action of the tool.

109Micro‑scars are likely to happen on the edges of the crystals exposed along the cutting edge. They have been observed on nearly all the experimental items, but they are neatly more numerous when working hard or medium-hard materials. On the items used for butchery, scars have been observed in the case of skinning and disarticulation. They are resulting from accidental contacts with bone and cartilage. They are discontinuous and often oblique with regard to the cutting edge, showing a longitudinal action. They are showing the same features as the macro-scars found on the tools dedicated to butchery: discontinuous, semi-circular or triangular with a bending initiation and a feather or step termination (figure 45b,g). Woodworking, especially by scraping, also created some scarring, the latter being sometimes continuous and oriented perpendicularly to the cutting edge. Hard animal material working is responsible for a large number of scars on the crystals, especially in the case of scraping, where they are mostly located on the trailing edge, perpendicular, continuous, sometimes superimposed and/or large in size (figure 48c). Finally, the cutting up of grasses produced a large number of scars on the crystals; the latter are located on both faces, often continuous, aligned or even superimposed (figure 49a). They are oblique or perpendicular and are sometimes themselves covered by the use-wear polish (figure 49c).

Figure 48 - Micro-traces related to cleaning or working hard animal materials

Figure 48 - Micro-traces related to cleaning or working hard animal materials

a and f: micro-polish; b: beginning of continuous erosion; c and h: scarring; d, e, and g: striations on crystals

Photographs: É. Claud

110Finally, micro-polish constitutes another trace observed on the cutting edges of quartzite tools. It is not always present on tools used for defleshing, disarticulation, sawing hard animal material and for cutting up and piercing dry hide, on the opposite to items that worked on fresh hide, on dry hide by scraping, on wood, that scraped bone and cut grasses; these latter showing a clearly readable polish identical to the one found on the flint cutting edges (figures 46b-d,f,h, 47a,e, 48a,f, 49b). It can be associated to rounding if the latter is found, and show striations, impacts, as in the case of hide working, or elongated depressions in the case of bone scraping. The polish is most often observed on the matrix (after the erosion of the crystals or on the initial matrix) even if in some cases we were able to see the development of micro-polish on the crystals, especially on their edges in the case of wood and grass cutting (figures 47d, 49b).

Figure 49 - Micro-traces related to other activities: cutting grasses (a to d) and digging soil (e and f)

Figure 49 - Micro-traces related to other activities: cutting grasses (a to d) and digging soil (e and f)

a and f: scarring; b: polish; c: scarring overlain with polish; d: striations and beginning of erosion of a crystal; e: intensive and large erosion of a crystal

Photographs: É. Claud

111On top of the different possible variations of each type of trace, their significance with regard to each other and their possible associations are determining elements for the identification of the worked materials:

  • in the case of butchery by cutting, erosion dominates, it is fine and regular (except in case of repeated bone contacts), keeping on the cutting edge a large number of crystals, sometimes eroded, sometimes scarred or striated, or quite often intact. Polish is rare. Some differences have been seen between the use-wear connected to the different stages of the butchery done by cutting (skinning, defleshing, disarticulation, and/or tendon removal, see Annex 2). Firstly, it appeared that defleshing does not produce or produces very little micro-polish, contrary to skinning (polish on the matrix and on the crystals), to disarticulation and to tendon removal (polish on the matrix only). Secondly, a micro-rounding, low in intensity, was observed only on the items used for skinning. Thirdly, striations would be rarer in the case of skinning than for the other stages. Fourthly, grain erosion would mostly be of a continuous type on the flakes used for skinning while it would be isolated and continuous on the other items. Erosion would also be fine and regular and micro‑scars rarer and smaller in the case of skinning and defleshing. Finally, the phenomenon of continuous breakage has only been seen in the case of skinning, in connection with the repeated contacts with the hide. Even if it is more tendencies than strict rules, the features of the micro-traces do seem to vary according to the stages of the butchery. The latter could thus be distinguished, on the basis of the mentioned criteria, on the archeological material, on the condition of very good preservation;

  • in the case of hide working, erosion also dominates but it is much more intense even for a cutting up work. Crystals can sometimes still be observed on the cutting edge, but they are, with some exceptions, modified by continuous erosion, that sometimes lead to a continuous breakage in the case of scraping. A rather intense rounding and a rough and soft coalescence polish are often seen. Scars are rare and small in dimension. Striations can also be observed;

  • in the case of woodworking, erosion is indeed present, the size and the regularity of the pitting vary, and the latter is sometimes associated to a slight rounding. Some crystals are still visible on the cutting edge but often altered by numerous striations or an ongoing erosion. The polish is well developed, and the scars, oriented perpendicularly to the edge, are visible;

  • in the case of the working of hard animal material, the main use-wears are scars on the crystals (trailing edge in the case of scraping), numerous, often superimposed and large in size striations and polish (leading edge). Erosion is also present but it preserves a large number of intact or nearly intact crystals (scarring or ongoing erosion) close to the edge. The erosion is coarse and leaves an irregular micro-topography on which the polish develops only in the most exposed areas;

  • in the case of the grass cutting, erosion is little intense, preserving a large number of crystals on the cutting edge. Striations are rare and short, limited to the surface of the crystals and to the polished surfaces. The scarring and polish, mostly localized on the crystals and not on the matrix dominate within the use-wear traces;

  • finally, on the item used for digging soil, the erosion and the rounding are very intense (continuous breakage) and highly covering. Striations and scars are rare, but the latter can be large in size. A moderately bright, rough and soft coalescence polish is sometimes associated to the erosion and to the rounding.

D - Synthesis of the results of the experimental reference collection study

112About the macro-traces, one of the main differences observed with regard to flint is the more systematic and more pronounced developing of a rounding of the cutting edge that superimposes on the first use-wear scars, whatever the material worked (hide, wood and bone) on the quartzite blanks. On flint a more significant superimposition of the scarring can be seen in the case of hard material working, and rounding that develops nearly only in the context of hide working. The development of these rounding could be connected to a rapid decrease in efficiency often noted for the quartzite cutting edges during our experiments and mostly in the context of working medium-hard and hard materials.

113Despite some reservations that we have put forward – notably due to some difficulties to characterize the scars, especially on medium-grained quartzite – our widely illustrated study shows that the macroscopic scale, by using a combination of criteria, can be relevant in the context of use-wear studies of quartzite industries of the Aure Valley and of the terraces of the Garonne River (figure 50). It allows in a first time to identify the used artefacts and to select them within numerically large assemblages, to locate the active edge, to identify the action mode and to evaluate the hardness of the material worked, or even, in the best of case, to propose an activity (see the scarring typical of butchery work). This idea is rather in opposition to the observations of J. F. Gibaja et al. (2009) who considered that the macro-traces did not bring sufficiently diagnostic criteria to determine the worked materials, because the scars are not as clear as on other rocks, they are less frequent and smaller. This dissenting point of view could be explained by a difference in the type of quartzite studied, the level of cohesion of the grains and the tenacity that conditions the manner the cutting edges will wear (Leipus, Mansur in Gibaja et al., 2009): by the loss of grains leading to a rounding or by breakage (scars). The quartzites studied by the authors, coming from the south of Portugal, could thus be especially sensitive to rounding and develop little scarring. On the other hand, the macroscopic approach requires, in our view, a good quality of preservation of the analyzed items, and if the conservation allows it, it must be completed by a microscopic examination that can bring a confirmation or a more precise interpretation of the modes of use proposed from the low magnification observation.

Figure 50 - Overview of the primary macro-traces from use observed on unmodified quartzite flakes, according to the hardness of the material worked and the mode of action

Figure 50 - Overview of the primary macro-traces from use observed on unmodified quartzite flakes, according to the hardness of the material worked and the mode of action

Photographs: É. Claud

114About the microscopic approach, the analyzed types of quartzite lend themselves well to the search for various traces of use-wear whose presence was evidenced by our colleagues on the quartzites from the Iberian Peninsula (Clemente‑Conte, Gibaja Bao, 2009; Gibaja et al., 2009). If the micro-polish and the micro-rounding seen on the flint and quartzites are comparable for a similar mode of use, on the quartzite there is a range of specific traces that need to be taken into account as they are capital in the interpretation of the modes of use. Indeed, erosion, striations and scars born by the crystals are constituting clues whose significance has been revealed by the study of this reference collection. Documenting the variability of all the micro-traces according the modes of use allows us today to have reading keys for the interpretation of archaeological artefacts and comparison traces (figure 51).

Figure 51 - Overview of the primary micro-traces observed on unmodified quartzite flakes, according to the nature of the material worked and the mode of action

Figure 51 - Overview of the primary micro-traces observed on unmodified quartzite flakes, according to the nature of the material worked and the mode of action

Photographs: É. Claud

7 - The experimental denticulates reference collection

C.Thiébaut, A. Coudenneau, É. Claud, M.-G. Chacón-Navarro

A - Why such a reference collection?

115For many years notched pieces have suffered from a bad reputation resulting from their confusion with objects simply altered by mechanical phenomena (Thiébaut, 2005, 2007a, 2007b, 2007c). Nevertheless, for the last two decades they have tended to be restored as human productions whose purpose, however, gives rise to numerous questions. Indeed, beyond questions connected to the taphonomic aspects of these items, the questions of their technological role (core, intermediary phase in the making of side scrapers or tools) and of their domestic function are raised.

116Some authors consider them as cores (see especially Geneste et al., 1997; Bourguignon et al., 2004), but it seems that this hypothesis cannot be generalized to all the notched pieces and that it is not presently demonstrated in the series of the Denticulate Mousterian (Bordes, 1961a, 1961b; Thiébaut, 2005, volume 1). As we realize that this terminology may appear today as slightly reductive to consider a cultural group (Thiébaut, 2013; Thiébaut et al., 2014), our intention is not here to discuss the relevance of F. Bordes’ facies or of what can be called today technocomplexes (Delagnes et al., 2007) but to apprehend the function of the notched pieces in order to identify a possible specialization of these tools. This approach will allow to better appreciate the cultural features that could be connected to this tool as it aspires to a better comprehension of the techno-economical behaviours of the human groups of the Middle Palaeolithic. The precise techno-economical study of lithic industries from various sites (especially Mauran, La Borde, Les Fieux layer K, Saint‑Césaire level Egpf) highlights the production of notched pieces on blanks resulting from full Discoid debitage, thus showing technological features common to various series (Thiébaut, 2005, volume 2). Thus these items would show the anticipation of a specific need that does not correspond to expedient tools as was proposed for other technological assemblages (Geneste, 1985; Meignen, Vandermeersch, 1986; Meignen, 1988). However, these elements do not constitute the main purpose of the debitage chaîne opératoire although they dominate the retouched toolkit, as in these series it constitutes less than 10 % of the produced blanks. The production is thus primarily aimed at obtaining of pseudo‑Levallois points and thick core-edge flakes intended to be used unretouched (Thiébaut, 2013). Parallel to these technical assemblages, a small number of series with denticulates made on flakes resulting from a Levallois debitage are found (Jonzac layer 8: Jaubert et al., 2008; Abri Brouillaud: Geneste, 1985).

117To this relative diversity of the blanks of the notched pieces, significant morphometric variations within the notches and denticulates can be added, which could correspond to distinct functions and/or modes of use (Thiébaut, 2003).

118From an archeozoological view point, several series are associated to a monospecific fauna (Mauran, Laborde, Les Fieux Kdenticulates: Jaubert et al., 1990; Farizy et al., 1994; Gerbe et al., 2014) but this is not systematic (Thiébaut et al., 2014). Their direct relationship with the acquisition and processing of carcasses of large bovid cannot be established without use-wear analyses. These items could be used for making wooden structures.

119Causing little interest amongst prehistorians, these items have not benefited from large scale use-wear analyses. Thus, the rare functional studies published about notched pieces generally concern very small samples of notches and denticulates within which the various types are not detailed (Gysels, Cahen, 1981; Anderson-Gerfaud, Helmer, 1987; Beyries, 1987a, 1993b; Shchelinskij, 1993; Lemorini, 2000; Martínez‑Molina, 2005). Therefore the absence of systematic overall study of these artefacts within a lithic series does not allow appreciating a possible specialization of the various notched tools.

120The experimental reference collections specific to this type of items are even rarer and more laconic and the only hypotheses and comparison data available concern attempts at using notches and denticulates in longitudinal or transverse actions on wood, published by F. Bordes at the beginning of the 1960s (Bordes, 1961a) and by S. Kantman (1970a, 1970b).

121In order to determine the function of these items, resorting to use-wear analyses was therefore unavoidable and the creation of an experimental reference collection essential.

B - The production of notched pieces

122To make experimental notched pieces, we favoured the production of tools similar to those excavated at Mauran, Les Fieux and Saint‑Césaire. Thus, we made Clactonian notches and denticulates with micro- and macro‑denticulation in flint and quartzite using thick blanks, issued from a Discoid debitage with cortical or core-edge backs but also items on thinner blanks like some pieces from Mauran and Saint-Césaire by using a flatter Discoid debitage. Two main techniques were used by two knappers (V. Mourre and C. Thiébaut), direct percussion with a hard hammer of the type of a pebble with convex touch for the production of Clactonian notches and denticulates with macro- and micro-denticulations and more tangential direct percussion with a hammer with dihedral touch of the type of a flake cutting edge for the making of denticulates with tighter micro-denticulation (see Thiébaut et al., 2009b). As for the raw materials used, we selected a large range of flint from various regions (Murs, Sault, Charente and Senonian from the Dordogne). For quartzite, first we mostly used medium and coarse grained quartzite from the Garonne River’s alluvium. But we quickly realized that these quartzites did not show the use‑wear traces observed on the archeological tools, as the wear on the experimental items was happening by grain loss, creating a rounding, rather than by the developing of scars as had been observed on the tools from Mauran. Afterwards, we favoured the production of blanks in fine quartzite from the alluvium of the Neste River. The latter, closer to the raw material used at Mauran and at Grotte du Noisetier, did record scarring.

C - Description of experiments and identified constraints

123In order to constitute our reference collection of traces, we carried out different activities involving various types of actions on varied raw materials (see tables 1, 4, 9, 10). The creation of this reference collection spread over six years, from 2007 to 2012, allowing to complete it alongside to observations carried out on the archeological material.

124Our reference collection includes 67 active areas, 35 in flint and 32 in quartzite, used in butchery (n=33, including two notches from the same denticulate used to clean the periosteum by scraping), hide working (n=8), wood (n=25) and, in a much more exceptional manner, horn (drilling, n=1 “bec”).

a - Woodworking

125Two main objectives guided the activities carried out: acquiring trunks or branches and their transformation to obtain poles for the drying of hide or for making hunting spears. Acquiring was done with a continuous contact (i.e anticipation of a non‑percussive) gesture and a longitudinal action. Their transformation was done with a continuous contact gesture by transverse actions with a more or less open leading angle (scraping, planing, pointing, and bark stripping). Different varieties were used allowing to work more or less hard woods (acacia, oak, hazel, beech, hackberry and poplar), green and dry. For sawing we favoured the use of more or less thick denticulates with micro‑denticulation, but we also used denticulates with macro‑denticulations and Clactonian notches to test the hypothesis proposed by F. Bordes and S. Kantman (see above). For these various experiments, the denticulates were held with bare hands or with a skin wrapping, but none was hafted.

126Firstly, about acquiring wood, the use of notched pieces in sawing appeared clearly less efficient than percussion with bifaces or flake cleavers (see Part I, chapters 2.9 and 2.10). The time needed varied between 20 minutes for trunks or branches with diameters under 3 cm and about one hour for diameters around 6.5 cm. The relation between the energy spent and result obtained appears rather unproductive to our 21st century Homo sapiens eyes. In detail, the sawing of soft wood, even dry (poplar), medium-hard like beech and hard like oak (but in a green state) did not prove very difficult with denticulates with micro-denticulation, even thick ones. However, it is important that the profile of the denticulate be regular and preferably rectilinear to have a greater length of usable cutting edge and an easier penetration. On the other hand, the use of a thick denticulate with macro‑denticulation in flint as well as sawing with Clactonian notches, with convexo-concave morphology in section, even for green hackberry, proved impossible. Although this could seem logical, we wanted to test this technique that has several times been mentioned in various publications (Bordes, 1961a; Kantman, 1970a, 1970b). Let us note that dry oak trunks could not be sawn with a denticulate held in the hand. The sawing of standing oak trunks required to take the tool in both hands, which was made possible by the large dimensions of the used blanks, larger by several centimeters than most of the tools from Mauran or Les Fieux.

127For sawing, whatever the variety of wood and its state of freshness, the use of a denticulate required to carry out a “diabolo” type motion to prevent the blank being stopped in its action by its own thickness (see figure 6c).

128Sawing with quartzite denticulates proved possible but only on soft or medium-hard wood of small diameter.

129The transformation of trunks and branches includes various activities: bark removal, planing, pointing and abrasion of the point (see Part I, chapter 1.2). Without great surprise, it was much easier to work on green wood than on dry wood and it was even impossible to plane a dry beech branch with a quartzite notch (see Part I, chapter 1.2). As the fibers clogged the hollows between the grains, the cutting edge needed regular cleaning. It also seemed to round more quickly, thus penetrating less into the worked material. Flint Clactonian notches also proved more efficient than those in quartzite to plane tree trunks (see figure 6). Notches with plano-concave morphology in section were more adapted to this activity (see Thiébaut, 2003). This remark is important as a large number of archeological Clactonian notches generally have convexo‑concave morphologies in section, thus little adapted to this type of activity. If the morphology in section of the notch and the width of its opening seem to be non-negligible criteria for a good efficiently of the tool, the angle of the notch seems to have had little impact, as long as it is between 45° and 70°.

b - Butchery activities

130For butchery activities, various animal species have been treated according to different methods: half a carcass of a boar and six half carcasses of red deer were skinned, defleshed, the tendons sometimes removed and some carcasses disarticulated (figures 13-15). A bison limb was defleshed and the tendons removed. In parallel, the bone of a cow was scraped with a notch to prepare the bone surface for fracturing and recovering the marrow.

131For the processing of two half carcasses, belonging to a boar and a red deer, we chose to use the same denticulate for the different steps of the chaîne opératoire (skinning, disarticulation, meat and tendons removal). This allowed to assess on one hand the life duration of a denticulate and on the other hand, its efficiency during the different steps. Both aspects were significant as we know that when the retouched toolkit is dominated by denticulates in an industry, the latter are few in relation to the unretouched flakes (Thiébaut, 2007b). Thus, it was interesting to be able to establish a possible link between the low number of these items and a duration of life possibly longer than for other tools (unretouched flakes, side scrapers). For the half-boar, only one denticulate in flint with macro-denticulation was necessary and at the end of the experiment, it was still usable. All the steps of the chaîne opératoire of the carcass processing were carried out, except for the removal of the ribs and tendons and the complete disarticulation of the limb bones (see Part I, chapter 1.3). The denticulate was used during 2 h 18 min. to remove the hide of the animal, to disarticulate the upper limbs and remove the meat from the limbs and the axial skeleton. It was efficient during the different steps, and especially for the skinning and the disarticulation of the limbs. However, although the meat removal was done, the tool seemed less adapted than a long unretouched cutting edge penetrating with more ease into the meat.

132For the first red deer half-carcass, four denticulates with medium and macro-denticulation were necessary. They were used until the cutting edge was not efficient any more, with durations of use varying between 1 h 10 min. and 2 h 50 min. Except for the complete disarticulation of the limb bones, the other steps of the chaîne opératoire of the carcass processing were carried out including the removal of the ribs and tendons. Thus, the denticulates were used for the skinning of the animal, for the disarticulation of the upper limbs, the removal of the meat, of the ribs (scraping and sawing) and of the tendons. The efficiency of the denticulates for the skinning, the disarticulation of the metapodials and the removal of the tendons was proven; on the other hand, their use to remove meat and disarticulate the femur seemed little adapted. The presence of macro‑denticulation associated to the thickness of the meat, larger for a deer than for a boar, did not allow a good penetration of the tool if the thigh was cut into from the outside. As we will see further, the disarticulation of the femur from the inside of the leg proves much more efficient and this is the case with any tool type, whatever its dimension.

133The second red deer half-carcass was treated with five denticulates of various morphotypes. This time we used the various items for a single activity: skinning, meat removal and disarticulation of the upper part of the limbs, removal of the tendons and disarticulation of the lower part of the limbs or cutting up the tenderloin in small pieces on a wooden support. This choice was guided by the will to characterize the traces connected to a specific step (skinning, disarticulation, defleshing) on the experimental material in order to possibly be able to precise the step during which the archeological butchery tools were used.

134For the processing of the third half-carcass, we also used various quartzite denticulates (n=6), with varying features according to the step of the butchery: the removal of the hide was done with a denticulate with micro-denticulation and a denticulate with medium denticulation, both hafted, the proximal part towards the outside, without glue or bindings in a small previously split wooden shaft. This technique makes prehension easier and allows applying more pressure on the tool. The disarticulation of the head and the front and back legs was done with two denticulates with medium and micro-denticulation, sometimes hafted. Denticulates with micro-denticulation were used for the removal of tendons and the disarticulation of the bottom of the legs. Finally, the meat was removed with two denticulates with macro-and medium denticulation.

135The various denticulates used for these two half carcasses were very efficient and the majority of those made of flint were still functional. However, the quartzite denticulates became rounded more quickly and at the end of the experiment, they were, except for one item used only for nine minutes, too rounded to be still efficient.

136The fourth red deer half-carcass was treated with five new quartzite denticulates by changing item at each step of the removal of the skin and of the meat. There was no disarticulation of the back legs. The various denticulates used were efficient except for TH09 Q3 that showed dimensions too small for allowing using the artefact with bare hands. Like for the planing of the branch, one of the inconvenient of quartzite pieces results on the fact that they need to be cleaned regularly. Indeed various particles from the animal treated fill the gaps between the quartzite grains and lower the efficiency of the cutting edge by making it less sharp.

137A last red deer carcass was treated with five quartzite denticulates for the disarticulation of the four limbs, the removal of all the meat, the skinning and the removal of the tendons of a single leg. Again, the regular cleaning required by the cutting edges made quartzite denticulates little adapted to this activity.

138For the experiment carried out on the bison, we used a flint denticulate to disarticulate the bones and remove the tendons of the left posterior limb and three quartzite denticulates to disarticulate, remove the meat and tendons of the right posterior limb. In order to answer the question raised by the association of small dimension denticulates to bison carcasses on the site of Mauran or Les Fieux, one of the objectives here was to test the efficiency of small size denticulates during the large-sized preys processing.

139The disarticulation of the humerus was carried out without difficulty with a small size blank as we incised the meat by the inside of the limb and not by the outside. Despite the necessary frequent cleaning of the quartzite blanks, all the tools allowed reaching the fixed objectives. Let us note that the experimenters, also helped by the advice of a butcher, were probably more skillful than in the first years, as this experiment was carried out in the last year of the Research Program.

140Finally, a cow femur was scraped with a flint denticulate during about one hour to remove the periosteum before fracturing it to recover the marrow. The various notches were used alternatively and proved efficient, except for the first notch, whose morphology in section was little adapted, as the starting face showed a disturbing convexity.

c - Hide working

141Notched pieces were also used to work on hide (figures 17, 19):

  • one “bec” to pierce fresh hide;

  • one quartzite denticulate and a flint denticulate to cut a dry humidified red deer hide;

  • two other denticulates, one made of flint and one of quartzite, to cut up a dry bovid hide;

  • finally two flint notches and one quartzite denticulate used as passive tools to soften strips of hide.

142We did not carry out any defleshing on fresh hide, as the notched pieces are little adapted to this type of activity because of the high risk of piercing the hide. Dry hides were generally laid on grass and cut up from the internal side of the skin (grain or hypoderm). One of the cow’s hide was set up on a dry wooden block, hypoderm upwards, and cut up into strips by parallel incisions, with a longitudinal gesture. Although the objective was reached, the used tools did not seem to us more efficient than an unretouched cutting edge. Cutting up with quartzite denticulates proved rather laborious, as well as the softening of the strips.

143The strips from the cutting up of cervid hides were softened with three notches, one of which was set up in a crack in a dry oak block allowing the softening to be carried out by just one experimenter. The softening was carried out by a bidirectional friction of the strip in the notch (figure 19). The other notches were set up vertically on a board and held by one of us with wooden sticks. The strips softened with this technique were softer than the ones softened by friction on an oak trunk.

d - Summary

144During the various activities carried out, the used denticulates presented a relative efficiency, except for the quartzite items used for hide working and wood scraping. Their use in butchery, although possible, required a frequent cleaning of the items. Indeed, the main inconvenience, noted by various experimenters, was the rapidity of the rounding of the quartzite cutting edges. However, the analysis of the latter with stereomicroscope did not show a significant rounding. Actually, when using the quartzite cutting edges, especially during activities including continuous contact gestures and linear or diffuse contacts (such as scraping or cutting activities) and more specifically when the cutting edges are in contact with soft or fibrous materials (fresh hide, fat, meat and wood), particles of the worked material fit closely to the surfaces of the used parts and fill the gaps between the quartz grains. This reduces the sharp and abrasive feature of the cutting edge. Thus quartzite denticulates need a regular cleaning of their cutting edge during use. This peculiarity had already been noted before, especially during experiments including unretouched quartzite cutting edges in Filonian quartz (Bracco, Morel 1998).

145About the efficiency of the notched tools, it is interesting to note that woodworking, by scraping or sawing, requires, for greater efficiency, specific cutting edges morphologies. A cutting edge rectilinear in profile and in delineation, rather biplane in section, is adapted to sawing. A significant length of the cutting edge allows the sawing of hardwood trunks, standing, by holding the tool with both hands. The presence of micro‑denticulation allows an easier penetration into the wood than with a simple untreated cutting edge. For scraping, plano‑concave or biplane morphologies in section are more adapted, but actually the presence of a notch is not essential.

146According to the step of the butchery chaîne opératoire, various morphologies of notched items can be used. The dimension of the blanks seems to us to be a little determining criterion for tool efficiency during butchery as even the smallest were efficient during the processing of large herbivore carcasses, whether for the removal of the meat or the disarticulation of the limbs. For skinning, the presence of a pointed denticle in the apical part of the tool allowed incising the hide more easily, but all the types of denticulates can be used during this step. About the removal of the meat, the irregular feature of the denticulate cutting edges in comparison to an unretouched cutting edge appears to have little effect on the efficiency of the tool. But we need to recognize that the use of denticulates for this step does not bring any major contribution in comparison to a simple unretouched cutting edge that penetrates with more fluidity into the thickness of the meat.

147On the other hand, as the disarticulation of the limb bones requires rather narrow blanks whose cutting edge morphology (delineation, cutting angle) allows penetrating easily into the articulations, denticulates appeared clearly less adapted than unretouched cutting edges. The removal of the tendons can be done with denticulates, especially with denticulates with micro-denticulation that will penetrate more easily into the thickness of the tendon.

148About the butchery activity in the broad sense, it is interesting to note that, finally, the use of denticulates does not appear to be a major advantage in comparison to a simple unretouched flint cutting edge. Their main advantage seems to be the sharp edge given by the denticles, allowing to cut upsoft but resisting materials such as hide or tendons, and in the case of flint denticulates, their durability is more significant than for an unretouched cutting edge.

149Finally, one must note that, whatever the activity carried out, the presence of a back opposite to the active edge allows better handling of the tool.

D - Use-wear traces on experimental tools

150We are documenting below the macro-traces visible on the active areas of the used notched pieces. Indeed, as previously noted in Part I, chapter 2.2.B), micro‑traces are more fragile and more rarely preserved in the old series and depend little on the morphology of the cutting edges, on the opposite to macro‑traces, whose study thus require the creation of specific reference collections.

151As for the possible macro-traces created by prehension or hafting of the rare hafted notched pieces, they proved to be absent during our observations with stereomicroscope. It is true that the hafted areas are showing an open angle (back) that does not encourage the forming of scarring.

a - Tools without traces or nearly

152Within our reference collection, 11 used tools, all in quartzite, do not bear any macro-traces, except for a possibly rounded grain (table 17). Various parameters can be concerned, in all likeli-hood acting in a combined manner:

Table 17 - Number of tools without macro-traces

Butchery

Hide working

Skinning

Defleshing

Dry hide cutting

Softening of strips

Medium-grained quartzite (n=6)

2 tools (between 9 and 40 min)

3 tools (9 to 30 min)

1 tool (21 min)

-

Coarse-grained quartzite (n=5)

1 tool (69 min)

3 tools (16 to 65 min)

-

1 tool (21 min)

  • the raw material of the blank: this parameter seems to be a major element, although it is not decisive by itself as will be seen below in the context of a use of the object in percussion (Part I, chapter 2.9). However, during activities involving a continuous contact gesture, it appears that the scars are developing much less on quartzite denticulates than on flint denticulates. This difference, connected to the raw materials, was also highlighted for the unretouched cutting edges (Part I, chapter 2.6). This data is important and will have to be mentioned in the interpretations resulting from the analysis of the archeological material in order to avoid any hasty conclusion in term of economics of the raw materials;

  • the duration of use: this parameter is as important as obviously the longer a cutting edge is used, the more traces have time to develop. However, it does not explain alone the absence of trace as some tools used for over one hour are lacking macro-traces;

  • the worked material: an important parameter to consider, since skinning for example, as we will see below, generally leaves no or few macro-traces, which would anyway be difficult to interpret. On the opposite, all the quartzite denticulates used on wood bear at least one type of macro-trace. Thus, like for unretouched cutting edges (Part I, chapter 2.5), the more flexible the contact material is, the less macro-traces are on the cutting edge, even for very long durations.

153To this number of blanks without traces, 15 tools (5 in flint and 10 in quartzite) can be added showing macro-traces that would not be, in an archeological context, interpretable with certainty as coming from the use of the cutting edge (table 18). These macro‑traces are little developed (few scars) and/or they show common morphologies with some natural alterations. The same parameters as those previously mentioned are combined to explain this observation. The degree of thinness of the raw material, the duration of use and/or the type of raw material worked thus determine the diagnostic or non-diagnostic feature of the macro-traces. Indeed, quartzite denticulates showing non-diagnostic macro-traces are either coarse grained quartzite blanks used on hard materials but during a short time, or tools on fine quartzite used longer on very soft materials. For two flint denticulates, the cutting of soft materials (skinning or meat cutting without bone contact) although used for relatively long (57 and 80 minutes), did not produce sufficiently characteristic macro-traces. It is also the case of a denticulate used for a short duration to saw a soft and green wood (figure 52).

Table 18 - Number of tools with macro-traces considered non-diagnostic of a particular use

Flint (n=6)

Fine/very fine grained quartzite (n=5)

Medium grained quartzite (n=2)

Coarse grained quartzite (n=2)

Skinning

1 tool (57 min)

Meat cutting

1 tool (38 min)

Butchery

Defleshing

1 tool (80 min)

2 tools (15 and 26 min)

1 tool (44 min)

Disarticulation + defleshing

1 tool (167 min)

Disarticulation

-

2 tools (10 and 14 min)

1 tool ( 26 min

Skinning of the legs + tendons removals

-

1 tool (40 min)

Hide working

Cutting dry hide

-

1 tool (19 min

Woodworking

Sawing (standing poplar)

1 tool (19 min)

Scraping (dry beech)

-

2 tools (2 and 7 min=

154Finally, 41 active areas out of 67 are showing diagnostic traces of use that is to say about 61 % of our reference collection. Although slightly disappointing, these results remain significant as they are underlining and reminding the interest and the limits of the low magnification use-wear approach to identify the activities carried out, even in ideal preservation state of the traces.

Figure 52 - Schematic illustration of the macro-traces observed on a flint denticulate used to saw green poplar

Figure 52 - Schematic illustration of the macro-traces observed on a flint denticulate used to saw green poplar

Photographs and CAD: C. Thiébaut

b - The different types of traces according to the activities carried out

155Three types of macroscopic traces have been observed (see Part I, chapter 2.4):

  • scars;

  • micro-fractures;

  • rounding.

156All the modifications observed on each item were sketched on a drawing representing the analyzed object (e.g. figures 52-62). This intermediary approach to the recording of traces in the data base allows the schematic reconstitution of some features proper to a mode of action (location, distribution) and to a raw material (morphology and type of traces).

157The reference collection of items showing traces interpretable as the result of the object use comprises 41 areas including 34 % of the quartzite blanks (11/32) and 86 % of the flint ones (30/35). In terms of worked material, it is obvious that the soft raw materials will be the less well represented in this part (table 19). Indeed, the tools that cut hide or meat without bone contact generally do not show any macro-trace (table 17) or traces too weak to be recognized or interpreted in the study of archeological tools (table 18). They are weak rounding and rare scars, small in dimensions and whose location, distribution and morphology are not characteristic of a use.

Table 19 - Number of tools presenting use-wear traces considered as diagnostic, according to modes of action and categories of material worked

Number of items bearing traces considered as diagnostic in archeological context

Total number of items

%

Longitudinal actions

23

45

51

Soft materials (hide, meat)

1

6

17

Soft materials with rare bone contacts

3

13

23

Soft to medium-hard materials (disarticulation, tendons)

7

11

64

Butchery (all steps)

2

2

100

Abrasive medium-hard materials (dry hide)

1

3

33

Hard to medium-hard materials (green soft wood to dry hard wood

9

10

90

Transverse actions

10

12

83

Abrasive medium-hard materials (dry hide)

2

3

67

Medium-hard materials (wood)

7

8

88

Hard materials (bone)

1

1

100

Rotary action

1

1

100

Soft materials (fresh hide)

1

1

100

1 - Distinctive features of the traces (figures 52-64)

158As for the unretouched cutting edges, we noted that the type of action carried out has effects on the position of the macro-traces. Longitudinal actions created more scarring on both faces (22/24). Only two quartzite denticulates used to saw a dry beech branch are showing either only one scar, or 3 scars on a single face. As for them, transverse actions (scraping, planing) have mostly left alterations on only one face (7/9). Without surprise, the scars and micro-fractures are concentrated on the areas where contacts are more frequent and/or more intense (more pressure), that is to say at the top of the denticles in the case of longitudinal actions, and on the faces of the notches used to saw or scrape hide or wood. As for rounding, it develops, in fact, on the active parts that are most in contact with the worked material: the denticles and the cutting edge. In three cases of dry wood sawing (poplar and oak), an invasive rounding on the ridges of the negatives of the notches was also observed. Finally, in the case of a quartzite item used to softened leather strip, the rounding was observed not on the cutting edge itself but on one or several grains/ crystals of the areas in which the contacts were more intense. Rounding is also present on the objects used in butchery (6 cases out of 14), for dry wood sawing (5 cases out of 14) or for working dry hide (3 cases out of 14).

Figure 53 - Schematic illustration of the macro-traces observed on a flint denticulate used in the butchery of a cervid

Figure 53 - Schematic illustration of the macro-traces observed on a flint denticulate used in the butchery of a cervid

Photographs and CAD: C. Thiébaut

Figure 54 - Schematic illustration of the macro-traces observed on a flint denticulate used to saw dry poplar

Figure 54 - Schematic illustration of the macro-traces observed on a flint denticulate used to saw dry poplar

Photographs and CAD: C. Thiébaut

Figure 55 - Schematic illustration of the macro-traces observed on a flint denticulate used to saw oak

Figure 55 - Schematic illustration of the macro-traces observed on a flint denticulate used to saw oak

Photographs and CAD: C. Thiébaut

Figure 56 - Schematic illustration of the macro-traces observed on a flint denticulate used to saw dry poplar

Figure 56 - Schematic illustration of the macro-traces observed on a flint denticulate used to saw dry poplar

Photographs and CAD: C. Thiébaut

Figure 57 - Schematic illustration of the macro-traces observed on a flint denticulate used to saw oak

Figure 57 - Schematic illustration of the macro-traces observed on a flint denticulate used to saw oak

Photographs and CAD: C. Thiébaut

Figure 58 - Schematic illustration of the macro-traces observed on a flint denticulate used to saw dry poplar

Figure 58 - Schematic illustration of the macro-traces observed on a flint denticulate used to saw dry poplar

Photographs and CAD: C. Thiébaut

Figure 59 - Schematic illustration of the macro-traces observed on a flint denticulate used in the butchery of a wild boar

Figure 59 - Schematic illustration of the macro-traces observed on a flint denticulate used in the butchery of a wild boar

Photographs and CAD: C. Thiébaut

Figure 60 - Schematic illustration of the macro-traces observed on a flint denticulate used to disarticulate the left hind limb bones of a bison

Figure 60 - Schematic illustration of the macro-traces observed on a flint denticulate used to disarticulate the left hind limb bones of a bison

Photograph and CAD: C. Thiébaut

Figure 61 - Schematic illustration of the macro-traces observed on a quartzite denticulate used to disarticulate the right hind limb bones of a bison

Figure 61 - Schematic illustration of the macro-traces observed on a quartzite denticulate used to disarticulate the right hind limb bones of a bison

Photograph and CAD: C. Thiébaut

Figure 62 - Schematic illustration of the macro-traces observed on a quartzite denticulate used to disarticulate the right hind limb bones of a bison

Figure 62 - Schematic illustration of the macro-traces observed on a quartzite denticulate used to disarticulate the right hind limb bones of a bison

Photograph and CAD: C. Thiébaut

- Scars

159The organization of the scars clearly depends, logically, on the action carried out, but also partly on the material worked. In the case of transverse actions, the scars are continuous and aligned for medium-hard materials (dry skin, soft wood) or superimposed for hard materials (dry hard wood and bone). In the case of longitudinal actions, the scars are in majority discontinuous and isolated or aligned. The superimposition of the scars is more frequent in a contact with a hard material, like the numerous contacts with bone during the carcass processing. The denticulates used in butchery or on hardwood have, logically, more frequently superimposed scars.

160The initiation of the scars is nearly always bending, some cone initiations being sometimes observed in the context of butchery with numerous bone contacts, as well as on the cutting edges that scraped the periosteum.

161The morphology of the scars appears connected to the hardness of the worked material, but also, to a lesser extent, to the action carried out. The quadrangular morphologies are very frequent whatever the worked material, except for the scraping of dry hide for which semi-circular morphologies dominate. This morphology is therefore not a discriminating criterion for a type of worked material. On the other hand, although they are not dominating, triangular morphologies and step or hinge terminations clearly refer to butchery work, as eight denticulates out of eleven are showing at least one removal of this type, whether they are flint or quartzite blanks. The presence of this scar and termination type could therefore, as in the case of unretouched flakes, be a feature of the items used in butchery. Let’s note that two out of the three items used in butchery, but not showing this type of scars, were used for cutting up soft material and/or for a short time (less than ten minutes) for disarticulating or cutting up tendons (one flint denticulate and one quartzite denticulate). The last item, made of fine quartzite, was used for 18 minutes for disarticulating limb bones, but is showing less than ten scars, the latter being semi-circular in morphology, with step termination.

162As for the working of medium-hard materials such as wood, it leads to the formation of generally quadrangular and/or trapezoid and half-moon scars. The termination is feathered. Let us underline that only one active area used for woodworking showed a triangular scar with step termination. The planing of dry wood spears sometimes created superimposed scars whose termination is, for some, step, therefore similar traces to that resulting of the scraping of osseous material. Indeed, the only item used for bone scraping shows quadrangular and trapezoidal scars with step or hinge terminations, similar to those observed on the unretouched cutting edges.

- Micro-fractures

163They are showing varied morphologies: burin-like (or twisted), bending or straight.

164Generally and logically, micro-fractures affect the denticles of the items used in forcing activities on hard or medium-hard materials and are found mostly on flint items used in longitudinal action during butchery or wood sawing (20 cases out of 23). For blanks used in butchery or for the cutting of dry skin, the various types of fractures are found with a majority of burin-like ones (figure 63a) followed by bending fractures, located mostly at the level of the tip of the denticles. On items that sawed wood, burin-like fractures have been observed in the case where hard and/ or dry woods have been worked (figure 63c). On the opposite, in the case of soft and/or green wood, the fractures are dominated by bending morphologies (figure 63b).

Figure 63 - Micro-fractures observed on the denticles of the experimental denticulates

Figure 63 - Micro-fractures observed on the denticles of the experimental denticulates

Photographs: A. Coudenneau

165Let us note that three items used in transverse action for very different activities (bone scraping, softening of strips and bark removal) show fractures on the denticles; two are bending and the last one is straight.

- Rounding

166As for the unretouched cutting edges, the presence or absence of rounding depends on the action carried out (longitudinal action versus scraping) as well as on the worked material (green wood, dry wood, fresh hide, dry hide, …). In the case of notched pieces, rounding is present on the items used in butchery (6 cases out of 14), for sawing dry wood (5 cases out of 14) and for working dry hide (3 cases out of 14).

167Only six items used in butchery (flint and quartzite) are bearing limited and localized rounding on the points or on part of the cutting edge.

168About woodworking, rounding is only found on blanks used for sawing dry wood (5 cases out of 8). Two coarse-grained quartzite denticulates used for the same activity are not showing any rounding. Is this connected to the raw material? It is possible that on the quartzite items and especially coarse-grained ones, rounding does not develop in such an intense way.

169The quartzite item used for softening a leather strip did not show any use-wear trace, except a grain rounding that appears to us as difficult to interpret in term of use.

170The three other items used to cut up or soften dry hide strips are showing rounding, whose intensity is stronger on the denticles and on the sharp edge than on the remainder of the cutting edge.

171As we can note reading these results, the only thing allowing to determine the hardness of the worked material and the gesture carried out, whatever the raw material of the tool, is taking into account the whole of the macro-traces found on a cutting edge, associated to their morphological features, their location and their position on the item, if they are in sufficient numbers. The same “use-wear rules” apply to notched pieces and to flint unretouched cutting edges (see Part I, chapter 2.5). For a similar duration of use, coarse-grained quartzite blanks appear more resistant to the developing of scars and rounding than the flint notched tools. The developing, on several denticulates used for dry wood sawing, of a rounding on the extremities and the ridges of the denticles is original as it has rarely been illustrated within unretouched flakes reference collections. This feature is probably in close connection with the specific morphology of the cutting edge of the denticulate tools, implying very intense and highly localized contacts on the protruding parts, thus able to bear much developed traces. Moreover, the duration of use of the denticulates used for sawing were relatively long. For some items, the fact that the sawn wood still had its bark could also explain the presence of rounding, as the bark is rich in abrasive particles that lead to the formation of well-marked rounding (see Part I, chapter 2.5). Finally, if the use-wear shown by the notched pieces do not differ from those observed on a simple unretouched cutting edge, the denticulates differ mostly by a distribution and an intensity of traces closely connected to the presence of the denticles: if one or several of them are included in the active area, they will bear the most intense and most characteristic use-wear, be it rounding or micro‑fractures.

2 - Diagnostic potential of macro-traces on the notched pieces

172In order to attempt quantifying the number of items that could be connected to a type of activities in an archaeological context in which conservation conditions would be optimal, we carried out, several years after the study of these items, a test done by the traceologist who had previously studied this reference collection by asking her, for each item, to propose a functional interpretation of the object.

173Out of the 11 items showing butchery use-wear traces (4 in quartzite and 7 in flint) and considered originally as interpretable as use traces, seven were connected without difficulty to a butchery activity (1 in quartzite and 6 in flint). The use-wear visible on two quartzite denticulates and one flint denticulate, which did not show triangular step scars (see above), were attributed to a longitudinal action on a medium‑hard material. A last denticulate, used for disarticulation, was identified as used for cutting up a soft to medium-hard material. Finally, amongst all the denticulates used in butchery, only one could be connected to a precise step, more precisely disarticulation.

174The notch used to scrape fresh bone in order to clean the surface before fracturing was identified as a tool that was used in a transverse action on medium-hard to hard material (wood or bone scraping).

175Amongst the items used for sawing wood, only three blanks, all made of flint could be connected with near-certainty to this precise activity. Two items in flint which sawed poplar and dry oak, were connected to a longitudinal action on medium-hard material. Three blanks (one in coarse-grained quartzite and two in flint) having sawn dry and hard wood were interpreted as having been used in a longitudinal action on medium-hard to hard materials in a context different from butchery. Finally, two items, one in quartzite and the other one in flint, were connected to a use in a transverse action (scraping) on wood (flint blank) or on an undetermined material (quartzite blank) resulting from the rarity or the absence of scarring on one of the faces.

176Amongst the items used for scraping wood, four have been interpreted rightly as resulting from a transverse action on a medium-hard material of the wood type. Two others, having scraped soft and dry wood (poplar), were showing spontaneous removals produced during the retouching, similar by their features to the scars generally visible during scraping of a hard material of the bone type. Their presence complicated in fact any functional interpretation of these tools.

177Finally, the two flint notches used as passive tools to soften dry hide strips have indeed been identified as such.

178Despite the questionable nature of such a “blind test” (the tester having known the results, but some years previously), errors of interpretation for the objects considered as informative are extremely rare. The degree of precision of determination of the worked materials nevertheless varies, and we must regret the difficulty, or even the impossibility, to replace each item used in butchery within a precise step of this activity.

E - Synthesis

179At the end of this study, several remarks, all of them highly significant for the functional interpretation of denticulates and archeological notches, must be stated:

  • compared to an unretouched cutting edge, denticulates appear to have greater action efficiency only when sawing a trunk or branch. They remain significantly less efficient than the use of heavier percussion tools for the same purpose: the acquisition of woody materials. In the context of butchery, their main asset lies in the presence of denticles allowing to cut more effectively into resistant materials such as thick skin or tendons. Clactonian notches may be of some benefit if the morphology in section of the cutting edge is plano‑convex during spear planing activities;

  • flint blanks always seem more efficient than quartzite blanks that require a more systematic cleaning of the cutting edges, during all activities;

  • macro-traces develop more on flint blanks than on quartzite blanks. Indeed, barely a third of the quartzite blanks used bear traces that can be interpreted as resulting from certain use;

  • denticulates have macro-traces generally comparable to those found on unretouched cutting edges used for the same activities (figure 64). Their presence is not systematic, probably in connection with the relatively open cutting angle of the cutting edges, at least of the concave (internal) part of the notches. Thus, in the case of longitudinal actions, the traces, taking the form of fractures, scars and/or rounding are concentrated on the denticles, which are both the most exposed and most fragile areas;

  • as with unretouched blanks and other tools (see Part I, chapters 2.9 and 2.10), the macro-traces related to the working of soft materials (hide, especially fresh, meat, soft and green wood), absent or tenuous, may not be identified as traces of use, and thus lead to problems of under-representation of one or more activities (hide working, light butchery, green and soft wood-working).

180In the end, about two-thirds of the items have macro-traces diagnostic of a specific use. It would therefore be a pity to do without the information available at this level of reading, while keeping in mind that some of the data is likely to escape us, especially if the micro-traces are not, in parallel, preserved and studied.

Figure 64 - Overview of the macro-traces observed on the experimental notched pieces, according to the material worked and the mode of use

Figure 64 - Overview of the macro-traces observed on the experimental notched pieces, according to the material worked and the mode of use

Photographs: A. Coudenneau ; CAD: C. Thiébaut and É. Claud

8 - The points reference collection

A. Coudenneau, F. Venditti, C. Lemorini, M.-G. Chacón-Navarro

A - Why such a reference collection?

181Middle Palaeolithic studies are punctuated with academic debates and reflections on how to understand the behavioural similarities between Homo neanderthalensis and Homo sapiens. Many Anglo‑Saxon researchers have argued that organized hunting could only be carried out by Anatomically Modern Humans (Rendu, 2007). Indeed, despite a growing number of archaeological data speaking in favour of controlled hunting, some authors are still believing that Neanderthals were mostly scavengers (Binford, 1985; Dibble, Mellars, 1992; Stringer, Gamble, 1994). The meat-rich diet of Neanderthals in western Europe is now firmly established: for example, isotopic analyzes have shown that their diet placed them amongst high-ranking carnivores (Bocherens et al., 1991; Richards, Trinkaus, 2009), but this information obviously does not allow saying if it is a food acquired by hunting or by scavenging. In fact, only archeozoological studies have shown that Neanderthal groups not only hunted but also had complex strategies, sometimes with game selection.

182These results are in contradiction with old studies that did not agree. The example of Grotte Vaufrey is telling: a 1988 study by L. R. Binford concluded that during the Saalian period the inhabitants of the cave were scavengers (Binford in Rigaud, 1988). However, a methodological review and a new study of the material made it possible to conclude that, on the contrary, the Neanderthals of Grotte Vaufrey had conducted selective deer hunting (Grayson, Delpech, 1994). Other studies have shown that in the Middle Palaeolithic there are examples of specialized slaughter sites for acquiring carcasses of large herbivores such as bison (Mauran: Farizy et al., 1994; Coudoulous: Jaubert et al., 2005; Les Fieux: Gerbe, 2010), aurochs (La Borde: Jaubert et al., 1990) or reindeers (Les Pradelles: Costamagno et al., 2006). In addition to these specialized sites, examples of archeozoological studies demonstrating the existence of intentional and organized predatory behaviour in the Mousterian are now numerous.

183However, determining which technical means were used for this predation according to the hunted game and the various groups of Neanderthals that populated Europe remains to be done. By technical means, we mean at the same time the hunting weapons used, the strategies and the collective or individual organization set up during the hunt. Regarding hunting weapons, numerous research conducted in the Near East have demonstrated the use of lithic elements as hunting points on sites of the Middle Palaeolithic, based on several specimens of Levallois points (Shea, 1988, 1997, 2003; Bergman, Newcomer, 1983; Solecki, 1992; Wendorf, Schild, 1993; Plisson, Beyries, 1998). But little data are available for the European continent.

184In 1995, three wooden spears were unearthed at a coal mining site in northern Germany, the Schöningen site (Dennel, 1997; Thieme, 1997). They were accompanied by knapped stone tools and the remains of ten horses with traces of butchery. The spears measure between 182 and 225 cm with a maximum diameter ranging from 2.9 to 4.7 cm. They were made from the trunk of a young pine tree whose branches were cut and the bark removed. The lengths of the weapons are not identical but they are showing the same model of construction. The basal part of the tree was used to make the distal part of the spear. Therefore, its pivotal point is always located in the upper third of the weapon. This balancing model corresponds to the standards used in the manufacture of modern javelins so as to obtain the best possible ballistics. These weapons could have been used as projectile weapons.

185Woody material elements are obviously exceptional and most direct clues of use of hunting weapons are to be found within the lithic assemblages. The site of Bouheben (Landes) delivered several points with traces of impact taking the form of face or transverse bending fractures with step termination (Villa, Lenoir, 2006). A Mousterian point discovered on the site of Angé (Loir-et-Cher) also shows traces of use as a hunting weapon (Soressi, Locht, 2010; Locht et al., 2015). In the Iberian Peninsula, recent research also highlights the probable presence of flint points used as hunting weapons (Lazuén, 2012b). While some doubts remain about some of the points that have been the subject of publications (for a more exhaustive synthesis and a discussion, see Part II, chapter 4.3), it is nevertheless conceivable that the Neanderthal groups made hunting weapons in stone and did not settle for wooden spears alone, even in Europe. However, the data remain rather disparate and without a systematic interdisciplinary analysis, it is difficult to restore more accurate hunting behaviours concerning the methods of prey acquisition by Neanderthal groups. The study of the bone remains is inseparable from that carried out on the stone points to understand the preferred hunting methods (see Part I, chapter 3).

186While faunal data are quite abundant and have been the topic of more systematic studies on the subject of hunting for several decades (Rendu, 2007; Gerbe, 2010), the question of hunting weapons remained a recurrent one and deserved a methodological and structured research to be set up on the lithic industries of western Europe, as no large scale functional study of the triangular elements had been done for this geographic area. This type of study is all the more legitimate since the Middle Palaeolithic series of Western Europe are full of triangular tools (flakes of triangular morphology, Levallois points, pseudo‑Levallois points, Mousterian points, convergent side scrapers). However, amongst all these products only three major categories of objects have been named points:

  • Levallois points: triangular flakes obtained by a Levallois debitage and whose debitage axis coincides with the technical axis (Bordes, 1961a). These points can be classified according to their more or less elongated morphology (ordinary Levallois points, ogive-like Levallois points, elongated Levallois points) or according to the presence of retouching (Soyons points, retouched Levallois points, Emireh points), or again according to their place within the debitage (first order, second order Levallois point, …);

  • Pseudo-Levallois points: triangular flakes most often obtained by Discoid debitage (but other processes are possible) and whose debitage axis does not coincide with the technical axis (Bordes, 1953b). The triangular flakes thus obtained are rather short and thick flakes with a butt and a back that take a significant place in the periphery of the object;

  • Mousterian points: they are side scrapers with two converging edges forming an acute angle and whose lower face is flat (Bordes, 1961a). This definition remains ambiguous in the sense that the notion of acute angle should be specified. In addition, the means for obtaining such objects are varied and they are to be defined in each series.

187In the context of the Research Program, a significant reference collection of triangular tools could be gathered and served as a base for the comparative use-wear analysis of various Middle Palaeolithic archaeological series rich in triangular elements from the sites of Payre, Mauran, Les Fieux and Coudoulous. This reference collection has also been used to study series integrated into a doctoral research (Coudenneau, 2013): Spy (Belgium), Beauvais (Oise) and Therdonne (Oise).

188We wanted to create a reference collection for the triangular items as exhaustive as possible, thus going beyond the elements only dedicated to hunting activities. Previously only two series of experiments had been conducted with this aim. One concerns a series of twelve points fitted and used as thrusting weapons (Plisson, Beyries, 1998). If this experiment allows a first approach, the tested corpus is insufficient to cover all the questions relating to the elaboration of solid identification criteria of the hunting traces on the Mousterian lithic elements. The other series of experiments was conducted by J. Shea to interpret industries in the Near East (Shea, 1988, 1997, 2003). It is problematic because it takes into account criteria for identifying traces that are too inaccurate for the Middle Palaeolithic of Europe. Indeed, these experiments focus mostly on Levallois points and take into account only the possibility of their use as a projectile. For our part, we have integrated into our experiments different types of points (Mousterian points, Levallois points and pseudo‑Levallois points and triangular flakes) used as throwing and as thrusting weapons.

B - Used tools and activities carried out

189The points intended for the experiments were made by different experimenters (J.-B. Boudias, V. Mourre and C. Thiébaut) in connection with the various techno-types of points encountered at the studied sites: Mousterian points, pseudo-Levallois points, Levallois points and triangular flakes. The raw materials used are also related to the studied sites: Coniacian flint from the Yonne region, Bedoulian flint from Murs, Forcalquier flint quartz and white quartzite from the Lot region, Pyrenean quartzite. In total, 198 pointed elements including 182 flint and 16 quartz and quartzite were used. The corpus gathers 198 triangular elements used according to different modes of action for varied activities. We also have 35 items dedicated exclusively to the fracturing (intentional or accidental) of the lithic blank, in order to be able to compare these fractures with those related to use, in particular as hunting points. With the exception of the points used during hunting activities, the points were generally held with bare hands or with a leather cover. The durations of use are variable: the points used during hunting activities were projected between one and ten times maximum. As soon as a fracture appeared, or, if necessary, after ten shots, the use of the point has been stopped. The duration of use of the other points is generally between 5 and 220 minutes. The methods of use correspond mainly to butchery, hide working and woodworking.

a - Hunting

190107 points (95 in flint and 12 in quartz / quartzite) were used as weapon extremities on various animal carcasses killed before the experiment and held in anatomical position (figure 10). At the beginning of the work of the Research Program, having little financial means and little experience, we had to resolve to use two carcasses of adult sheep. In a second time, three cervids (does and red deer) were used. The presence of cervids being confirmed on many sites of the Middle Palaeolithic, it was indeed interesting for us, as for other specialists, to study identical species to those hunted by the Neanderthals. The first ewe was gutted during the shots, unlike the carcasses that followed. The points were attached to the handle with bindings of plant cord (linen) reinforced by an adhesive made of wax, ocher and resin, heated and poured on the insertion area of the point into the handle.

191The first experiment involved 15 flint points (7 unretouched and 8 retouched) shot mechanically, by a system consisting of an underwater shotgun, on the gutted ewe. The points were fitted on shafts of light poplar wood. These shafts were drilled in the central length to allow the arrow of the rifle to come into the shaft. The purpose of this first session was to obtain preliminary information on projectile fractures, by limiting the parameter of variation induced by hand throwing. We could also determine that it is best not to gut the animal before the experiment to add additional inertia to the target and perhaps a little more realism to this experimental activity. Finally, it allowed us to note that, whatever the size of the point, its effectiveness is determined by two conditions: the sharpness of the point and the protruding of the edges relative to the handle. The more the point protrudes from the handle, the easier it will penetrate into the animal.

192The second experiment involved 30 retouched flint points used as throwing thrusting weapons on a non-gutted ewe. The objective this time was to check whether there are significant differences between the morphology of fractures obtained by mechanical propulsion and that of the fractures obtained with thrusting weapons on the retouched blanks. All the haftings of the points were axial. The shafts used were beech and pine studs 1 to 2 m long and 2 to 3 cm in diameter.

193Having found no fundamental differences between the traces observed on the points that were thrown mechanically and those used in thrusting weapons, the 32 flint points used for the third experiment were in thrusting weapons, but this time, on a carcass of cervid (adult deer). These are 14 unretouched points and 18 retouched points. In this session, we wanted to check if the thickness of sheep wool could have an influence on the formation of the fractures by comparing the fractures obtained during the experiments on the sheep and those obtained on an animal whose hair is less thick and compact. This experiment allowed us to observe that the hide of the red deer was easier to penetrate. It was also an opportunity to test some lateral hafting: seven retouched points were dedicated to this test. We found that the morphology of our points was not appropriate to this type of hafting: as the bindings did not lend themselves to side hafting, we used the adhesive alone. The problem is that a large number became dislodged during the experiment. The shafts were similar to those of the previous experiment.

194The last two experiments with red deer involved 20 flint points (5 unretouched and 15 retouched) and 12 points in quartz and quartzite. The quartz and quartzite points were used in the same way as the flint points, that is to say as thrusting weapons with an axial hafting and fixed to the handle by means of plant rope bindings and of an adhesive made of a mixture of resin, wax and ocher.

b - Butchery

195Regarding butchery activities, in the broad sense, 55 active areas were dedicated to various actions carried out for the carcass processing of different species: Deer, Fox, Boar, Reindeer, and Bison. With the exception of two deer metapodials that were simply disarticulated, all the butchery operations from skinning to disarticulation were carried out on the other species. The objectives of the actions conducted out were multiple:

  • recover the hide;

  • remove the meat;

  • retrieve the tendons;

  • recover the bones.

196A total of 15 unretouched flint points, four unretouched quartzite points, two retouched flint points and nine flint Mousterian points were used during the different stages of the butchery chaîne opératoire. Six points were used for skinning, three for disarticulation, eight for defleshing, nine points for complete butchery except skinning and four were used at all stages of the butchery chaîne opératoire from skinning to disarticulation.

197The points have proved effective for most stages of this chaîne opératoire. The point allows to open with great efficiency the various tissues and the cutting edges allow a good cutting of the flesh.

c - Hide working

198For the working of hide, nine active parts from eight points were used for different activities: two points were used to deflesh fresh hide (rabbit and sheep) in tangential cutting, a Mousterian point (two active areas) was used to deflesh a dry, ash coated sheep hide by scraping the grain to remove the last residues of flesh and fat. Finally, five points were dedicated to the piercing of hide: two were used to pierce fresh hide in a rotational motion before setting it on a frame, three were used to pierce dry and tanned hides to make sewing holes. One of these points was used in a continuous contact gesture (i.e non-percussive) and by applying pressure on the point placed perpendicular to the leather without rotational movement. The other two points were used in a continuous contact gesture with again a punctiform contact but by applying a rotational movement. The points have proved very efficient for this work. The punctiform contact actions constitute a functional specificity of these points.

d - Acquiring and processing wood

199Different activities, from the acquisition to the making of objects in wood were carried out with 21 flint points:

  • for the first phase, only the acquisition by sawing (continuous contact gesture, longitudinal action) has been implemented. A poplar branch of small diameter was sawn with a retouched point;

  • during the second phase, for the preparation of the wooden supports, two retouched points were used to plane. The small branches that could interfere with the making of objects were sawn with an unretouched point and a retouched point;

  • the processing of the acquired and prepared wood into finished objects was carried out using thirteen points, having worked the wood in the dry state (boxwood and pine), for the manufacture of handles and spears, using the “diabolo” technique. Two points were used in grooving (retouched), four in scraping (two unretouched, two retouched) and seven in planing (six retouched and one unretouched). With the exception of the longitudinal grooving for which the point can guide the gesture by creating a distinct groove, for the other operations the presence of a point is completely useless without being inconvenient;

  • we also drilled dogwood and oak wood to obtain a hole in which we could pass a tie (e.g. for suspension). For this activity, three points were used: two Mousterian points on dry wood and a Mousterian point on green wood (figure 7). The points have proved very efficient for this use, the drilling of green wood being easier. The thickness of the wood may be a limiting factor for the success of this operation. When the diameter of the handle to be drilled was too large (> 20 mm), we made a hole on both sides of the handle until both resulting holes met.

e - Bone working

200Two distinct objectives guided the activities carried out on the bones: a food objective and a technical objective. In the context of butchery activities and more specifically for the bone marrow extraction, six unretouched points were used to scrape the periosteum before fracturing of the bones. Well aware that the making of bone objects is rare in the Middle Palaeolithic, different actions have nevertheless been carried out on this material in order to discriminate the traces obtained from those produced by the activities on wood: two points, one unretouched and one retouched, were used to scrape a fresh bone to prepare the surface before grooving, three unretouched points were used to saw bones and three retouched points to groove a fresh bone.

f - Seashell working

201To complete the reference collection, five points were used to drill very hard materials like seashells (Nucella lapillus).

C - Description of use-wear traces

a - Frequent macro-traces

202Macro-traces are present on almost all the active parts of the used points. These macro-traces are mainly scars, except for the points used during hunting activities, for which the observed macro-traces are mainly fractures. Only five points have no trace. These are points used on fresh hide: defleshing and piercing of the hide before setting them on a frame. This scale of observation is not negligible, whether the points are unretouched or retouched.

b - Differences with unretouched cutting edges

203Macro-traces on the cutting edges of the retouched points can be compared to those on the unretouched cutting edges with equivalent angle described in the literature (e.g. Gonzáles Urquijo, Ibáñez Estévez, 1994; Lemorini, 2000). The cutting edges of the retouched points follow the same use-wear modes as the unretouched edges. Without radically modifying the features of the macro-traces noted on the unretouched edges, we can nevertheless observe some notable differences in the case of the butchery activities and on the tools used to work wood. For activities involving materials of greater relative hardness, these differences are almost non-existent.

204The main differences are:

  • the scars of the retouched active parts tend to be smaller in size;

  • on the retouched points, the retouched face shows less numerous scars, more often discontinuous and isolated;

  • the scars present on the retouched edges have more frequently a step termination.

205On the other hand, the position, the morphology and the direction of the scars are little affected by the presence of retouch. It is the same for rounding and blunting.

c - Description of macro-traces according to the modes of use

Traces linked to hunting activities

Flint points

206The traces associated with hunting activities have particular features compared to those produced during other activities. They are mainly fractures (80 % of cases), sometimes accompanied by scars (figures 65-67).

207The fracture analysis of experimental hunting points allows us to make several observations:

  • the fractures observed on the points used as hunting weapon are bending fractures with step termination or apical scarring (Coudenneau, 2013: 41-45; table 12, figures 65-66) with step termination. The presence of apical scarring seems to be a real feature of this mode of use since we do not find them for any other tested activity;

  • the presence of a single item with a bending fracture and a step termination in an archaeological serie cannot be a proof of the use of this item as a hunting weapon. Indeed, other causes can produce this type of fracture. And the presence of an apical scarring on a single item in a series should also be interpreted with caution. This type of scarring actually constitutes the physical translation of a particular application of the forces, but it is not impossible that other domestic activities or taphonomic events could also produce such fractures;

  • - there are variations in the fracturing modes of the artefacts also related to the presence or not of retouch and to the relative thickness of the fur of the hunted animal. Thus the denser the coat (sheep), the more the resulting fractures tend to be bending fractures. Conversely, fractures produced during hunting actions on red deer are more frequently apical scarring. However, these variations are at most nuances and should be handled with care;

  • - the presence of a double fracture does not seem to be related to a particular factor amongst those we tested. As during debitage, it is very likely that a less marked or more diffuse contact with the material causing the fracture may produce such an effect. It is also possible that, during its journey inside the animal, the item meets bones several times and thus breaks twice. Double fractures are most often an association between a bending fracture and an apical scarring;

  • - the scars of the lateral edges are not, strictly speaking, a discriminating feature of this type of use, but quadrangular scars with step termination, or combinations of scars associating half-moons and quadrangular or trapezoidal morphologies with transverse or step terminations may be an additional evidence of use as a hunting weapon. This evidence gains weight when associated with a fracture such as those described above;

  • - in all cases, the recurrence of traces considered as diagnostic within the same archaeological series will be a significant element to guarantee interpretations based on these experimental analyses.

Figure 65 - Examples of points used experimentally as hunting weapons

Figure 65 - Examples of points used experimentally as hunting weapons

a: ventral surface before and after use and detail of the apical scarring with a step termination; b: dorsal and ventral surfaces before and after use and detail of the apical scarring with a spin off termination (A) and the associated scars (B and C)

Photographs and CAD: A. Coudenneau

Figure 66 - Examples of fractures observed on points used experimentally as hunting weapons

Figure 66 - Examples of fractures observed on points used experimentally as hunting weapons

a: apical oblique scarring with a step termination and a step-terminating bending transverse fracture; b: step-terminating bending fracture; c: apical scarring with a step termination; d: apical oblique scarring with a spin off termination

Photographs: A. Coudenneau

Figure 67 - Examples of scarring observed on points used in hunting

Figure 67 - Examples of scarring observed on points used in hunting

a: association of quadrangular and half-moon scarring with step terminations; b: quadrangular scarring with step terminations; c: combination of quadrangular, semi-circular, and trapezoidal scarring with step terminations. The arrows indicate the direction of movement

photographs: A. Coudenneau).

The quartz and quartzite points (F. Venditti, C. Lemorini)

208Before giving the functional results of experimental impacts for quartz and quartzite points, it is necessary to describe briefly the features of these raw materials and the methodology used to analyze them.

209Some quartz and quartzite are composed of coarse quartz grains that produce an irregular surface with many prominent ridges (Knutsson, 1988a: 42). It can happen that the traces of use are not distributed evenly along an active edge, not allowing detecting them easily. It is therefore necessary to locate them on the different planes of the crystals, which means observing them on highly localized surfaces. Moreover, on each part, the degree of development of the traces can be very different, even if it is the product of the same activity. Therefore, if one can use any method of observation for fine-grained rocks, in order to detect traces of use on the materials referred to here, one must carefully analyze all the edges of the object, to locate the most developed traces that allow the best interpretation. It is obvious that the time required by this type of analysis is necessarily longer than for other types of raw materials.

210In addition, the surfaces of the quartz crystals are almost never plane and smooth and they may show imperfections related to the origin of the crystals themselves or abrasions due to knapping activities. These imperfections or abrasions may have a morphology that resembles some of the micro-traces. It is therefore necessary to have casts of the surface before the use of the item to be able to control each crystal composing it. For this purpose, we used a two-component silicone (Provil Novo Light Fast Set, Heraeus) that, in addition, helped us to attenuate the phenomenon of light reflection typical of quartz crystals and allowed a better observation of the lithic surface (see Part I, chapter 2.3).

211Following the methodology developed for this analysis, we observed all the points using a stereomicroscope in reflected light to identify scars or fractures produced by the impact of the point on the animal.

212After low magnification observation, we performed the analysis of the casts of the used points with a metallographic microscope with 100 ×, 200 × and 500 × magnifications, provided with reflected light and an interferometry system.

213Out of the 12 points used, three were not analyzed because they did not have pre-use casts; amongst the nine others, five points show a combination of traces typical of the action carried out.

214We are presenting here the description of each analyzed test and the description of the recognized traces.

215Point no. 2:
Point no. 2 was used twice without being able to penetrate into the animal. The faint traces visible on the surface of some crystals are chaotic abrasions and a stronger abrasion that has destroyed the left side of the crystal edge (figure 68).

216Point no. 5:
Point no. 5 was used three times, then we stopped using it as it moved in its handle. Slight converging abrasions developed on the ventral surface of the apical part (figure 69
b).

217Pointe no. 6:
Point no. 6 was used several times and the fifth shot hit a rib causing an invasive fracture with a
snap termination (figure 70c). In this case, we found a large crystal that, in the pre-use cast, presented converging technological abrasions that disappeared after the use of the item. It is possible that the contact with the subcutaneous fat has caused the disappearance of the abrasions present on the crystal by obstructing the small holes produced by the abrasion.

218Indeed, fresh hide working and skinning activity did not produce diagnostic traces.

219The subcutaneous fat may cause alterations in the crystals that homogenize the appearance of their surface without developing a distinctive and recognizable feature. It thus becomes difficult, if not impossible, to recognize this type of trace on archaeological materials.

Figure 68 - Point no. 2

Figure 68 - Point no. 2

a: experimental set-up; b: point before use; c: unused crystal on the edge before use; d: abraded crystal on the left edge

Photographs a-b: PCR Des Traces et des Hommes; c-d: C. Lemorini and F. Venditti

Figure 69 - Point no. 5

Figure 69 - Point no. 5

a: experimental set-up; b: micro-traces produced during experimentation: crystal showing a row of superficial abrasions after use

Photographs a: PCR Des Traces et des Hommes; b: C. Lemorini and F. Venditti

Figure 70 - Point no. 6

Figure 70 - Point no. 6

a: experimental set-up; b: point before use; c :point fracture ; d: micro-traces due to experiment ; e: cristal before use

Photographs a-b: PCR Des Traces et des Hommes; c-e: C. Lemorini and F. Venditti

220Pointe no. 7:
With point no. 7, ten shots were done that hit the stomach of the animal. There was only one micro-trace of contact with meat near the point; it affected the edges and the central part of the crystal (figure 71). Contact with the fleshy tissues may affect the surfaces of the crystals on the cutting edge either by slight abrasions with a light bottom that change the generally smooth surface of the crystals or by deeper abrasions, with a dark gray bottom, that can be attributed to more violent contact with the tissues or sometimes with the bone (for example in butchery activities). These abrasions can be located either on the edge of the crystals, by breaking them up, or towards the center of their surfaces and they can have a distinct direction or a chaotic orientation.

Figure 71 - Point no. 7

Figure 71 - Point no. 7

a: experimental set-up; b: point before use; c: micro-traces produced during the experiment: crystal showing abrasions that can be attributed to contact with meaty tissues after use

Photographs a-b: PCR Des Traces et des Hommes; c: C. Lemorini and F. Venditti

221Pointe no. 4:
Point no. 4 showed the most significant traces of the series. We encountered three different types of traces but in combination and in agreement with the action carried out (figures 70, 73-74). This point ran through the deer twice: the first shot penetrated the fleshy tissue through the skin and touched a rib. The second shot caused a
snap fracture and only the proximal part of the item remained inserted into the shaft (figure 72).

222By comparing the crystals before and after their use, the entire part of the point that came into contact with the hide, the fleshy tissues and the bone shows several crystals with:

  • superficial and slight abrasions typical of contact with fleshy tissues;

  • deeper abrasions with a direction perpendicular to the point and therefore in agreement with the impact movement;

  • crystals with smooth and luminous surfaces probably due to contact with the subcutaneous fat.

223Out of the 12 points thrown, 5 points bear micro-traces, but in only one case the three types of micro-traces described above (superficial abrasions, deep and perpendicular abrasions, smooth and luminous crystals) were present concomitantly. For three cases, the impact with the carcass caused fracturing: this is snap fracturing of the distal or mesio‑distal part of the point.

224It is obvious that only the combination of different traces on the same item can give a high degree of reliability to its final interpretation as a hunting point.

225Experimentation has shown that this combination does not always develop and therefore, from an archaeological point of view, it is likely that some impacts will never be reliably identified. The reasons are as follows:

  • in some cases the points remain intact;

  • the single presence of snap fractures is not evidence of impact given that this morphology can develop also for technological or taphonomic reasons;

  • the single presence of traces of fleshy tissue could be due to the butchery activity;

  • the single presence of abrasions is not an obvious evidence since this type of trace can be found on the natural surface of the crystals.

Figure 72 - Point no. 4

Figure 72 - Point no. 4

a: experimental set-up; b: point before use; c: proximal part of the point remaining in the haft after fracture

Photographs: PCR Des Traces et des Hommes

Figure 73 - Point no. 4

Figure 73 - Point no. 4

a, c, and e: crystals observed before use; b: smooth and polished crystal after use; d: crystal with a slight abrasion oriented perpendicular to the point, after use; f: crystal bearing traces typical of butchery, after use

Photographs: C. Lemorini and F. Venditti

Cutting soft animal materials without bone contact

226These operations are certainly the most difficult to identify archaeologically on the basis of macro-traces because they only occur very rarely, especially when treating a fresh hide. For these operations, it will therefore be necessary to rely more on micro-traces.

227Of the seven points used to deflesh hides, only two show scars. They are small scars that develop only on one face, they are continuous, aligned, semi-circular (anecdotally quadrangular and triangular scars are seen), with feather and oblique terminations. A point used in defleshing hide also bears a very slight rounding on the tip of the cutting edge.

Cutting soft animal materials with bone contact (soft to medium-ard materials)

228The defleshing disarticulation and possibly skinning operations produce scars that reflect accidental bone contact. The macro-traces observed on the points and characterizing the butchery activities are:

  • fractures: in almost half of the cases, they appear on the experimental items. They are face and burin‑like bending fractures with a feather termination. These fractures appear only under certain conditions: a bone contact, in particular in the context of disarticulation, and the presence of a pointed distal part;

  • scars (figures 75-76): these are scars that appear on both faces of the tools alternately from one face to the other. They are arranged in series, by combinations of scars of various morphology. The combination of various morphologies, and in particular the presence of triangular scars is a reliable indicator of the use of the cutting edges for a butchery activity. The terminations are step or hinge, and the axis of symmetry is deviated (oblique orientation).

Figure 75 - Scarring observed on experimental points used in butchery activities

Figure 75 - Scarring observed on experimental points used in butchery activities

a: combination of quadrangular and triangular scarring with hinge and step terminations; b: combination of quadrangular, trapezoidal, semi-circular, and triangular scarring with feather and step terminations; c: view along the tip of the edge, showing a pattern alternating scarring; d: combination of quadrangular and trapezoidal scarring at the distal extremity of a point

photographs: A. Coudenneau

Figure 76 - Experimental point used in butchery activities and the macro-traces produced during use

Figure 76 - Experimental point used in butchery activities and the macro-traces produced during use

a-c: combinations of scarring observed

Photographs and CAD: A. Coudenneau

Scraping of soft abrasive organic material (hide)

229Hide scraping activities do not produce fractures. It is mainly activities involving contact with dry hide or leather (tanned and softened skin) that produce scars. These scars depend on the nature of the contact. The two Mousterian points used to scrape the dry hide, covered with ashes before tanning, bear bifacial, discontinuous, aligned to superimposed scars, with a combined morphology: quadrangular, semi-circular and trapezoidal, with fine or step and oblique termination. This is a rather atypical distribution for this type of contact that usually produces traces on the face opposite to the contact face. The main feature of the macroscopic traces observed on these items is the presence of a very strong rounding on the active edge. This rounding starts from the cutting edge towards the interior of the item on the face in contact with the hide.

Drilling soft to medium-hard material (hide)

230The actions of punctiform contact on the dry hide produced fractures, but in this case, they are traces little characteristic of the material. They are rather demonstrating the action. These are bending fractures with feather termination (figure 77). More than ever, these fractures alone cannot be diagnostic and must be combined with other legible traces. In the absence of other traces, it will be impossible to conclude when passing to the study of the archaeological material. The leather piercing and perforating activities produced scars that develop alternately on both edges of the point in the distal part. These scars are isolated to aligned. They are all with quadrangular morphology and hinge termination. They are oblique and very small. A marked rounding is also present on the items used in drilling on the edges and on the ridges of the distal part, in contact with the hide. On the other hand, no rounding is observed on the tem used in perforation.

Figure 77 - Mousterian point used in punctiform contact on a tanned hide and the associated macroscopic and microscopic traces

Figure 77 - Mousterian point used in punctiform contact on a tanned hide and the associated macroscopic and microscopic traces

a: scarring and bending fracture

Photographs and CAD: A. Coudenneau

Longitudinal action on medium-hard materials (wood)

231The longitudinal contact actions produced scars on both faces of the active edge. These scars are distributed discontinuously by alternating from one face to the other, thus forming a sinuous cutting edge. There is no exception to this statement amongst the nine points used in longitudinal contact action. The organization of the scars is aligned to superimposed (figure 78a-b). On a point used to incise the bark to remove it, the scars are isolated. The morphologies of the scars are generally quadrangular. These quadrangular scars are very often combined with various morphologies: quadrangular and half-moon or quadrangular and trapezoidal. The mode of action and the state of freshness of the wood do not seem to condition the morphology of the scars. The terminations of the scars are snap, hinge or step. On two points that worked on green wood, we observe feather terminations. We find scars with step termination indifferently on dry or green wood. All the observed scars are oblique.

Figure 78 - Scarring on points related to working wood

Figure 78 - Scarring on points related to working wood

a-b: scarring produced during continuous longitudinal contact; c: scarring produced during percussive longitudinal contact; d: scarring produced during transverse continuous contact; e-f: scarring produced during punctiform contact, observed on the dorsal surface of the point (e) and the ventral surface (f )

photographs: A. Coudenneau

232A rounding is visible on two of the points used in longitudinal action. It is a rounding of low intensity, which develops on the ridges of the negatives of removals of the retouch. This rounding seems related to the friction of the upper parts of the retouched face of the object against the wood during the action.

233We finally notice that there is no significant difference in the development of macro-traces obtained on hardwood or softwood.

Transverse action on medium-hard materials (wood)

234The transverse actions produced scars that develop on only one face: the trailing face, that is to say the face opposite to the leading face. Only three active areas have bifacial scars. These scars are distributed continuously along the active edge and they are aligned with each other. The morphologies of the observed scars are very predominantly quadrangular (11 cases out of 16, that is to say 69 % of the cases). They are rarely combined with other morphologies of scars and when it happens it is with scars of trapezoidal or half-moon morphologies. The terminations of the observed scars are usually feather (figure 78d). For the retouched edges, a hinge or step termination is not excluded. Scars obtained during a transverse contact action on wood are perpendicular. A single point has a very weak rounding on the cutting edge.

Drilling medium-hard materials (wood)

235The points make it possible to perform rotational actions to drill wood. This is one of their functional specificity. We are therefore checking the specificity of the macro-traces obtained for this type of use on the wood. Three points have been dedicated to this use. For each experiment, the distal part of the two edges as well as the convergent extremity were involved. The scars are positioned alternately from one edge of the tool to the other. They are continuous throughout the active area and superimposed. They are scars of quadrangular morphology, but we also observe semi-circular scars on a point used on a dense and dry wood. The terminations of these scars are step (figure 78e-f). The symmetry is variable and depends on the axis in which the action took place. The tool used on green wood has a rounding on the ridges of the point that have been in contact with the wood.

Longitudinal action on hard materials (bone)

236Longitudinal actions produced scars in all cases. They are bifacial and discontinuous scars that alternate from one face of the tool to the other. They are numerous, large in dimensions, aligned to superimposed, of quadrangular and half-moon morphology and with snap and step terminations. No rounding has been produced, certainly because of the continuous scarring of the active edges that prevents the preservation of a rounding.

Transverse action on hard materials (bone)

237The transverse contact actions produced scarring in all cases. These scars appear most often on the face opposite to the trailing face but some of them are found on the leading face. These chips are continuous and superimposed. They are quadrangular and trapezoidal scars with step termination. They are perpendicular (figure 79). No rounding was observed.

Figure 79 - Traces observed on experimental points that came into contact with bone: microscopic scarring produced by scraping the periosteum

Figure 79 - Traces observed on experimental points that came into contact with bone: microscopic scarring produced by scraping the periosteum

Photograph: A. Coudenneau

Punctiform action on hard materials (bone and shellfish)

238The punctiform actions on hard materials produced macro-traces in all cases. The scars are either bifacial or located on one face alternately all along the active area. They are continuous and superimposed. They are quadrangular with a step termination, sometimes feather when it comes to drilling shellfish (figure 80). They are oblique. No rounding was found. At the point, there are fractures and crushing. These are twisted transverse bending fractures. The termination is step. The twisted aspect of the fractures is a valuable indication of the type of motion, as well as the direction of the scars and their position on the item. Crushing is visible on the distal extremity, but also on the sharp ridges of the point. No rounding is observed probably because of the constant scarring that prevents the preservation of a rounding.

Figure 80 - Traces observed on experimental points used to perforate shells. Mousterian point before and after use

Figure 80 - Traces observed on experimental points used to perforate shells. Mousterian point before and after use

a: scarring; b: transverse bending fracture and associated crushing

photographs: A. Coudenneau

d - Synthesis of results from the experimental reference collection study

239As a result of this study, we find that points have been effective for many activities.

240The reference collection of unretouched and retouched points in flint and quartzite was mainly constituted in order to have criteria of interpretation of the traces of use-wear in connection with hunting activities. As such, several results can be retained:

  • the most recurrent fractures of the hunting points are bending fractures (face, transverse or burin-like) with step termination (60 %) and axial, oblique or burin-like apical scarring with step termination (20 %). But only the latter proved to be truly diagnostic of the use of points as throwing or thrusting weapons. They are characterized by the combination of a cone initiation with a “languette” of the bending fracture type and result from a pair of forces that include the impact of the point in the animal and the pressure exerted by the oscillation of the shaft of the spear (in the case of a javelin type throwing) or by its twisting (in the case of a thrusting mode) during the penetration of the point into the animal (Coudenneau, 2013);

  • bending fractures are quite numerous on the points used as hunting weapons, but they can also result from the carcass processing. ”Languettes”, reaching large dimensions, were thus observed on two used points: one, Mousterian, used in complete butchery, disarticulation included, has a transverse burin-like bending fracture 9.8 mm in length and an unretouched pseudo‑Levallois point used to deflesh a rabbit hide has a transverse bending fracture whose “languette” is 3 mm long. Thus, although quite rare on the points used in a continuous contact gesture, their presence alone on a point cannot, even less on an isolated object, indicate with certainty a use as hunting points. These new data thus feed into the debate initiated by our predecessors and we should remain cautious about the interpretation of bending fractures on triangular elements;

  • lateral scars are not strictly speaking a discriminating feature of this type of use. Nevertheless, the presence of quadrangular scars with step termination, or of a combination of half-moon, quadrangular or trapezoidal scars with snap or step terminations, may be a further indication of use as a hunting weapon. These clues gain even more weight if they are associated with an apical scarring.

241The nature and the frequency of the impact traces vary notably according to the morphology of the points in parallel with other variables like the nature and the organization of the materials crossed or touched during the shots (hide, meat, bone: ribs or long bones, …) or the fixing modes of the point on the shaft, as recently mentioned by A. Coudenneau (2013) and V. Rots and H. Plisson (2014). The latter have recently insisted on the need to use an experimental reference collection specific to each chrono-cultural period or even to each case study and to consider the other use- wear traces possibly associated with fractures, without which the risk of misinterpretation could be important. Moreover, several authors, such as J. Pargeter (2011) or L. Chesnaux (2014) have highlighted the fact that fractures usually considered as diagnostic, such as complex bending fractures (whose “languette” is longer than 2 mm) or burin‑like fractures (see for example Fisher et al., 1984; O’Farrell, 2004) can also occur during the debitage of the blank and during trampling. It is therefore important to ensure that the traces are posterior to retouching if retouch is present and to be particularly vigilant in the case of unretouched points, especially if the fractures are observed punctually at the scale of a site (one or two items only, no standardization of the concerned items). The repetition of impact traces on several points of the same serie coupled with the knowledge of the economic and taphonomic context of the site can make it possible to overcome the problems of convergence with other modes of use.

9 - Experimental flake cleaver reference collection

M. Deschamps, É. Claud, D. Colonge, V. Mourre, Ch. Servelle

A - Why this reference collection?

242Prehistorians noted the presence of large tools made from coarse-grained stone in Middle Palaeolithic assemblages from the Vasco‑Cantabrian region from the out set of the 20th century. These tools, referred to as flake cleavers by H. Breuil (1910), were identical to a tool type previously described from Lower Palaeolithic contexts, where they were more frequent. Apart from rare exceptions, Middle Palaeolithic flake cleavers are known uniquely from the Franco-Cantabrian area of the Pyrenees.

243The occurrence of these tool types in Middle Palaeolithic assemblages of this region was initially interpreted as a regional cultural phenomenon (Bordes, 1953a) and continues to be a subject of considerable debate (Benito del Rey 1972-1973, 1983; Jordá Cerdá, 1977; Rodríguez Asensio, 1983; Cabrera Valdès, 1983; Freeman, 1994; Rodríguez Asensio, Arrizabalaga, 2004; Mazo et al., 2012; Álvarez-Alonzo, 2014), with several hypotheses held to account for the presence of flake cleavers in the Franco‑Cantabrian area of the Pyrenees.

244The ‘deterministic’ model for the Lower and Middle Palaeolithic (Villa, 1983; Chauchat, 1985; Rolland, 1988, 1990; Santonja, 1996) sees local raw material properties, coupled with variations in their accessibility according to climatic oscillations, as the principle factors underlying changes in these industries. C. Chauchat considered Mousterian flake cleavers to be a component of Middle Palaeolithic bifacial industries, forming part of the variability of bifaces from Mousterian of Acheulean Tradition (MTA) assemblages. According to this view, “the removal of large flakes was imposed by a raw material that permitted only a minimal degree of retouch” (Chauchat, 1985: 238). This idea is, however, inconsistent with the fact that, on one hand, flint and quartzite bifaces are often found in assemblages containing flake cleavers (Deschamps, 2014) and, on the other hand, that regional flint varieties are primarily used in parallel with the production of flake cleavers at certain sites (Deschamps, 2014).

245The hypothesis proposed by Freeman (1969-1970), where flake cleavers were manufactured to fulfil specific functional needs, implies that the production of different tool types was influenced by the their use and would suggest a functional complementarity between sites with assemblages composed of different tool types. While true that the raw materials employed and the overall morphology of flake cleavers – relatively abrupt retouched lateral edges and an unmodified distal cutting edge opposite a thick base – seem to argue in favour of these heavy‑duty objects being used in activities requiring considerably robust tools, particularly a resistant cutting edge (see for example Latapie, 1956), functional studies of flake cleavers remain extremely rare. A previous functional analysis concerning 11 flake cleavers, most likely Acheulean in age, recovered during surface surveys in the Najerilla Valley, Spain (Utrilla, Mazo, 1996), showed five to exhibit use-related micro‑polishes (4 quartzite examples bore a bifacial micro‑polish typical of woodworking and one flint example had micro-polish attributed to working meat). In addition to this analysis, a more recent study using an experimental approach was designed to test the efficiency of working wood with flint and quartzite flake cleavers (Domingo-Martinez, 2013).

246Finally, seven pieces (three in ophite and four in micaceous lutite) interpreted as flake cleavers from level VII of Amalda (Guipuzcoa, Spain) were studied by J. Rios, who, based on macro-traces, interpreted their having been used for cutting or percussion during heavy duty-activities (Rios-Garaizar, 2012).

247The ways in which flake cleavers were used therefore appears largely unknown.

248In order to explore to what extent cultural, functional and environmental factors underlie the manufacture of flake cleavers in this particular geographic zone during the Middle Palaeolithic, one of us carried out a techno-economic analysis of several Vasconian assemblages (Deschamps, 2014). These initial results concerning the production of flake cleavers and their particular morpho‑functional characteristics subsequently formed the basis for an experimental and use-wear approach designed to investigate their function (Claud et al., 2015). This approach had multiple objectives:

  • first, to evaluate the feasibility of a use-wear analysis of flake cleavers by assessing the preservation of several lithic assemblages and identifying potential use-related wear;

  • build a reference collection of flake cleavers similar in shape and size to those found in archaeological assemblages and used in multiple activities (see Part II, chapter 1);

  • identify the main characteristics of use-related wear from the reference collection in order to analyse archaeological flake cleavers;

  • and finally, determine how Mousterian flake cleavers were used, the nature and hardness of the worked materials as well as the actions employed and potential prehension modes (see Part II, chapters 2 and 4 for results).

B - Production of experimental pieces

a - Flake cleavers

249Blanks were made from quartzite collected from the Pyrenean valleys of the Nive and Neste or the alluvium of the Garonne River in the northern foothills of the Pyrenees, near Mauran, and ophite collected uniquely from the Nive Valley. Both raw materials occur in the form of alluvial pebbles.

250Quartzite blanks were detached from cores stabilised on the ground using hard hammers held in both hands. For the significantly harder ophite, blanks were debited using percussion on a passive hammerstone (figure 81b-c), which proved the only effective technique for detaching large flakes although difficult to identify archaeologically (Mourre, Colonge, 2011). Retouch was carried out exclusively with hard hammers and a tangential blow (figure 81e).

251Blanks were selected according to size, as they needed to be sufficiently large to allow the lateral edges and base to be retouched while maintaining the tool’s large size. The eventual active edge angle also played a decisive role in blank selection. All examples were consistent with a previous analysis of 83 flake cleavers from Abri Olha I that demonstrated working edges to be generally convex or straight with edge angles ranging between 25° and 65° but with a majority measuring between 40° to 50° (Claud et al., 2015). Finally, blanks with neocortical edges were often selected in order to reflect their predominance in Middle Palaeolithic assemblages (type 0 in Tixier, 1956).

252On the other hand, additional characteristics, such as type of edge modifications or base morphology, were not taken systematically into account given the considerable variability of these traits in archaeological contexts. During our experiments, modifications primarily concerned blank morphology and were influenced by whether the flake cleaver would be hafted. For example, pieces sometimes required modification to fit the size of the slot on the handle.

b - Haft production

253The production of hafts takes considerable time for different reasons. First, manufacturing hafts is complex and requires the acquisition and transformation of wood in various stages. Second, multiple hafting arrangements needed to be tested in order to identify which were most consistent with the use-wear traces identified on the archaeological pieces (e.g. straight or curved handle, male or juxtaposed hafting arrangements, parallel or perpendicular tool placement, see Stordeur, 1987; Part I, chapter 2.4).

Figure 81 - Steps in the manufacture of experimental flake cleavers.

Figure 81 - Steps in the manufacture of experimental flake cleavers.

a: the selection of ophite cobbles and a passive hammerstone from the alluvium of the Nive River; b: abandoned core and passive hammerstone; c: refitting of the blanks produced from the core; d: retouch of the blanks; e: production of large blanks; f: photograph of a flake cleaver before use; g: mould of active edge before use

Photographs: PCR Des Traces et des Hommes

254Both straight and curved handles were used during our experiments (figure 82). Flake cleavers hafted onto curved handles were positioned transversally, in juxtaposition or mounted using a male hafting arrangement, while tools mounted onto straight handles involved an exclusively male hafting arrangement. In order to stabilise the haft, voids between the handle and tool were sometimes filled with an adhesive composed of ochre, resin and wax. Finally, leather bindings soaked in water were used to tighten the wood around the tool to avoid it splitting the handle following repeated shocks. Once dry, these bindings proved highly resistant.

255Flake cleavers mounted using a male hafting arrangement were the most effective. However, the addition of an adhesive turned out to be ineffective; either the stability of the haft was compromised from the outset, as the adhesive broke and failed to prevent the tool coming loose from the haft, or the piece wedged itself into the wood during the first moments of use, rendering the resin superfluous or hindering the flake cleaver from becoming fixed in the wood.

256The durability of the hafts also presented certain problems. The same hafts were re-used over a three-year period for each series of experiments. When the wood dried (and the fibres became less supple), the flake cleavers were more difficult to fix in the slots. We therefore soaked the hafts in water for 24 hours, inserted the flake cleavers in the damp wood and left them to dry for another 24 hours. This technique proved to be effective for better fixing the flake cleavers in the haft.

257Trapezoidal flake cleavers with bases slightly straighter than the distal portion were best adapted to a male hafting arrangement, as the base gradually became securely fixed in the slot. The morphology of the distal active edge also played a role in the stability of the hafting arrangement, notably when used to fell trees. It was equally clear that, for flake cleavers with substantially convex distal cutting edges to be effective, the blow needed to be perfectly centred on the convexity, as off-centred shocks tended to wrest the flake cleaver from the handle. Flake cleavers with two edges delineation also presented problems of stability when hafted.

Figure 82 - Hafting used in the experiments in percussion

Figure 82 - Hafting used in the experiments in percussion

On the left, male hafts with straight handles; on the right: juxtaposed and male hafts with curved handles

Photographs: PCR Des Traces et des Hommes

C - The effectiveness of flake cleavers for various activities

258The overall experimental assemblages comprised 52 flake cleavers that were used in a variety of activities reflecting those suspected to have been carried out by Neanderthal groups. These activities primarily concerned the acquisition and exploitation of either plant or animal resources according to different stages of the overall chaîne opératoire (Deschamps et al., 2011; Part I, chapter 1).

259As ophite and quartzite proved to function in a similar fashion in all activities, these two raw materials are presented together.

260Tools were used hafted, hand-held or wrapped in leather. Hafted flake cleavers were most frequent, represented by 31 examples fixed in straight (n=27) or curved (n=4) handles. These hafted tools were primarily used to work wood, followed by the breaking down of carcasses and fracturing long bones (n=9). Use-time varied, ranging from 5 minutes to several hours in order to evaluate and compare edges at different stage of wear.

261Direct percussive motions were most often employed for two reasons. First, an initial evaluation of the archaeological material (see Part II, chapter 2) demonstrated flake cleavers to exhibit large scars and fractures, whose intensity was consistent only with use in a percussive motion (see Part I, chapter 2.5). We therefore decided to test this assumption by using flake cleavers in a variety of ways, varying both the material worked and the hafting arrangement. It is important to note that, while these types of removals have previously been interpreted as reflecting resharpening episodes, their often bifacial distribution, discontinuous organisation, and more abrupt angle compared to that of the initial unmodified edge argue for a use-related origin. This damage resulting from taphonomic processes can also be ruled out given the morphology of the removals (frequent invasive removals with bending initiations), their large size (natural phenomenon tend to produce smaller scars or fractures) and location (systematically present on the distal cutting edge).

262Second, the active zone of flake cleavers comprises an unmodified edge and the types of damage sustained by this type of edge during contact activities (scraping, cutting) was investigated in detail using the quartzite flake reference collection (see Part I, chapter 2.6). With that said, several experiments did involve flake cleavers used to scrap and cut in order to compare the formation of use-wear traces with those exhibited on flakes, particularly ophite flakes, during contact and percussive motions. This primarily concerned the quantity and size of scars, as well as the effectiveness of flake cleavers for these activities.

a - Woodworking

263Twenty-five flake cleavers were used in a variety of woodworking activities involving different types of motions (see figures 4-6), but primarily concerned the felling of trees (18 flake cleavers). Only three examples were hafted as adzes on curved handles, and two were used in the bare hands with a leather wrapping. Not surprisingly, felling trees with a flake cleaver held in the bare hands proved both laborious and painful, with the arms of the user being struck numerous times. For the same species (ash) with comparable diameters (between 8 to 10 cm), 752 blows were required with an unhafted flake cleaver versus 140 and 480 blows in the two tests using a flake cleaver hafted onto a straight handle. The second unhafted flake cleaver failed to fell a tree with a 12 cm diameter trunk; successive scarring blunted the edge and a reduced transfer of force compared to a hafted cleaver led to a dramatic reduction in effectiveness.

264Curved handles also proved less efficient than straight examples due to the significantly greater number of blows needed to fell the same species of tree with equivalent diameters (approximately 1 300 versus 625 blows on average). Moreover, numerous percussion experiments using hafted flake cleavers were carried out as this mode of use produced damage of comparable intensity with that frequently observed on archaeological examples (see Part II, chapters 2 and 4).

265In order to explore whether damage varied according to the hardness of the wood, we felled both softwood (e.g. poplar, will, and maple) and hardwood species (e.g. ash and boxwood), as well as species between the two (e.g. cherry). Trunk diameters ranged between 5 and 17 cm, with the number of blows delivered counting between 16 and 1697. Certain flake cleavers saw very little use due to difficulties keeping the tool securely fixed in the haft. These tests also allowed the intensity of use-wear formation on the active edges of flake cleavers to be evaluated. During the tree felling experiments, scars, which sometimes measured several centimetres, formed during the first few minutes of use. The working edge than stabilised before a further series of scars formed, considerably modifying the edge and resulting in an edge angle of around 80°. However, this did not have repercussions for the effectiveness of the tool, which is likely connected to its inbuilt inertia.

266Three flake cleavers were used to shape wood while either held in the bare hands or hafted. One example was used in scraping and longitudinal percussive motions to shape a handle from an ash branch. While the use of a short handle (operable with a single hand) substantially increased efficiency, the weight of the tool sometimes proved ill-adapted to the precision required to shape wood, which demands less force and more precision compared to felling trees.

267One flake cleaver hafted onto a curved handle also served to scrap and pound the bark from a poplar trunk using an oblique transverse motion. The bark was first notched by light percussion and then scraped or torn away by hand. The tool proved poorly adapted to this type of activity, as it penetrated too deep unto the trunk due to the narrow positive rake angle. Finally, removing bark by hand or scraping following the incision of several notches using a longitudinal action proved quicker and more efficient.

268In the end, a smaller number of experiments involving the removal of bark and shaping wood were carried out given the ineffectiveness of flake cleavers for these activities.

269One piece was used to split freshly cut poplar logs set on their sides. The inertia of the tool, which was hafted onto a particularly long and heavy handle, was sufficient to split the logs. This splitting experiment is, however, incompatible with the type of wood splitting that was likely carried out during the Palaeolithic given the regularity of the logs cut with a modern saw. In the ethnographic literature, wood splitting generally concerns entire trunks posed horizontally and split lengthwise using laterally posed wedges (Pétrequin, Pétrequin, 1993). Our wood splitting experiment therefore has little archaeological relevance. Moreover, we did not test splitting entire trunks using flake cleavers as wedges due to the fact that no archaeological examples exhibited traces of percussion on the base consistent with this potential use.

270Finally, we tested the effectiveness of bucking dry, dead, or burnt wood, with the goal of reproducing damage associated with working the hardest possible wood. This damage is potentially more consequential than that produced when felling trees. Two flake cleavers hafted onto straight handles were used to split logs using oblique longitudinal percussion. Splitting dried wood and partially burnt wood proved impossible; the tool penetrated the first few centimetres beneath the bark or into the partially burnt areas, but the heart of the wood or non-burnt areas was extremely hard. The tool bounced off the surface without penetrating the wood fibres.

b - Working animal materials

271The processing of animal resources groups together several different types of activities (see Part I, chapter 1.3), whose main objectives include the removal of meat for consumption as well as the skin, tendons and certain bones for a variety of needs (retouchers, accessing marrow). Twenty-seven flake cleavers were used to skin, remove meat, disarticulate carcasses, collect tendons, fracture long bones, and flesh skins (figures 11, 17c-d, 18c). Three deer and a lamb carcass, the leg and axial skeleton (spinal column, sternum, ribs) of a bison, two cow femurs and the spinal column of a horse were fractured.

272Skinning, fleshing, meat removal and disarticulation were exclusively performed using hand-held tools or tools wrapped in leather. On the other hand, disarticulation and fracturing long bones were for the most part done using hafted cleavers. Certain pieces served uniquely to remove meat (n=4), fracture bone (n=3) or flesh carcasses (n=3), while others were used for both skinning and disarticulation (n=3), removing meat and cutting articulations (n=4), disarticulation and fracturing bones by percussion (n=8), or multiple tasks until the tool was no longer effective (n=2). In the end, all stages of the butchery processes were represented in the experiments.

273Flake cleavers proved to be particularly effective for skinning, as the long convex edges minimised the risk of piercing the skin. On the other hand, while flake cleavers were useful for removing large segments of meat, they were poorly adapted to removing meat from the lower portions of the limbs, as the large size of the tool reduced precision. Moreover, the weight of the objects (between 100 and 900 g) introduced an additional difficulty (i.e. muscle fatigue).

274Cutting articulations also posed problems; the morphology of the edge, especially the lack of a point or angular zone, and the overall thickness of the tool were incompatible with the precision required for this activity.

275Conversely, flake cleavers proved well adapted to disarticulating carcasses using percussion and fracturing. Although difficult to completely disarticulate a carcass when hand held (it was possible to separate the ribs from the sternum of the lamb or deer, but not to fracture it), hafted flake cleavers allowed the vertebrae to be disarticulated and the ribs to be separated from the sternum. They were also successful used to section the bison sternum and fracture the horse long bones with a limited number of blows (between 3 and 43). This often led to the rapid development of large use-related damage on the edges of hafted cleavers without reducing the tool’s effectiveness. In the end, more time was needed to complete the same tasks using hand‑held cleavers compared to hafted examples, despite the former being used to disarticulate smaller carcasses (deer and lamb versus bison and horse).

276Fleshing skins consisted of removing meat and fat residues from fresh or dried skins stretched in a wooden frame or on the ground with pegs. A quartzite and an ophite cleaver were used un-hafted on fresh skins in a tangential motion and an ophite cleaver with leading angle greater than 90° was used to scrape a hide pre-treated with cinder in order to avoid it rotting.

D - Description of use-wear traces on the experimental flake cleavers

a - A low-powered approach

277The experimental flake cleaver reference collection was analysed using a stereomicroscope at low magnifications for several reasons:

  • a significant number of tests involving percussive motions produced very little microscopic wear. In fact, heavy edge wear left little time for micro-polishes to develop, meaning they were generally absent or insufficiently developed to be diagnostic (Part I, chapter 2.5.B);

  • micro-polishes typical of working each tested material with different actions were already documented during the study of the quartzite flake reference collection (Part I, chapter 2.6);

  • the archaeological assemblages studied as part of this project were not sufficiently well preserved to permit a high-powered approach (Part II, chapter 2).

278Of course, an eventual microscopic analysis of the flake cleaver reference collection is still possible, especially should flake cleavers with preserved micro-polishes be discovered.

279Our objective was to identify the key macroscopic traits that help identify how Vasconian flake cleavers functioned.

280As no differences were observed between the types of wear that formed on quartzite and ophite cleavers used in the same activities, the two raw materials are presented together.

b - Characteristics of macro-traces according to action and contact material hardness

Use-wear produced during continuous contact with the material

281- Cutting soft animal materials without contact with the bone.
The two flake cleavers used to flesh fresh hides exhibited very little wear. Scarring was absent or extremely limited, small in size (less than 1 mm) and isolated on the piece. Oriented obliquely, scars were half moon‑shaped with bending initiations and transverse terminations. The cutting edge was unaltered with no clearly visible rounding, which could be linked to the duration of use or the accumulation of residues (e.g. fats, grease, tissue) in the lower portions of the tool’s surface precluding the development of wear. Cutting edge angles remaining unchanged.

282- Cutting soft animal materials incurring accidental contact with the bone
Flake cleavers used in skinning, fleshing and disarticulation (cutting) exhibited comparable wear between specimens but more marked wear compared to the examples described above (figures 83
a, 89). Scarring was discontinuous, isolated or aligned, and small in size (3.9 mm long on average and 2 mm wide for the largest scar observed on the same piece), invisible or practically invisible to the naked eye.
Different scar morphologies could occur on the same edge (semi-circular, half moon‑shaped, triangular) and initiations can be bending or coned. Terminations were often feathered or stepped, rarely transverse, and occur at extremely low angles.
Scarring was of variable elongation and orientated obliquely to the edge. Although more than 20 scars were sometimes present, the cutting edge was most often left intact or presented a light rounding. Scarring did not alter the edge angle.
No single criterion or suite of macroscopic criteria was sufficiently diagnostic of the different butchery stages involving continuous contact with the material (skinning versus meat removal or disarticulation). With that said, a significant number of flake cleavers were used in consecutive butchery stages in order that they reflected a general use in butchery activities and allowed differences between use-wear traces produced by other activities, particularly percussive motions, to be reliably discerned.

283- Scraping soft abrasive materials (cinder-covered dry skin)
An ophite flake cleaver used to scrape dry skins exhibited three discontinuous, medium-sized (3 × 1 mm), abrupt to semi-abrupt scars with transverse and stepped terminations on one surface. This damage occurred at the beginning of use as the edge stabilised. Moreover, while the cutting angle remained unchanged, a clear transverse rounding developed on the tool’s lower surface (figure 83
b).

284- Scraping medium-hard materials (wood)
The flake cleaver used to shape a wooden handle using a mix of direct percussion and scraping exhibited scarring typical of a transverse action on a medium-hard material; scars were small, short, and quadrangular with bending initiations (figure 83
c).

Use-wear connected to percussive motions

285- Percussion of medium-hard (woody) materials with hafted flake cleavers
Felling trees with hafted flake cleavers used in oblique percussion produced large scars on the distal end of tools that were easily visible to the naked eye (figures 83e, 86-87). The largest scar observed on each piece was on average between 14 mm long and 7 mm wide. Most scars were continuous and aligned with bending initiations, creating an abrupt aspect that significantly increased the edge angle (23° greater on average). Semi‑circular and half-moon shaped with both feathered and stepped terminations were most frequent. Certain scars removed a large portion of the edge in the form of complex bending fractures with languettes measuring between 4 and 14 mm. The cutting edge rarely remained intact. In fact, small, difficult to characterise scars (never more than 20) or crushing could be responsible for the loss of crystals that produced a lightly rounded or regularised edge. The observation of pieces during use also showed the formation of larger scars to potentially remove previous traces of wear. The majority of flake cleavers were several millimetres shorter after use.

286Although scarring was predominantly bifacial, the number and size of scars was organised asymmetrically on the two surfaces. The examination of wear formed during the use of several flake cleavers revealed large scars to generally develop on the trailing edge, which, given the oblique motion used, would be the upper surface of the tool. With flake cleavers hafted onto straight handles, turning the piece in the slot or inversing the way in which the handle was held, modifying the direction of percussion, logically produced scarring on the opposite surface. The development of scars on this surface could also result from an intentional or accidental change in the angle of movement during percussion as the edge became stuck in the wood, twisted and then broke.

Figure 83 - Overview of macro-traces produced during experimentation according to gesture and the hardness of the material worked

Figure 83 - Overview of macro-traces produced during experimentation according to gesture and the hardness of the material worked

Photographs and CAD: É. Claud and M. Deschamps

287These general descriptions obviously masks a certain degree of variability in use-wear traces that results from a combination of factors, principally the duration of use, the angle of the active edge, haft type (straight versus curved), and the hardness of the wood. As use time increases, the number of scars grows and the tool becomes shorter. Conversely, the edge is more resistant as the edge angle increases (around 60°), resulting in the formation of fewer scars. It was also the case that half-moon shaped scars with transverse terminations developed more readily on edges with cutting angles of less than 50°. The regularisation of the cutting edge is more intense and particularly visible on one of the two surfaces of flake cleavers hafted in curved handles. In terms of the influence of wood hardness, our analysis equally showed that the rare cases of superimposed scars resulted uniquely from felling ash or boxwood, the two hardest species tested (figure 83h).

288Flake cleavers used to pound and remove the bark of poplar trees or shape wooden objects exhibited scars comparable to those produced when felling trees. The size of scars resulting from these activities nevertheless fell within the lowest values recorded for felling trees (8 and 9 mm in length and 3 and 5 mm in width).

289- Percussion of medium-hard materials (wood) with a hand-held flake cleaver
Although felling trees with flake cleavers held in the bare hand resulted in the same types of wear described for hafted tools (figures 85-86), the scars were smaller (see above; 12 and 7 mm in length and 5 and 3 mm in width for the largest scars, figures 83
d, 85).

290- Percussion of hard materials (dry and burnt wood) with hafted flake cleavers
The two flake cleavers used to buck dry and burnt ash (figure 84) exhibited particularly large scars (14 and 22 mm long and 8 and 33 mm wide) that fell at the upper limit of the size variability documented for use-wear traces from felling trees (28 mm in length and a maximum of 20 mm in width). The characteristics of the scars were similar to those recorded for felling trees, although scarring on the second flake cleaver was sometimes superimposed (i.e. two visible scarring episodes). This piece also incurred a lateral fracture (complex oblique bending fracture) during use, which can be linked to the hardness of the wood worked (dry ash). Finally, this flake cleaver exhibited no edge crushing due to it being abandoned after the formation of the large scars and the fracture. It should be recalled that we were unsuccessful in bucking wood with a flake cleaver, leaving open the question as to whether Neanderthal groups used flake cleavers for this task.

291- Percussion of hard materials (bone) with flake cleavers held in the bare hand
The two flake cleavers used in the bare hands to section the sternum of an adult deer and an immature sheep exhibited scarring barely visible to the naked eye (4 × 3 mm, figure 88). This damage was bifacial, continuous, aligned, sometimes superimposed, and numerous, with the number of scarring generations ranging between 1 and 3 (figure 83
f,i). Scar morphology was variable (semi‑circular, trapezoidal, triangular, half-moon), terminations were either feathered or stepped with cone initiations being most frequent, removals were at low-angles (shallow) and oriented obliquely or perpendicularly. The cutting edge remained intact or was slightly crushed, with the cutting angle only slightly altered.

292- Percussion of hard materials (bone) with hafted flake cleavers
A hafted flake cleaver used in perpendicular percussive motions to fracture carcasses exhibited damage that varied in intensity depending on the animal (figures 90-92), which would be consistent with differences in the hardness and thickness of bones.

Figure 84 - Schematic illustration of the macro-traces observed on a hafted flake cleaver used in percussion on dry oak

Figure 84 - Schematic illustration of the macro-traces observed on a hafted flake cleaver used in percussion on dry oak

Photographs and CAD: M. Deschamps

293The flake cleavers used to process the bison leg, the axial skeleton of the horse and bison, as well as the cow femur, all from adult individuals, bore heavy use-wear scars, except for one example that was used to split a cow long bone using three blows. Scarring was bifacial and visible to the naked eye, large in size (10 and 38 mm in length and 8 and 30 mm in width for the largest examples), numerous (>20), and superimposed (up to five generations figure 83j). Damage developed preferentially on the trailing edge when the motion was oblique. Scar morphologies were semi‑circular, trapezoidal, or triangular. Cone initiations were most frequent, although the occasional bending initiation was also recorded, and terminations were stepped and more rarely feathered. One bending fracture without a languette was produced obliquely to the lateral portion of the flake cleaver. Finally, average cutting angle increased markedly by 13° and 26°.

294On the other hand, flake cleavers used to breakdown the lamb carcass exhibited smaller scars (7 and 9 mm long and 2 mm wide for the largest examples, figure 83g) that were rarely superimposed. Half-moon scars with feather terminations occurred alongside semi-circular, trapezoidal and triangular examples. The cutting angle remained unchanged or was only slightly altered.

295For both types of carcasses, scars were low-angled in a majority of cases, including along the cutting edge. The few scars with bending initiations sometimes tended to give the cutting edge a semi‑abrupt aspect, although this was considerably less flagrant than in the case of percussion on wood (see above). Moreover, the cutting edge remained sharp or intact, without any crushing regularising the edge.

296No macroscopic hafting traces were identified on any of the hafted flake cleavers except for one example where a small, approximately 5 mm scar, indistinguishable from retouch, was detached from the base. This damage is, however, not diagnostic of the presence of a haft.

Figure 85 - Schematic illustration of the macro-traces observed on a hand-held flake cleaver used in percussion on green ash wood

Figure 85 - Schematic illustration of the macro-traces observed on a hand-held flake cleaver used in percussion on green ash wood

Photographs and CAD: M. Deschamps

Figure 86 - Schematic illustration of the macro-traces observed on a hafted flake cleaver used in percussion on green ash wood

Figure 86 - Schematic illustration of the macro-traces observed on a hafted flake cleaver used in percussion on green ash wood

Photographs and CAD: M. Deschamps

Figure 87 - Schematic illustration of the macro-traces observed on a hafted flake cleaver used on green poplar

Figure 87 - Schematic illustration of the macro-traces observed on a hafted flake cleaver used on green poplar

Photographs and CAD: M. Deschamps

Figure 88 - Schematic illustration of the macro-traces observed on a hand-held flake cleaver used in percussion in the course of butchery activities

Figure 88 - Schematic illustration of the macro-traces observed on a hand-held flake cleaver used in percussion in the course of butchery activities

Photographs and CAD: M. Deschamps

Figure 89 - Schematic illustration of the macro-traces observed on a flake cleaver used for cutting during butchery activities

Figure 89 - Schematic illustration of the macro-traces observed on a flake cleaver used for cutting during butchery activities

Photographs and CAD: M. Deschamps

Figure 90 - Schematic illustration of the macro-traces observed on a hafted cleaver used in percussion for the disarticulation of a horse carcass

Figure 90 - Schematic illustration of the macro-traces observed on a hafted cleaver used in percussion for the disarticulation of a horse carcass

Photographs and CAD: M. Deschamps

Figure 91 - Schematic illustration of the macro-traces observed on a hafted cleaver used in percussion for the disarticulation of a lamb

Figure 91 - Schematic illustration of the macro-traces observed on a hafted cleaver used in percussion for the disarticulation of a lamb

Photographs and CAD: M. Deschamps

Figure 92 - Schematic illustration of the macro-traces observed on a hafted cleaver after its use in percussion for the disarticulation of a bison

Figure 92 - Schematic illustration of the macro-traces observed on a hafted cleaver after its use in percussion for the disarticulation of a bison

Photographs and CAD: M. Deschamps

E - The informative potential of macro-traces on flake cleavers

297All of the experimental flake cleavers exhibited some form of macroscopic use-wear, most frequently scarring, demonstrating a low-powered approach to be informative despite the common perception of quartzite and ophite edges being highly resistant.

298Our study of macroscopic use-wear traces on experimental flake cleavers produced recurrent traits characteristic of both the type of use and differences between distinct functions, confirming their diagnostic character. Fixed butchery actions (cutting and scraping), woodworking and hide processing produced edge rounding and, more frequently, microscopic scars. Their characteristics vary according to the material worked and the type of action employed, similar to what was seen with the quartzite flake reference collection (Part I, chapter 2.6). Identifying and interpreting this type of evidence on archaeological pieces should therefore be unproblematic provided edges are sufficiently well preserved. It should also be highlighted that flake cleavers used to cut soft animal materials without touching the bone (fleshing hides) exhibited very light macro‑traces, similar to what was seen with unmodified flakes or denticulates. This evidence is difficult to reliably interpret, even in cases of well-preserved archaeological assemblages. The identification of tools used to cut these materials is therefore dependent on preservation conditions and the use of a microscope.

299Percussion motions produced more heavily developed use-wear traces; primarily in the form of relatively frequent scarring of variable size. Hafted flake cleavers clearly exhibited larger scars compared to hand-held examples for the same action and raw material. This damage was consistently visible to the naked eye, except on flake cleavers used to process the relatively less resistant lamb carcass.

300A certain degree of variability in macro-traces was evident with flake cleavers used to breakdown carcasses and fell trees. This variability was connected to working edge angle, hardness of the wood or bone, duration of use, and type of hafting arrangement. Two groups of scars were nevertheless identifiable, certain of which were uniquely associated with working wood (a medium-hard material) and others with breaking down carcasses (frequent contact with hard material, i.e. bone) (figure 83). Several flake cleavers exhibited scarring whose characteristics fell somewhere between or were shared by two modes of use (e.g. superimposed but abrupt edge scarring or low-angled scars with limited overlapping, figure 83g-h). These convergent traits are probably partially due to continuity in the hardness of the contact material. “Edge condition” (i.e. intact or crushed), a trait that can help determine the hardness of the material worked, proved reliable only in cases of well-preserved edges, which was rarely the case with archaeological flake cleavers studied in this project. A certain grey area therefore existed for use-wear traces that precluded reliably determining the hardness of the material worked by certain archaeological flake cleavers (Part II, chapter 2).

301This experimental reference collection therefore provides a means for assessing use-related wear on archaeological flake cleavers in order to identify the type of action (percussion versus continuous contact) and the type of material worked when the most characteristic wear is present. The use of experimental flake cleavers in different activities equally allowed the relative effectiveness of this tool type to be tested, confirming its suitability for performing heavy-duty tasks, namely felling trees and breaking down carcasses.

10 - The experimental biface reference collection

É. Claud, S. Maury, V. Mourre, M. Brenet, D. Colonge

A - Why this reference collection? Current understandings of biface function and use

302Since the 19th century, bifaces have often been considered as multi-functional tools given their morphology and traces of use (e.g. Macalister, 1921; Posnansky, 1959; Soriano, 2000; Soressi, 2002). Although these tools are a consistent component of Acheulean and Middle Palaeolithic assemblages in Europe, dedicated use-wear analyses remain relatively rare (see tables 20-21 and Claud, 2008 for a detailed synthesis) and generally concern small samples (but see Veil et al., 1994; Mitchell, 1998; Soressi, Hays, 2003). While traces referable to butchery are most frequent on Acheulean bifaces, these tools were also sometimes used to work wood or dig and occasionally exhibit damage linked to percussion on hard materials. Mousterian bifaces also exhibit use-wear connected to butchery, as well as damage typical of cutting wood or skins. On the other hand, bifacial points from the African Middle Stone Age or Eastern European Middle Palaeolithic seemed to have been used primarily to arm hunting weapons. Morpho-functional analyses of Mousterian bifaces (Soriano, 2000; Soressi, 2002) demonstrated their potentially multifunctional edges to have prolonged use-lives. This is connected to both the durability of their edges and their plano‑convex cross‑sections that permit multiple episodes of resharpening. Techno‑economic analyses have also demonstrated the high mobility of these tools (see for example Geneste, 1985; Turq, 2000; Faivre, 2006; Turq et al., 2017) in relation to their long use-life and ability to fulfil multiple tasks. However, given the rarity of use-wear analyses, is it not perhaps slightly premature to consider bifaces as genuine “Swiss army knives”? Moreover, S. Soriano (2000) and É. Boëda’s (2001) work with techno-functional units differentiated bifacial tools from bifacial tool blanks. Similarly, bifaces with different morphologies could reflect distinct uses (Albrecht, Müller-Beck, 1988; Phillipson, 1997), which are susceptible to change as the bifaces passes through various stages of reduction (Boëda et al., 2004; Brenet et al., 2004). This being the case, the multi-functional hypothesis needs to be tested against both techno-morphological data and use-wear evidence.

Table 20 - List of assemblages attributed to the Acheulean in which use-wear on bifacial pieces is mentioned, based on studies published as of 2010

Table 20 - List of assemblages attributed to the Acheulean in which use-wear on bifacial pieces is mentioned, based on studies published as of 2010

A cross indicates that the number of pieces or active zones is unknown; in this case, the minimum number (one) is used in the calculation of totals“

Table 21 - List of assemblages dated to the Early or Middle Palaeolithic (aside from the Acheulean) in which traces of use were mentioned on bifacial pieces, as published in 2010

Table 21 - List of assemblages dated to the Early or Middle Palaeolithic (aside from the Acheulean) in which traces of use were mentioned on bifacial pieces, as published in 2010

A cross indicates that the number of pieces or active zones is unknown; in this case, the minimum number (one) is used in the calculation of totals“.

303With this in mind, we undertook new functional analyses of bifaces from Mousterian of Acheulean Tradition (MTA) assemblages recovered from sites in south-west of France; Chez-Pinaud (Jonzac) in the Charente-Maritime, Fonseigner in the Dordogne, Bayonne le Prissé in the Pyrénées‑Atlantiques, as well as three open-air sites from the Bergerac region of the Dordogne (Combe Brune 2, La Graulet and La Conne de Bergerac) (see Claud, 2008, 2012, 2014b; Part II, chapters 1 and 2). The construction of an experimental biface reference collection was therefore instrumental for identifying criteria capable of distinguishing use-related wear from technological evidence and natural alterations as well as differences in use-wear patterns that could potentially reflect different uses. Here we adopted a low-powered approach using a stereomicroscope for several reasons. First, it is now well understood that macro-traces (fractures, scarring and rounding) observed under low-magnification are informative for identifying tool function, especially for lithics (see Part I, chapters 2.2 and 2.5). Secondly, macro-traces are larger and generally better preserved than micro-traces. Finally, scarring characteristics have been shown to be heavily influenced by edge morphology (see Part I, chapter 2.2), a factor that is currently under documented and poorly illustrated. The effect of retouch on the formation of wear, for example, has not been systematically explored and available reference collections concern primarily unmodified edges.

B - Experimentation

a - Biface production

304Experimental bifaces were manufactured using Middle Palaeolithic techniques and methods, notably those documented for the MTA. Blanks – often flakes – were shaped and sharpened by soft-hammer percussion using an organic or mineral hammer and tangential blows. The majority of the bifaces replicas had two convergent, retouched edges and a non-cutting base (Soressi, 2002), typical of ‘classic’ MTA bifaces (figures 93, 95). The angle between the two surfaces of these plano‑convex tools was shown to measure between 40° and 50°. Bifaces with biplane cross-sections and transverse cutting edges with equivalent edge angles were less frequent and similar to those found on certain sites excavated during work on the Bergerac bypass road (figure 94). Several side scrapers on bifacial blanks were also used. All pieces were manufactured from brown (“Grain de Mil”) or black Senonian or Masstrichtian flint. More details concerning the technological characteristics of the experimental bifaces can be found in Annex 3 (M. Brenet).

b - Activities performed and observations concerning technical effectiveness

305One hundred and thirteen tests were carried out involving 81 bifacial tools. A single piece can in fact have multiple potential working edges (see for example figure 93). Apart from several hafted pieces, bifaces were generally used in the bare hands or wrapped in leather or plant fibres (see Annex 3). The lateral convergent edges, point, and transverse cutting edges were most often used.

Figure 93 - Schematic illustration of the macro-traces observed on a flint biface used to scrape and saw dry wood

Figure 93 - Schematic illustration of the macro-traces observed on a flint biface used to scrape and saw dry wood

Photographs and CAD: É. Claud and M. Coutureau

306Duration of use varied between 3 and 110 minutes, but most often averaged between 10 to 30 minutes. Experimental bifaces were primarily used in butchery, woodworking, and hide processing using various actions (cutting, scraping, percussion, piercing).

Woodworking

307Sixty-two tests were carried out using different species of fresh, dry, or burnt wood (primarily oak and poplar).

308Twenty-eight edges were used to scrape bark or thin or regularise the surface of wooden objects such as handles or spears. Scraping, although possible, was difficult due to the fact that the prehensile zone of bifaces with convergent edges is not directly opposite the active edge.

309A smaller number of bifaces were used to groove (4 active zones), saw (6), split (10) or pierce (3) wood. Bifaces proved ill adapted to sawing, as this activity requires a regular straight profile adequate denticulation and a certain degree of finesse. The tips of bifaces also proved unsuitable for piercing given the lack of a solid trihedral geometry and an edge profile combining convex and flat surfaces. Moreover, the often-thin tips of bifaces are fragile and the relatively wide angle formed by the convergent edges makes piercing a deep hole difficult. Creating a deep hole is, in fact, only possible by investing a significant amount of time and removing a considerable amount of wood.

Figure 94 - Schematic illustration of the macro-traces observed on a flint biface used in percussion to fell a tree

Figure 94 - Schematic illustration of the macro-traces observed on a flint biface used in percussion to fell a tree

Photographs and CAD: É. Claud and M. Coutureau

Figure 95 - Schematic illustration of the macro-traces observed on a flint biface used in cutting during butchery

Figure 95 - Schematic illustration of the macro-traces observed on a flint biface used in cutting during butchery

Photographs and CAD: É. Claud and M. Coutureau

310Eleven bifaces, some hafted, were used in a percussive motion to fell small trees (trunks measuring around 10 cm). The active zones concerned a lateral edge or base for hafted examples and a lateral or transverse edge for those wrapped in leather or plant fibres (see figure 94 for examples). We tested pinch hafts and male, lateral, axial and transverse hafting arrangements bound with bindings made from tendons or plant fibres or adhesives composed of pine resin, bees wax, and ochre. The working edges were sufficiently resistant and sharp for chopping wood, with the haft providing a comfortable grip that allowed for large movements and increased force in cases of heavy handles held in both hands. With that said, we did not find a sufficiently stable combination that allowed the tool to be used for more than 15 minutes. In fact, the bifaces gradually began to shift and eventually came loose from the haft or they remained stable for a few minutes before the resin cracked or the piece broke and was violently ejected from the haft. The same problem was encountered for hafted flake cleavers used for percussion (see Part I, chapter 2.9). Experiments using unhafted bifaces demonstrated that, although it was possible to fell trees, after a few minutes the blows became painful for the hands and arms. This being the case, if bifaces were used to fell trees, we could expect them to have either been hafted in an arrangement we didn’t test or have a different morphology (i.e. form of the base) providing more stability under the conditions we tested.

Butchery

311Bifaces were used in 22 butchery experiments involving different animals (sheep, deer, cow, horse). Tools were used in all the different stages of the process: skinning (1), disarticulation and cutting away meat (12), percussive motions (7), and scraping the periosteum (2).

312Bifaces proved particularly effective butchery tools, with the sharp tip formed by the two resistant edges easily able to penetrate the flesh and perforate the skin. Compared to unmodified flakes retouch gives the edge the ‘biting’ capacity necessary to cut certain strong tissues such as aponeuroses. The effectiveness of bifaces in butchery activities, notably experimental replicas of Acheulean examples, had been documented in previous experiments as well as those carried out at the same time as our collective research project (Jones, 1980; Mitchell, 1995, 1998; Gálan, Domínguez-Rodrigo, 2014).

313During skinning, particularly tangential movements designed to remove the very tender fatty tissue attaching the skin to the carcass, the tip did not prove especially advantageous, as it often pierced holes in the skin despite careful attention to avoid doing so. On the other hand, the tip is extremely useful for cutting into the skin at the outset of the skinning process.

314Meat was removed with longitudinal cutting motions, except in the cases of bifaces with transverse cutting edges similar to examples found during work on the Bergerac bypass road. The distal edges of these pieces exhibited scarring consistent with percussion on a medium-hard material. The low-angle of the transverse cutting edge was used to cut away meat and disarticulate the hind legs of a cow. While we were sceptical at first this technique ultimately proved effective for removing meat from carcasses: striking the tissue enveloping the muscle both ruptured it and split the muscle in two. Additionally, percussion using edges of this type severed thick tendons (diameters greater than 1 cm) in one or two blows. Bifaces therefore proved to be particularly efficient for disarticulating carcasses. The lateral edges of these bifaces were equally effective for cutting muscles or tendons in difficult areas or when they weren’t taut. In fact, when tendons were detached from the bone on one or both sides, it became impossible to cut them by percussion. On the other hand, using the distal cutting edges in percussive motions was only effective for the first 15 to 20 minutes, as the edge wore rapidly, whereas bifaces with convergent edges used in cutting motions had much longer use‑lives, becoming less efficient only after around an hour of use in cutting away meat or disarticulating carcasses; the edge tended to slide over the meat or required considerably more force than was originally necessary.

315In contrast to those of Gálan and Domínguez-Rodrigo (2014), our results did not show bifaces to be less efficient for disarticulation compared to removing meat. Of course, while we did not adopt the same experimental protocol (recording the time necessary for each stage of carcass processing for the same species according to each broad tool type: unmodified and retouched flakes, bifaces), the differences between the two studies more likely reflect that fact that, based on the photos provided, they used more squat, less pointed and probably less sharp bifaces than those tested in our project. This illustrates the difficulties, even impossibility of comparing the effectiveness of tools, or any other type of data for that matter, using relatively broad categories such as “unmodified flakes”, “retouched flakes” or “bifaces”. Each of these categories subsumes an extremely high degree of both technological and morphological variability, which could imply equally different functional potential. Finally, this divergence could also be due to differences in the butchered species (deer for Gálan and Domínguez-Rodrigo, 2014 versus cow, sheep and deer in our study), as large bovids do not necessarily present the same constraints during disarticulation as cervids.

316Butchery experiments using bifaces with their proximal-lateral portions fixed in a male hafting arrangement demonstrated the haft not to impede the tool’s use provided that the distal portion of the haft is finely shaped and does present an obstacle for penetrating the flesh. A composite tool can be used with both hands, increasing the force applied. Nevertheless, the manufacture and management of the composite tool requires a heavy time investment (e.g. hollowing out the area for the biface, working the distal end, preparing the adhesive, gluing the biface, removing it for resharpening and then re-gluing it in the haft) compared to the benefits gained, especially as hand-held bifaces proved to be equally effective for butchery.

Hideworking

317Seven hideworking experiments were carried out using bifaces to cut (3), scrape (2), and pierce (2) dry and fresh hides (deer, sheep, figure 18a-b). The long, regular edges, rectilinear in profile and convex in plan, were used to flesh fresh hides using a tangential cutting motion in order to remove as much flesh as possible without piercing the skin (Plisson, 1985). The biface edges proved well suited to this type of activity, although no more so than unmodified flakes or side scrapers. Similarly, fleshing skins treated with cinders was possible with the retouched, convex and regular proximal end of a hafted biface (figure 18b), as was cutting strips from dried hides (figure 96). Piercing hides, although possible, posed the same problem as encountered when working wood.

Other activities

318A significantly smaller number of experiments concerned hard animal materials, grasses and soil. Five involved bone (scraping, sawing, percussion), three used dry deer antler (scraping and sawing), grasses were cut in three, and four involved digging. Sawing hard materials posed the same problems as those identified for woodworking and were connected to the unsuitable profiles of bifaces. Two small bifaces were hafted onto the end of wooden shaft as armatures. Finally, five additional experiments concerned transporting bifaces; two bifaces were taken in and out of a bark sheath for 10 minutes, and three bifaces were transported in either bark or leather sacs during a 10‑hour hiking trip.

Figure 96 - Schematic illustration of the macro-traces observed on a flint biface used to cut leather thongs from dry hide

Figure 96 - Schematic illustration of the macro-traces observed on a flint biface used to cut leather thongs from dry hide

Photographs and CAD: É. Claud and M. Coutureau

c - Resharpening bifaces

319The goal of the resharperning experiment was to explore whether use-wear traces from previous activities were still identifiable on resharpened bifaces and resharpening flakes.

320Six bifaces produced by S. Maury were first used in three different activities:

  • two were used to cut meat from a cow at the Grands Abattoirs de Bordeaux (Bordeaux Slaughterhouse) for 1h 45 each;

  • two served to plane and scrape bark (using positive and negative rake angles) and then thin a dry ash branch for 1h 10 and 1h 50, respectively;

  • and, finally, two were used to dig (using a percussive motion) sandy, compact and humid soils, each for 30 minutes.

321While each of these uses produced fairly intense wear, it varied between activities, which is fundamental to the potential identification of use-traces after resharpening. In fact, wear produced during butchery was more moderately distributed compared to the more extensive traces produced when working wood (transverse motion) or while digging, which left traces all across the biface.

322The pieces were subsequently washed, photographed, and examined for use-wear traces using a stereomicroscope (10 to 30 × magnification) and a metallographic microscope (100 to 200 × magnification) The presence and intensity of wear was recorded for each of the six bifaces (Bou 47 and 48, BS 39 and 40, T7 and 8). Finally, the used edges were photographed and impressions were made in order to preserve their condition prior to resharpening.

323The bifaces were resharpened by S. Maury using both soft organic (reindeer antler and a holm oak billet) and softstone (sandstone) hammers. The resulting flakes were collected and sorted by resharpening stage (first or second generation) and by resharpened or reworked zone (base, point, right or left edge). The total number of flakes or flake fragments larger than or equal to 5 mm was recorded, as were those detached uniquely from the used area of the biface (table 22). Each flake or fragment was observed unaided and then using a stereomicroscope. Higher magnifications were used to examine use-wear traces or verify potential traces identified under lower magnifications. The edges and surfaces of the bifaces were also examined under high and low magnification.

Table 22 -List of bifaces subjected to resharpening after use and number of pieces still bearing traces of use after resharpening, according to mode of use

Number

Activity

Resharpening flakes

Resharpening flakes from the active zone

Resharpening flakes from the active zone with use-war traces

Resharpened bifaces with use-wear traces

% of flakes with use-wear traces

Bou 47

butchery

45

27

0

0

0

Bou 48

butchery

23

16

0

0

0

BS 39

wood scraping

85

37

1

0

3

BS 40

wood scraping

77

52

2

0

4

T7

soil digging

136

105

8

1

8

T8

soil digging

115

99

14

1

14

C - Characterising use-wear traces on experimental pieces

a - Manufacturing traces

Resharpening

324A low-powered approach to retouched pieces, such as bifaces, requires distinguishing between intentional retouch, notably the final phase of edge regularisation, and scars resulting from use. Based on the general tenets of the low-powered approach (Tringham et al., 1974; Odell, Odell‑Vereecken, 1980), we identified relevant criteria for differentiating the two by comparing the un-used edges of bifaces retouched with different types of hammers (organic: deer and reindeer antler, boxwood, helm oak; mineral: limestone and sandstone) with scarring produced during use.

325First, retouch is never isolated on the edge and rarely discontinuous (i.e. never on a limited portion of the edge), unlike scarring resulting from a majority of longitudinal actions (figure 97a-b). Moreover, shallow retouch creates a regular cutting angle along the edge, which tends to result in a finer edge (designed to be sharp) unlike use‑induced damage that generally increases edge angle either due to more abrupt removals or bending initiations. Removal direction is consistently perpendicular. Retouch is often superimposed, reflecting multiple resharpening episodes designed to maintain the same low edge angle. The retouch negatives also share the same morphology, which is not the case with use-related damage that sometimes occurs in highly variable forms (half‑moon with semi-circular, triangular, …). These removals are never associated with crushing, which is both typical of working hard materials and increases the cutting edge angle. They can be highly invasive and extremely narrow, which is never the case with use-induced damage.

326Certain zones resharpened with soft-hammers exhibited retouch with bending initiations (figure 97b) that tended to increase the cutting edge angle, although in these cases, additional criteria, notably the extension and regularity of the retouch, sets it apart from use-related damage.

Technologically related abrasion and percussion

327The examination of bifaces, unused biface manufacturing flakes and waste produced by different types of hammers (quartzite for preforming, and sandstone, limestone, reindeer and deer antler for shaping and sharpening) and abraders (sandstone, schist) allowed us to better identify technological traces as well as distinctive criteria separating them from use-related damage. The edges and surfaces of bifaces, as well as the platforms, bulbs and flake edges, were subsequently examined using low and high magnification.

328The rare percussion mark took the form of different types of striations. One edge (on the non-retouched surface) of a biface exhibited isolated, wide, very shiny additive striations that were shallow, discontinuous and long (figure 97c). Traces of a hard, domed, marginal and shiny polish were also occasionally present on the edge and oriented perpendicular to it (figure 97d). Two smooth, deep, wide and long parallel striations were observed on the platform of a flake detached during the preform stage using a quartzite pebble (figure 97e). This type of damage is rare, highly localised and isolated, making it impossible to be confused with use-related wear.

329Given previously documented evidence for abrasion on MTA bifaces (Soressi, 2002: 116, 119), the experimenters sometimes used an abrader to prepare the edges for shaping or retouching, with the resulting traces evident on:

  • the platforms of biface manufacturing flakes, occasionally extending onto the proximal portion of the dorsal surface;

  • and more rarely, on the non-retouched areas that “escaped” the final regularisation of the edges (i.e. abraded but unretouch).

Figure 97 - Microscopic view of the retouch removals from sharpening and microscopic traces of percussion on the experimental bifaces and biface manufacturing flakes

Figure 97 - Microscopic view of the retouch removals from sharpening and microscopic traces of percussion on the experimental bifaces and biface manufacturing flakes

a: sharpening retouch; b: sharpening retouch (aligned, invasive removals, bending initiation) (superimposed removals, cone initiation); c: additive streak related to percussion during sharpening retouch; d: spot of polish on an exposed zone related to contact with an organic hammer; e: smooth-bottomed striations related to percussion with a quartzite hammerstone

Photographs: É. Claud

330This crushing created a heavy blunting associated with striations that could occasionally be observed at low magnification (figure 98a-c). These numerous long striations, occurring in parallel groups, were either smooth bottomed, fine and deep (figure 6d) or additive, wide and shallow (figure 6e-f).

331These zones cannot be mistaken for active areas used to work mineral materials given both their short length and the fact that they were overlain by subsequent resharpening removals.

Figure 98 - Photographs of the experimental macro- and micro-traces related to technical abrasion on the residual zones of the edges of bifaces and the platforms of biface manufacturing flakes

Figure 98 - Photographs of the experimental macro- and micro-traces related to technical abrasion on the residual zones of the edges of bifaces and the platforms of biface manufacturing flakes

a: crushing produced by scraping with a sandstone hammer (platform, biface manufacturing flake); b: striations produced by abrasion from scraping with a sandstone hammer (platform, biface manufacturing flake); c: detail of crushing produced by abrasion from scraping with a sandstone hammer (platform, biface manufacturing flake); d: microscopic striations produced by abrasion on schist; e: microscopic striations produced by abrasion with sandstone on the proximal end of a biface manufacturing flake e (platform); f: microscopic striations produced by abrasion with sandstone on the proximal end of a biface manufacturing flake (dorsal surface)

Photographs: É. Claud

b - Use-related wear

Frequent macro-traces

332Of the 100 active zones on the experimental bifaces, 90 exhibited macroscopic use-wear, most often in the form of scarring. Edges lacking use-related damage were used to work soft, fibrous wood, such as poplar, with the bark already removed (sawing for 10 to 15 minutes, scraping for 15 to 40 minutes, grooving for 25 minutes), cut meat already removed from the carcass (30 minutes), and flesh fresh hides using a tangential cutting motion (40 minutes). A low-powered approach to the study of bifaces is therefore informative despite the more resistant retouched edges compared to lower-angled unmodified ones on which scarring has often been described.

Differences compared to unmodified edges

333Macroscopic damage on the edges of bifaces was compared to unmodified flakes with equivalent angles described in the literature (see Part I, chapters 2.4 and 2.5). The development of use-wear on bifaces follows the same “rules” as unmodified flakes: the same parameters (hardness of the contact material, movement, …) induce differences in the characteristics of use-related wear and a combination of criteria (e.g. morphology, distribution, initiation) needs to be considered when interpreting them. Notable differences have, however, been observed in cutting meat and cutaneous tissue as well as all types of woodworking. The damage resulting from working hard animal materials and minerals, soil, grasses, as well as scraping hides, share more features with those found on unmodified edges.

334The main differences concern:

  • the size of scars, which tend to be smaller;

  • their number, disposition and distribution: scars are generally less numerous, often discontinuous and isolated;

  • their morphology: certain morphologies are not found on bifaces, such as half-moons, while others are more frequent (triangular);

  • and their elongation: longitudinal actions tend to produce less elongated scars compared to unmodified edges of the same angle, while transverse actions tend to result in narrower and therefore more elongated scars.

335The position, number of generations, initiation and orientation are less affected by the presence of retouch. This is also the case for edge-rounding and blunting.

Description of macro-traces according to action and hardness of the contact material

336The different types of uses and associated macroscopic edge-damage are described below, accompanied by a synthetic example (figure 99).

- Non-percussive uses

337Cutting soft animal materials without touching the bone
This includes cutting fresh and dried hides (fleshing and cutting) as well as meat already removed from the carcass. These activities are generally more difficult to identify archaeologically based on macroscopic use-wear due to the fact that they are not produced systematically: one out of two bifaces used to flesh fresh hides using a tangential motion for 40 minutes exhibited no damage, as did an example that served to cut meat already removed from the bone for 30 minutes. For these activities, microscopic use-wear traces need to be taken into account. With that said, other bifaces used to cut dried hides for an hour or flesh fresh hides using a tangential motion for 30 to 40 minutes exhibited wear consistent with longitudinal actions involving soft materials, which set them apart from damage produced by other materials: for fleshing, this comprised minute, triangular scarring with stepped terminations located uniquely on the unretouched surface, oblique to the edge, and associated with a weakly‑developed bifacial edge-rounding; for cutting dry hides, very limited (less than 5), small semi-circular scars oblique to the edge with feather terminations and marked bending initiations occurred on the same surface and were also associated with a weakly developed bifacial edge-rounding (figure 96). In archaeological contexts, reliably interpreting this macroscopic evidence as resulting from use can be difficult.

338Cutting soft animal materials while touching the bone (soft to medium-hard materials)
This category groups meat removal, disarticulation and skinning, which produce scars reflecting accidental contact with the bone. These characteristic scars develop relative to the force exerted and the number of contacts with the bone and therefore vary according to the caution employed and duration of use (ranging from 5 to 20 with variable sizes). Even pieces used for the shortest amount of time (30 minutes) and with precision (i.e. avoiding the bone), nevertheless exhibited interpretable scars, which were bifacial, discontinuous, isolated or aligned, sometimes organised in one or two generations, triangular or trapezoidal, often with marked step terminations, cone or weakly‑developed bending initiations, low to average elongation and an oblique orientation (figure 95).

Figure 99 - Overview of the characteristic macro-traces produced during experimental use of bifaces, according to the hardness and nature of the materials worked and the mode of action

Figure 99 - Overview of the characteristic macro-traces produced during experimental use of bifaces, according to the hardness and nature of the materials worked and the mode of action

Photographs and CAD: É. Claud

339A final important feature is that, even if the entire length of a lateral edge is used, macroscopic use-wear traces often concentrate in the distal third, with the tip exhibiting the heaviest scarring (in terms of numbers and generations). A weakly developed edge-rounding can also be observed in cases of skinning.

340No significant differences were noted between macroscopic damage produced during the different butchery stages involving cutting (skinning, meat removal, disarticulation).

341Cutting soft plant materials (grasses)
The distinguishing feature of this activity is the brightness of the polish that develops all along the cutting edge, sometimes covering several millimetres of the surface. The edge no longer feels sharp to the touch. No scarring was observed when cutting fresh grasses, while several scars were produced when dry grass was cut. These very small scars were bifacial, discontinuous, isolated, semi-circular, and short with stepped terminations.

342Scraping soft and abrasive organic materials (hides)
Use-wear traces on bifaces used to scrape hides with different degrees of freshness are similar to those documented on side scrapers (Lemorini, 2000; Claud, 2008).
Scraping
fresh hides results in weakly rounded edges and a macroscopic brightness, as well as limited scarring on the trailing edge, which is short, small, perpendicular to the edge and associated with feather terminations. All of these macroscopic traces are not always present (e.g. a biface used for 20 minutes exhibited no damage), but when present, they allow a function to be determined. It is, however, easier to base interpretations on the more developed and diagnostic micro-traces. The scraping of dried hides with an abrasive is easily identifiable by a well‑developed edge rounding, including on pieces used for relatively short periods (20 minutes), and the frequent presence of associated scarring, which is short, small, semi-circular, perpendicular to the edge and associated with feather terminations and cone initiations.

343Piercing soft to medium hard materials (dried hides)
Scarring was concentrated on the tip, small in size, and found even on pieces used for a limited amount of time (10 to 20 minutes). This characteristic damage developed on both surfaces and both edges and occurred as discontinuous, aligned or isolated scars of variable length, semi-circular or quadrangular in morphology, perpendicular to the edge, with feathered or lightly stepped terminations and bending initiations. Scarring from piercing can be distinguished from that observed with woodworking due to its smaller size and less well-developed bending initiations (for woodworking, the inflection point and abrupt portion of the initiation are easily visible and give the edge a steep profile).

344Digging loose mineral materials (soils)
This activity is identifiable by a combination of damage that develops rapidly (10 minutes): shiny polish and heavy rounding of the point that diminishes towards the base, short, bifacial, discontinuous scarring, aligned or superimposed (one or two generations), variable in size (large and perpendicular at the point and average to small in size on the lateral edges), semi-circular, quadrangular or trapezoidal in morphology, with feathered or hinged terminations and cone initiations.

345Working medium-hard materials with a longitudinal motion (wood)
While sawing and grooving wood sometimes left no macroscopic damage, scarring exhibited by edges used to work fresh wood (two pieces used for 15 and 20 minutes to work, respectively, oak and ash) or dry wood (scarring formed after only 10 minutes) are easily distinguishable from damage produced by other activities. This bifacial scarring was continuous, aligned or isolated, never superimposed and relatively limited (between 2 and 10 scars). The small to average sized scars were often semi-circular, although they could at times be trapezoidal or slightly elongated. Scarring was organised obliquely to the edge with marked bending initiations and feathered terminations. Scarring occurred in the most fragile areas of the tool, with other areas sometimes exhibiting a macroscopic sheen.

346Working medium-hard materials with transverse motions (wood)
Bifaces used to scrape fresh, relatively soft and fibrous wood (e.g. poplar) with the bark already removed for between 15 and 40 minutes exhibited no diagnostic edge damage, meaning that this type of use can only be identified microscopically. Edges used to remove the bark from and then scrape fresh wood all bore characteristic scarring, even those used for a short duration (15 minutes). Scarring occurred primarily on a single surface (i.e. the trailing edge) and could be continuous or discontinuous, isolated or aligned (but never superimposed, apart from in the case of heat treated wood, which can be considered a hard organic material, see above), limited in number (between 2 and 10), semi-circular or quadrangular, rarely trapezoidal, short to average in length, perpendicular to the edge with feathered or, more rarely, stepped terminations and often well-developed bending initiations that remove a large portion of the cutting edge, giving it an abrupt cross-section (figure 93). If the trailing surface was retouched, scarring was small or sometimes very small and more difficult to identify than when it occurred on the non-retouched surface, where scars are generally larger in size.

347Piercing medium-hard materials (wood)
Scarring was found on the tip and was consistently present, even on pieces used for a limited time (10 minutes). This damage is comparable to what was observed for piercing dried hides. Damage produced when piercing wood, however, is distinguishable by the large size of scars (small to average for fresh wood and average to large for dry wood) and significantly more marked bending initiations that produce an abrupt cutting edge.

348Working hard organic materials
These activities include scraping the periosteum and transforming hard animal (scraping, sawing) and very hard plant materials (burnt wood). Scarring was consistently present, even on edges used for brief periods (10 minutes) and had a characteristic morphology that is fairly similar to the type of damage observed on unmodified edges: numerous, superimposed scars with semi-circular or trapezoidal morphologies that are short to long with often stepped terminations for transverse motions and hinged or transverse terminations in the case of smaller scars (small to average in size) resulting from longitudinal motions. Crushing can sometimes be observed on the edge, although unlike what was observed with hard mineral materials (e.g. shell, stone), the edge remained sharp.

- Use-wear resulting from percussive motions

349Percussion of medium-hard materials (wood)
All bifaces used to chop medium-hard materials exhibited characteristic scars that, although variable, were generally large to very large, limited in number (less than 10), bifacial, discontinuous, most frequently isolated, rarely superimposed (one or two generations), perpendicular or oblique to the edge, average to long, semi-circular or trapezoidal with stepped terminations and well-developed bending initiations (figure 94). Scar size, well-developed bending initiations, and the absence of superimposed scars set this damage apart from that produced while working hard materials, such as bone, or softer materials, including hides and meat. The position, size and orientation of the scars also distinguish them from damage typical of other woodworking activities (scraping, sawing).

350Bifaces as intermediate pieces for splitting medium-hard materials (wood)
While all of the bifaces used in this activity exhibited clear evidence for percussion on their non-active zones (i.e. the base), demonstrating their use as intermediary pieces, the active zones did not always carry scarring. Simple bending fractures can occur and extremely limited scarring (less than 5 scars) was present on the tip. This scarring was bifacial, discontinuous, isolated or aligned, moderately elongated, average to large in size, semi-circular, quadrangular or trapezoidal with feathered or slightly stepped terminations and clearly visible bending initiations.

351Percussion of soft to medium-hard to hard organic material (carcasses)
Removing meat, disarticulating carcasses and fracturing the sheep sternum by percussion all quickly produced intense scarring characteristic of violent contact with hard organic materials such as bone or cartilage. The abundant scars were large to very large in size, superimposed (one to four generations), bifacial, semi-circular, trapezoidal or triangular with marked stepped terminations, and cone and or concave bending initiations. This damage renders the edge highly irregular, and differs from scarring produced by the percussion of medium-hard organic materials such as wood (see above).

352The two bifaces used as projectile tips and thrown into a carcass bore similar traces, except for their discontinuous organisation on the two edges. These traces occurred as numerous large, superimposed bifacial scars with semi-circular and trapezoidal morphologies that gave the edge a ‘shredded’ aspect due to the lack of continuous contact regularising the profile. These features are characteristic of a single, violent contact with a hard organic material.

Overview of macroscopic use-wear on bifaces

353To conclude, recurrent features of macroscopic edge-damage were observed for the specific uses described above, as were differences between activities, confirming the diagnostic potential of edge damage on bifaces for determining use modes. These types of damage allow both the action to be identified and to distinguish uses on soft animal materials, soft and medium-hard plant materials, hard organic materials, as well as loose (i.e. soils) and hard mineral materials (see above). However, determining the exact activity is more difficult in the absence of micro‑polishes, as each of the above categories includes multiple materials, for example, meat and hides (soft animal materials) as well as bone, cervid antler, hardwoods (hard organic materials). Moreover, macroscopic use-wear alone may be insufficient for documenting certain activities tested in our experimental program. This would include cutting soft animal materials without contact with the bone (fleshing skins with tangential motions and cutting meat already removed from the carcass), scraping fresh hides for short durations, and shaping relatively soft wood, such as poplar, using both longitudinal and transverse motions. These activities do not always produce macroscopic damage, meaning that use-related polishes are necessary to establish function. While polishes were systematically present on experimental pieces, unfortunately they were not always preserved on archaeological pieces.

c - Differences with other types of wear

354Our results confirm that macroscopic use-wear traces can be distinguished from both natural alterations and damage linked to prehension and transport (for more detail see Claud 2008). This distinction is fundamental for use-wear analyses designed to determine tool function.

355The same diagnostic features identified in experimental reference collections of unretouched tools can also be applied to bifaces (Tringham et al., 1974; Odell, Odell-Vereecken, 1980; Rots, 2004).

356No macroscopic traces linked to prehension or hafting were observed on the experimental bifaces, with only very occasional microscopic use-wear traces evident on pieces used while wrapped or hafted. Wear resulting from these two configurations was also much less frequent and considerably less well developed compared to use-induced wear. Hafting traces are, however, are also relatively rare and more difficult to identify, which is probably connected to the use of adhesives (Rots, 2004). Finally, prehension traces differ from use‑related wear, as they are limited to contact with fingers or the limits of the haft rather than being regularly distributed along the edge. The absence of macroscopic hafting or prehension evidence and the low intensity of the observed micro-wear, combined with that fact that Middle Palaeolithic assemblages are often patinated, in our opinion, renders the identification of potential hafts difficult if not impossible for bifaces from this period.

357Traces of transport are consistent with previous descriptions, particularly their random distribution across the piece (Odell, Odell‑Vereecken, 1980; Rots, 2002a). Longer periods of transport only altered the extent and organisation of the traces, which first affect the higher than lower areas of the tool’s surface (Rots, 2002a). These traces are therefore easily distinguishable from those linked to hafting. Transporting bifaces in a sac with other tools equally leaves diagnostic use-wear traces (scarring, bright spots, striations with variable orientations) that are almost certainly identifiable in archaeological contexts.

d - Residual use-wear traces on resharpened bifaces

358In the case of bifaces used for butchery, neither the tools themselves nor resharpening flakes exhibited macro- or microscopic traces of use (table 22). Unprepared removals are small (often less than 1 cm) and often broken; the platform is very frequently missing, which precludes observing use-wear traces. Prepared removals, on the other hand, are generally larger and preserve the platform, which often exhibits evidence for crushing linked to the transverse motion of an abrader or hammer. This preparation removes the cutting edge, and with it any traces of use. Finally, no or very few striking platforms preserve a portion of the bifacial edge.

359Bifaces resharpened after having worked dry wood exhibited no residual traces of use. Flakes are often broken and only rarely preserve a portion of the biface’s edge. Three flakes had clear use-wear traces, which were easily identifiable under high-magnification. Detached during the first resharpening episode, these flakes had both prepared (2) and unprepared (1) platforms. Use-wear traces on one example were preserved on a ridge not directly adjacent to the cutting edge and therefore unaffected by edge preparation during resharpening, while on another flake, traces were observed on an unprepared platform preserving a portion of the original bifacial edge. Finally, use-wear traces were present on a lightly prepared platform.

360Traces of use were therefore preserved on resharpening flakes detached from bifaces used to scrape and plane dry wood. While sufficiently well preserved to be interpreted on experimental examples, identifying use-wear traces on resharpening flakes from archaeological contexts can be limited by their small size (less than 3 cm) and the fact that only three flakes exhibited use-related wear.

361Both bifaces used to dig exhibited use-wear traces away from the edge, in zones unaffected by the resharpening episodes. Although this wear was less well developed compared to the original traces on the edges, its morphology and distribution are sufficiently diagnostic to identify the bifaces as having previously worked a very supple abrasive material. Following the two experiments, 22 flakes bore traces of previous use: 9 on the platform and dorsal surface and 13 uniquely on the dorsal surface. Fifteen flakes were detached during the first resharpening phase and 7 from the second, the latter exhibited use-wear traces in the distal portion of the dorsal surface rather than on the platform and proximal area. Two flakes with unprepared platforms can be described as lip flakes, meaning that their thick platforms have marked lips. These two features (complete and lip flakes) seem to be connected to a mode of use that significantly alters the edges (scarring, edge-rounding, polishes, high edge angles). In fact, unlike bifaces used in woodworking, the experimenter detached complete flakes without systematic platform preparation during the first phase of resharpening. All of the flakes were small (less the 2 cm) and, despite a clear difference in the results obtained for bifaces used in butchery and scraping wood, the proportion of resharpening flakes exhibiting traces of use remained low (n=22, approximately 10 %).

362In sum, both resharpened bifaces and associated resharpening flakes exhibited use-wear traces. The likelihood of observing these traces varies, however, according to their nature, particularly their extent: the more extensive the wear, the more frequently residual use-related evidence can be observed. With that said, the proportion of flakes with use-related wear remains low. For example, no residual traces of use were observed on bifaces or flakes used in butchery.

363This differential preservation of use-wear traces therefore poses a problem for interpretations based on non-quantifiable observations. It should, however, be born in mind that some parts of the edge are not resharpened and thus may provide evidence of previous activities regardless the extent of the original use-wear. The identification of residual use-wear evidence is, in the end, limited by the small proportion of flakes that preserve traces of use and, especially, their small size. These artefacts are, in fact, more susceptible to breakage and transport, notably by surface water run-off. Moreover, when preserved and recovered during excavations, their study often requires the sieve residue to have been sorted.

11 - Synthesis of the lithic tools reference collections

É. Claud, C.Thiébaut, A. Coudenneau, M. Deschamps, V. Mourre

364The reference collection of lithic tools constituted in the context of the Research Program totals 510 active areas, used for butchery (164 areas), acquiring and working of wood (149 areas), shooting on carcasses (107 points), hide working (49 areas), hard animal material working (bone, antler, horn, shellfish 32 active areas), and to a lesser extent grass harvesting (4 areas) and soil digging (5 areas). Flint and quartzite are the most used raw materials, alongside some elements in ophite, schist and volcanic rock.

365Among the items used, there are unretouched and retouched points (198 areas), bifaces (108 active areas and 5 transported bifaces), unretouched flakes (most of them in quartzite: 69 areas), notched pieces (67 areas) and flake cleavers (52 tools).

366Practicing different activities, according to various modes of action and, to a lesser extent, of prehension, made it possible to confront the constraints and advantages of the different tested tools types. A general remark can be formulated for quartzite, whatever the activity carried out: the cutting edges get dirty quickly and require regular cleaning during work in order to preserve their efficiency.

367With few exceptions, the tools have been relatively well adapted to the intended objective. Some tools, however, seemed to us to have more advantages than others. We are summarizing below the constraints and advantages encountered during the main activities.

368In the context of woodworking, unretouched flakes proved less suitable for sawing than denticulates with micro- and medium denticulation, which made it possible to achieve the stated objective: sawing a trunk or a branch. Long cutting edges, generally rectilinear in plan and especially in profile and thin blanks facilitate the working. Unsurprisingly, acquiring by percussion gesture proved much faster than sawing, especially in the case of hafted tools. The hafted bifaces allowed to achieve the stated objectives (felling of a tree) but the number of blows required was greater than with hafted flake cleavers. The different tests of lateral hafting of the bifaces did not allow us to find a stable hafting system over the duration of the experiment, while some flake cleavers remained in their handle during all their functioning time. For the latter, hafting into straight handles, in the manner of axes, allowed faster felling than the use of curved handles, of the adz type, for comparable varieties and trunk diameters. The morphology of the active area seemed to condition the stability of the flake cleaver in the handle: a rectilinear or slightly convex and symmetrical delineation is preferable to a very convex, asymmetrical or two sides delineation cutting edge. Despite the strong scarring of the cutting edges for each experiment, the flake cleavers remain relatively efficient. The large size of the handles, adapted to that of the flake cleavers and not allowing to carry out precise gestures, was an obstacle to the shaping of wooden elements although handles of smaller dimensions, usable with one hand, had been produced for this activity. Regarding the scraping of wood, rectilinear or concave cutting edges in delineation, with biplane or plano-concave section, seem better adapted. On the contrary, using quartzite notches, or flint notches with a convexity (convexo‑concave or convex-plane section) on the flat face, appeared unsuitable for this functioning mode, because the cutting edge failed to cut into the material.

369During our butchery activities, the presence of a pointed part (point or denticle) has proved very useful for penetrating the flesh, for example to disarticulate and to incise the hide at various stages of skinning by applying concentrated pressure on a reduced portion of the cutting edge. For removing the hide, the use of pointed or angular active parts was of less interest. For defleshing, a long cutting edge, unretouched or retouched, was well suited although the removal of the meat of the hind limbs can be achieved without problem with pseudo‑Levallois points of small dimensions. With regard to disarticulation, although the presence of denticles may facilitate the cutting of tendons, the relatively obtuse angle of some denticulate cutting edges and the thickness of their blanks has sometimes been an obstacle to penetrating between the articular surfaces. Unlike flint blanks, quartzite blanks required frequent cleaning. Their cutting edge also seems to become blunted and lose sharpness faster than flint blanks. Flake cleavers, too large too massive and presenting a difficult prehension and a limited maneuverability, appeared inappropriate for butchery cutting. On the other hand, their use to fracture the rib cage, hafted in percussion, made it possible to reach the objective very quickly. Their use in percussion made it possible to fracture, in just a few blows, bovine femurs to extract the marrow. Bifaces with convergent edges have proved very effective for all stages of butchery; the point, the regular retouching of the edges and the acute to moderately open angle give them good penetration capacity, and a significant sharp edge and resistance. Those with distal transverse edge have also proved useful for butchery cutting, with one of the lateral edges and the angle formed with the distal edge, but also for butchery in percussion, with the transverse distal edge, for the purpose of defleshing or disarticulating. On the other hand, an abrupt decrease in sharpness was noted for bifaces used in percussion to cut meat or tendons in the context of these activities. We did not test the use of hafted bifaces for rib cage fracturing so we cannot compare their efficiency with flake cleavers for this functioning mode, but two converging-edges bifaces used with bare hand in percussion allowed to fracture the rib cage of a red deer.

370Except for piercing and softening straps, hide working requires regular and convex cutting edges, whether unretouched or retouched. They allow applying sufficient pressure on the hide to deflesh it, shave it or thin it, without the risk of piercing it. The use of flint notches as stationary tools adapted to the width of the straps for softening was a success. As with butchery cutting, the large dimensions of the flake cleavers, even held with bare hands, made these tools difficult to handle for hide working, compared to simple unretouched quartzite flakes.

371The use of handles was a definite advantage in the case of bifaces and flake cleavers used in percussion for acquiring wood and fracturing of the rib cage. The hafting of two small pseudo‑Levallois points (used for butchering the bison) in a simple partially split handle was also appreciable. The handle, whose manufacture was very quick (split handle, without bindings or adhesives) and which was well adapted to the size of the points, did not interfere with the penetration of the tool in the flesh. Apart from several particular situations (percussion gestures, small size items) and, of course, from the hunting points, the use of handles did not bring a neat gain in efficiency. Their manufacture has sometimes been relatively time-consuming, especially in the case of large tools. In particular, this was the case for the bifaces used in butchery cutting, for which the handle had to be both large enough to hold the tool and sufficiently fine, therefore adjusted to the tool, not to affect its ability to penetrate the flesh.

372The use-wear study of experimental tools, at low and more rarely at high magnification (quartzite flakes reference collection), confirmed the informative potential of the utilization traces born by the various tool types.

373The interest of taking into account macro-traces was underlined for each of the reference collections. Observations at low magnification allow the identification of fractures, scars and rounding that are often diagnostic of the motion carried out and of the hardness of the material worked. Some activities, such as butchery, produce very characteristic scars, also identified on all types of studied tools and raw materials: triangular morphology, oblique orientation, bending initiation and step termination, distributed discontinuously and located on both faces. They are often associated with other scar morphologies (mostly semicircular and trapezoidal). The tools used for hide working, for their part, carry little or no scars, always small in size, and a more or less marked rounding, depending on the duration of use, the state of the hide or the presence of abrasive particles. Woodworking, even in percussion, produces relatively few, semicircular, trapezoidal or quadrangular scars, with clearly bending initiation, which increases the cutting angle of the edge used. Macro‑rounding of the cutting edge was observed on quartzite tools that scraped wood, as well as on denticulates used for sawing dry wood. The working of hard organic materials such as bone, antler or bone cleaning by scraping produces on the contrary numerous and superimposed scars. Some convergence has been observed between the marks produced by the working of a hard material such as bone and hardwood, materials that may have similar hardness, depending on the species / variety, the age of the animal (immature / adult) and the state of the support (fresh, dry, heated). This convergence has been observed particularly in the case of unretouched quartzite flakes used for scraping (wood, bone) and of quartzite and ophite flake cleavers used in percussion for acquiring wood and fracturing the rib cage.

374Nevertheless, within all the reference collections, some tools, sometimes representing up to a third of the tools making up the reference collection (for notched pieces) have no macro-traces or traces that are too tenuous to be considered as characteristic and interpretable in archaeological context. These tools were used for cutting meat or skinning, without contact with bone or cartilage, to scrape fresh and soft wood, and to a lesser extent for hide defleshing. Low durations of use and open cutting angles limit the formation of traces and therefore their ability to be recognized. To identify these functioning modes, a microscopic approach is essential. However, it is possible only in the case where the preservation of microscopic surfaces and cutting edges is correct, which is unfortunately not always the case in the studied assemblages (see Part II, chapter 2). Thus, these activities are probably underestimated when restoring activities carried out by Neanderthal groups.

375The creation of specific reference collections for a raw material, quartzite, and for several types of tools made it possible to illustrate the peculiarities of each in terms of their wear behavior, particularly the development of macro-traces.

376Thus, on the bifaces and the points, use scars appeared less numerous and smaller in size than on the unretouched flakes. They are also less elongated in the case of longitudinal actions. Denticulates used for sawing and butchery cutting have the particularity of developing scars and rounding on the denticles, the most prominent parts, which therefore wear out more significantly. In the case of an active zone comprising a pointed part (unretouched and retouched points, bifaces, …), the use-wear tends to concentrate at the level of the point and of the cutting edge adjacent to it, their intensity decreasing when moving away from it. The distribution of use-wear related to scraping actions seemed less dependent on the morphology of the cutting edge, which varies little according to the tool types.

377Let us note that the quartzite blanks from the Nive and Garonne Rivers have developed macrowear and micro-traces characteristic of their utilization mode. The micro‑traces on flint are not transposable to quartzite, hence the interest of a specific reference collection. Indeed, the matrix generally does not show any polish, the traces have to be looked for on each crystal, hence the need for an examination with a 50 lens (500 × magnification). On this material, the legibility of the traces depends on the size of the crystals (fine-grained quartzite versus medium-grained quartzite): the macro-traces are easier to read on fine-grained quartzite than on medium-grained quartzite, and, conversely, the scarcity of the matrix on medium-grained quartzite makes the reading of micro-traces easier.

378The reference collection of points made it possible to highlight the formation, on the latter, of a new type of fracture that can be diagnostic of an impact, the apical scarring. It also allowed putting in perspective the diagnostic aspect of bending fractures with tongue and step termination since these can also result from other activities than hunting or from natural phenomena (Pargeter, 2011). The presence of fractures on the points used as projectile insets for hunting is far from systematic, and among the fractures observed, only a small number of them could be considered as diagnostic. The number of pieces showing potentially diagnostic traces of an impact within the same series (very anecdotal or not), the overlapping of retouches by fractures in the case of retouched points, as well as the taphonomic and economical context of the occupations are crucial data to be able to offer a reasonable functional interpretation of the artifacts bearing traces compatible with a use as hunting points.

379Concerning the question of identifying of a particular stage within the butchery (skinning, defleshing, disarticulation, tendon removal), it turns out that the latter have produced macro- traces that are not significantly different from each other. In fact, it is mainly the accidental contact with bones or cartilage that creates the scars typical of butchery, and these are likely to form at all stages of the butchery. A macro‑rounding is sometimes observed on the tools used for skinning, but its intensity, always low, limits its interest for the interpretations of the archaeological material because it could be mistaken with a slight natural rounding. Only micro-traces (see Part II, chapter 2.6 for quartzite and Claud, Thiébaut in Thiébaut et al., 2011 for flint: 230-258) might make it possible to identify the different stages of butchery in the event that the tool was used for a particular stage.

Haut de page

Notes

1 This resin is perfectly adapted to producing highly-accurate moulds of macro-traces. However, while our experience showed micro-traces to be equally well replicated, the micro-topography of the moulds proved fragile, requiring considerable caution when manipulating (removing from the mould or during observations) and storing (arranged with minimal contact) in order to preserve polishes. For example, rubbing a mould of an edge for several while wearing latex gloves produced a domed, coalescence polish.

2 Certain authors use the terms ‘activity’ to designate the mode of use (van Gijn, 1989b) or action (Odell, Odell- Vereecken, 1980; Gutierrez Saez, 1993), and the term ‘task’ for what we have described as an ‘activity’.

3 We prefer the terms continuous contact and percussive motions to fixed percussion (grinding) or percussion (pounding) as discussed by Leroi-Gourhan (1971).

Haut de page

Table des illustrations

Titre Figure 21 - Example of the data collection form for stone tools used during the experiments (excluding spear-points)
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-1.jpg
Fichier image/jpeg, 437k
Titre Figure 22 - Example of the data collection form for objects used as thrusting or projectile point
Crédits CAD: A. Coudenneau
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-2.jpg
Fichier image/jpeg, 164k
Titre Figure 23 - Terminology employed: gestures
Crédits CAD: C. Thiébaut
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-3.jpg
Fichier image/jpeg, 33k
Titre Figure 24 - Terminology employed: types of contact
Crédits CAD : C. Thiébaut
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-4.jpg
Fichier image/jpeg, 32k
Titre Figure 25 - Terminology employed: movements
Crédits CAD: C. Thiébaut
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-5.jpg
Fichier image/jpeg, 29k
Titre Figure 26 - Position of the face, the trailing face and the associated angles
Crédits Modified from Pasquini, 2003-2004
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-6.jpg
Fichier image/jpeg, 39k
Titre Figure 27 - Value of the leading angle according to the mode of action of an edge applied in continuous, unidirectional, transverse action (positive and negative rake angle)
Crédits Modified from Gassin, Garidel, 1993
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-7.jpg
Fichier image/jpeg, 40k
Titre Figure 28 - Existing combinations of the different hafting attributes
Légende Fem.: female; jux.: juxtaposed
Crédits Modified from Stordeur, 1987
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-8.jpg
Fichier image/jpeg, 117k
Titre Table 11 - Terminology adopted for the description of scars
Crédits Illustration modified from Gonzáles Urquijo, Ibáñez Estévez 1994; Coudenneau 2004; Prost 1989; Lawrence 1979
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-9.jpg
Fichier image/jpeg, 168k
Titre Table 12 - Terminology adopted for the description of fractures
Crédits Illustration of fracture removal A. Coudenneau, all others after O’Farrell 1996, modified, and Fischer et al. 1984
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-10.jpg
Fichier image/jpeg, 106k
Titre Figure 29 - Types of micro-rounding observable on quartzite
Légende a: scraping fresh lamb hide (hide defleshing); b: scraping of a fresh cow bone (removal of periosteum)
Crédits Photographs: É. Claud
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-11.jpg
Fichier image/jpeg, 95k
Titre Table 14 - Terminology adopted for the description of micro-polishes
Crédits Photographs: É. Claud; illustration modified after Plisson 1985
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-12.jpg
Fichier image/jpeg, 212k
Titre Table 15 - Terminology for the description of striations
Crédits Photographs: É. Claud
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-13.jpg
Fichier image/jpeg, 116k
Titre Table 16 - Supplementary terminologies for the micro-traces observed on quartzite
Crédits Photographs: É. Claud
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-14.jpg
Fichier image/jpeg, 202k
Titre Figure 30 - Overview of the macro-traces observed on flint pieces
Crédits Photographs: É. Claud and A. Coudenneau; CAD: É. Claud and M. Coutureau
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-15.jpg
Fichier image/jpeg, 476k
Titre Figure 31 - Overview of the micro-traces observed on flint pieces
Crédits Photographs: É. Claud and A. Coudenneau; CAD: É. Claud and M. Coutureau
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-16.jpg
Fichier image/jpeg, 459k
Titre Figure 33 - Macro-traces observed on the tools that were used on soft materials such as fresh hide (skinning, hide defleshing), meat (defleshing) and soft to medium-hard materials (defleshing, removal of tendons)
Légende The white lines indicate zones of rounding
Crédits Photographs: É. Claud; CAD: C. Thiébaut and M. Coutureau
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-17.jpg
Fichier image/jpeg, 379k
Titre Figure 34 - Macro-traces resulting from scraping, cutting and perforating dry hide
Légende a, c and d: rounding; b, e, f: semi-circular scars with a flexion initiation. The white lines indicate zones of rounding
Crédits Photographs: É. Claud; CAD: C. Thiébaut and M. Coutureau
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-18.jpg
Fichier image/jpeg, 307k
Titre Figure 35 - Macro-traces related to scraping, sawing and perforation of wood
Légende The white lines indicate zones of rounding
Crédits Photographs: É. Claud; CAD: C. Thiébaut and M. Coutureau
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-19.jpg
Fichier image/jpeg, 322k
Titre Figure 36 - Macro-traces observed on edges used to clean or work hard organic materials (bone, horn)
Légende The white lines indicate zones of rounding / blunting; d, e, and i show them in detail
Crédits Photographs: É. Claud; CAD: C. Thiébaut and M. Coutureau)
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-20.jpg
Fichier image/jpeg, 331k
Titre Figure 37 - Schematic illustration of the macro-traces observed on an unmodified flake in quartzite used to scrape wood
Crédits Photograph and CAD: C. Thiébaut
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-21.jpg
Fichier image/jpeg, 153k
Titre Figure 38 - Schematic illustration of the macro-traces observed on an unmodified flake in quartzite used to skin a carcass
Crédits Photograph and CAD: C. Thiébaut
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-22.jpg
Fichier image/jpeg, 151k
Titre Figure 39 - Schematic illustration of the macro-traces observed on an unmodified flake in quartzite used to deflesh a fresh hide
Crédits Photograph and CAD: C. Thiébaut
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-23.jpg
Fichier image/jpeg, 152k
Titre Figure 40 - Schematic illustration of the macro-traces observed on an unmodified flake in quartzite used to disarticulate a carcass
Crédits Photograph and CAD: C. Thiébaut
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-24.jpg
Fichier image/jpeg, 125k
Titre Figure 41 - Schematic illustration of the macro-traces observed on an unmodified flake in quartzite used to disarticulate a carcass
Crédits Photograph and CAD: C. Thiébaut
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-25.jpg
Fichier image/jpeg, 138k
Titre Figure 42 - Schematic illustration of the macro-traces observed on an unmodified flake in quartzite used to scrape a dry hide
Crédits Photograph and CAD: C. Thiébaut
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-26.jpg
Fichier image/jpeg, 105k
Titre Figure 43 - Macro-traces observed on edges used to cut grasses and dig soil
Légende The white lines indicate zones of rounding/ blunting
Crédits Photographs: É. Claud; CAD: C. Thiébaut and M. Coutureau
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-27.jpg
Fichier image/jpeg, 224k
Titre Figure 44 - Edges before use, viewed with a microscope (200 x magnification)
Légende a: flint; b: fine-grained quartzite from the Neste River
Crédits Photographs: É. Claud
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-28.jpg
Fichier image/jpeg, 93k
Titre Figure 45 - Micro-traces resulting from cutting activities related to butchery
Légende a: continuous breakage and striations; b, g: scarring; c: rough, fluid or soft polish on a zone of erosion; d-f: continuous erosion; h: fine and superficial oblique striations on a crystal
Crédits Photographs: É. Claud
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-29.jpg
Fichier image/jpeg, 354k
Titre Figure 46 - Micro-traces observed on tools used to work fresh or dry hides
Légende a: continuous erosion; b, f and h: rounding and polish with a rough, fluid or soft coalescence on the matrix; c: continuous breakage: the rounding and the polish (with craters) occur mostly in a line devoid of crystals; d: crystal in the process of dissipation by continuous erosion; e, g: start of continuous erosion and short, isolated striations in g
Crédits Photographs: É. Claud
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-30.jpg
Fichier image/jpeg, 312k
Titre Figure 47 - Micro-traces related to woodworking
Légende a: micro-polish; b: fine, regular erosion, rounding and striations on the matrix, c-d: striations on crystals, isolated and continuous erosion in c and isolated erosion with micro-polish on the edge in d; e: continuous erosion, striations and micro-polish; f, g and i: erosion of varying intensity and striations on the crystals; h: large and deep pitting (extremity of the active zone
Crédits Photographs: É. Claud
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-31.jpg
Fichier image/jpeg, 341k
Titre Figure 48 - Micro-traces related to cleaning or working hard animal materials
Légende a and f: micro-polish; b: beginning of continuous erosion; c and h: scarring; d, e, and g: striations on crystals
Crédits Photographs: É. Claud
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-32.jpg
Fichier image/jpeg, 337k
Titre Figure 49 - Micro-traces related to other activities: cutting grasses (a to d) and digging soil (e and f)
Légende a and f: scarring; b: polish; c: scarring overlain with polish; d: striations and beginning of erosion of a crystal; e: intensive and large erosion of a crystal
Crédits Photographs: É. Claud
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-33.jpg
Fichier image/jpeg, 219k
Titre Figure 50 - Overview of the primary macro-traces from use observed on unmodified quartzite flakes, according to the hardness of the material worked and the mode of action
Crédits Photographs: É. Claud
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-34.jpg
Fichier image/jpeg, 201k
Titre Figure 51 - Overview of the primary micro-traces observed on unmodified quartzite flakes, according to the nature of the material worked and the mode of action
Crédits Photographs: É. Claud
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-35.jpg
Fichier image/jpeg, 184k
Titre Figure 52 - Schematic illustration of the macro-traces observed on a flint denticulate used to saw green poplar
Crédits Photographs and CAD: C. Thiébaut
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-36.jpg
Fichier image/jpeg, 122k
Titre Figure 53 - Schematic illustration of the macro-traces observed on a flint denticulate used in the butchery of a cervid
Crédits Photographs and CAD: C. Thiébaut
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-37.jpg
Fichier image/jpeg, 153k
Titre Figure 54 - Schematic illustration of the macro-traces observed on a flint denticulate used to saw dry poplar
Crédits Photographs and CAD: C. Thiébaut
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-38.jpg
Fichier image/jpeg, 150k
Titre Figure 55 - Schematic illustration of the macro-traces observed on a flint denticulate used to saw oak
Crédits Photographs and CAD: C. Thiébaut
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-39.jpg
Fichier image/jpeg, 157k
Titre Figure 56 - Schematic illustration of the macro-traces observed on a flint denticulate used to saw dry poplar
Crédits Photographs and CAD: C. Thiébaut
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-40.jpg
Fichier image/jpeg, 161k
Titre Figure 57 - Schematic illustration of the macro-traces observed on a flint denticulate used to saw oak
Crédits Photographs and CAD: C. Thiébaut
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-41.jpg
Fichier image/jpeg, 158k
Titre Figure 58 - Schematic illustration of the macro-traces observed on a flint denticulate used to saw dry poplar
Crédits Photographs and CAD: C. Thiébaut
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-42.jpg
Fichier image/jpeg, 153k
Titre Figure 59 - Schematic illustration of the macro-traces observed on a flint denticulate used in the butchery of a wild boar
Crédits Photographs and CAD: C. Thiébaut
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-43.jpg
Fichier image/jpeg, 146k
Titre Figure 60 - Schematic illustration of the macro-traces observed on a flint denticulate used to disarticulate the left hind limb bones of a bison
Crédits Photograph and CAD: C. Thiébaut
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-44.jpg
Fichier image/jpeg, 145k
Titre Figure 61 - Schematic illustration of the macro-traces observed on a quartzite denticulate used to disarticulate the right hind limb bones of a bison
Crédits Photograph and CAD: C. Thiébaut
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-45.jpg
Fichier image/jpeg, 98k
Titre Figure 62 - Schematic illustration of the macro-traces observed on a quartzite denticulate used to disarticulate the right hind limb bones of a bison
Crédits Photograph and CAD: C. Thiébaut
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-46.jpg
Fichier image/jpeg, 102k
Titre Figure 63 - Micro-fractures observed on the denticles of the experimental denticulates
Crédits Photographs: A. Coudenneau
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-47.jpg
Fichier image/jpeg, 60k
Titre Figure 64 - Overview of the macro-traces observed on the experimental notched pieces, according to the material worked and the mode of use
Crédits Photographs: A. Coudenneau ; CAD: C. Thiébaut and É. Claud
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-48.jpg
Fichier image/jpeg, 256k
Titre Figure 65 - Examples of points used experimentally as hunting weapons
Légende a: ventral surface before and after use and detail of the apical scarring with a step termination; b: dorsal and ventral surfaces before and after use and detail of the apical scarring with a spin off termination (A) and the associated scars (B and C)
Crédits Photographs and CAD: A. Coudenneau
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-49.jpg
Fichier image/jpeg, 179k
Titre Figure 66 - Examples of fractures observed on points used experimentally as hunting weapons
Légende a: apical oblique scarring with a step termination and a step-terminating bending transverse fracture; b: step-terminating bending fracture; c: apical scarring with a step termination; d: apical oblique scarring with a spin off termination
Crédits Photographs: A. Coudenneau
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-50.jpg
Fichier image/jpeg, 121k
Titre Figure 67 - Examples of scarring observed on points used in hunting
Légende a: association of quadrangular and half-moon scarring with step terminations; b: quadrangular scarring with step terminations; c: combination of quadrangular, semi-circular, and trapezoidal scarring with step terminations. The arrows indicate the direction of movement
Crédits photographs: A. Coudenneau).
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-51.jpg
Fichier image/jpeg, 82k
Titre Figure 68 - Point no. 2
Légende a: experimental set-up; b: point before use; c: unused crystal on the edge before use; d: abraded crystal on the left edge
Crédits Photographs a-b: PCR Des Traces et des Hommes; c-d: C. Lemorini and F. Venditti
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-52.jpg
Fichier image/jpeg, 254k
Titre Figure 69 - Point no. 5
Légende a: experimental set-up; b: micro-traces produced during experimentation: crystal showing a row of superficial abrasions after use
Crédits Photographs a: PCR Des Traces et des Hommes; b: C. Lemorini and F. Venditti
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-53.jpg
Fichier image/jpeg, 112k
Titre Figure 70 - Point no. 6
Légende a: experimental set-up; b: point before use; c :point fracture ; d: micro-traces due to experiment ; e: cristal before use
Crédits Photographs a-b: PCR Des Traces et des Hommes; c-e: C. Lemorini and F. Venditti
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-54.jpg
Fichier image/jpeg, 214k
Titre Figure 71 - Point no. 7
Légende a: experimental set-up; b: point before use; c: micro-traces produced during the experiment: crystal showing abrasions that can be attributed to contact with meaty tissues after use
Crédits Photographs a-b: PCR Des Traces et des Hommes; c: C. Lemorini and F. Venditti
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-55.jpg
Fichier image/jpeg, 120k
Titre Figure 72 - Point no. 4
Légende a: experimental set-up; b: point before use; c: proximal part of the point remaining in the haft after fracture
Crédits Photographs: PCR Des Traces et des Hommes
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-56.jpg
Fichier image/jpeg, 131k
Titre Figure 73 - Point no. 4
Légende a, c, and e: crystals observed before use; b: smooth and polished crystal after use; d: crystal with a slight abrasion oriented perpendicular to the point, after use; f: crystal bearing traces typical of butchery, after use
Crédits Photographs: C. Lemorini and F. Venditti
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-57.jpg
Fichier image/jpeg, 115k
Titre Figure 75 - Scarring observed on experimental points used in butchery activities
Légende a: combination of quadrangular and triangular scarring with hinge and step terminations; b: combination of quadrangular, trapezoidal, semi-circular, and triangular scarring with feather and step terminations; c: view along the tip of the edge, showing a pattern alternating scarring; d: combination of quadrangular and trapezoidal scarring at the distal extremity of a point
Crédits photographs: A. Coudenneau
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-58.jpg
Fichier image/jpeg, 91k
Titre Figure 76 - Experimental point used in butchery activities and the macro-traces produced during use
Légende a-c: combinations of scarring observed
Crédits Photographs and CAD: A. Coudenneau
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-59.jpg
Fichier image/jpeg, 263k
Titre Figure 77 - Mousterian point used in punctiform contact on a tanned hide and the associated macroscopic and microscopic traces
Légende a: scarring and bending fracture
Crédits Photographs and CAD: A. Coudenneau
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-60.jpg
Fichier image/jpeg, 179k
Titre Figure 78 - Scarring on points related to working wood
Légende a-b: scarring produced during continuous longitudinal contact; c: scarring produced during percussive longitudinal contact; d: scarring produced during transverse continuous contact; e-f: scarring produced during punctiform contact, observed on the dorsal surface of the point (e) and the ventral surface (f )
Crédits photographs: A. Coudenneau
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-61.jpg
Fichier image/jpeg, 175k
Titre Figure 79 - Traces observed on experimental points that came into contact with bone: microscopic scarring produced by scraping the periosteum
Crédits Photograph: A. Coudenneau
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-62.jpg
Fichier image/jpeg, 59k
Titre Figure 80 - Traces observed on experimental points used to perforate shells. Mousterian point before and after use
Légende a: scarring; b: transverse bending fracture and associated crushing
Crédits photographs: A. Coudenneau
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-63.jpg
Fichier image/jpeg, 141k
Titre Figure 81 - Steps in the manufacture of experimental flake cleavers.
Légende a: the selection of ophite cobbles and a passive hammerstone from the alluvium of the Nive River; b: abandoned core and passive hammerstone; c: refitting of the blanks produced from the core; d: retouch of the blanks; e: production of large blanks; f: photograph of a flake cleaver before use; g: mould of active edge before use
Crédits Photographs: PCR Des Traces et des Hommes
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-64.jpg
Fichier image/jpeg, 383k
Titre Figure 82 - Hafting used in the experiments in percussion
Légende On the left, male hafts with straight handles; on the right: juxtaposed and male hafts with curved handles
Crédits Photographs: PCR Des Traces et des Hommes
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-65.jpg
Fichier image/jpeg, 60k
Titre Figure 83 - Overview of macro-traces produced during experimentation according to gesture and the hardness of the material worked
Crédits Photographs and CAD: É. Claud and M. Deschamps
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-66.jpg
Fichier image/jpeg, 302k
Titre Figure 84 - Schematic illustration of the macro-traces observed on a hafted flake cleaver used in percussion on dry oak
Crédits Photographs and CAD: M. Deschamps
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-67.jpg
Fichier image/jpeg, 246k
Titre Figure 85 - Schematic illustration of the macro-traces observed on a hand-held flake cleaver used in percussion on green ash wood
Crédits Photographs and CAD: M. Deschamps
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-68.jpg
Fichier image/jpeg, 262k
Titre Figure 86 - Schematic illustration of the macro-traces observed on a hafted flake cleaver used in percussion on green ash wood
Crédits Photographs and CAD: M. Deschamps
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-69.jpg
Fichier image/jpeg, 228k
Titre Figure 87 - Schematic illustration of the macro-traces observed on a hafted flake cleaver used on green poplar
Crédits Photographs and CAD: M. Deschamps
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-70.jpg
Fichier image/jpeg, 206k
Titre Figure 88 - Schematic illustration of the macro-traces observed on a hand-held flake cleaver used in percussion in the course of butchery activities
Crédits Photographs and CAD: M. Deschamps
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-71.jpg
Fichier image/jpeg, 249k
Titre Figure 89 - Schematic illustration of the macro-traces observed on a flake cleaver used for cutting during butchery activities
Crédits Photographs and CAD: M. Deschamps
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-72.jpg
Fichier image/jpeg, 166k
Titre Figure 90 - Schematic illustration of the macro-traces observed on a hafted cleaver used in percussion for the disarticulation of a horse carcass
Crédits Photographs and CAD: M. Deschamps
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-73.jpg
Fichier image/jpeg, 241k
Titre Figure 91 - Schematic illustration of the macro-traces observed on a hafted cleaver used in percussion for the disarticulation of a lamb
Crédits Photographs and CAD: M. Deschamps
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-74.jpg
Fichier image/jpeg, 197k
Titre Figure 92 - Schematic illustration of the macro-traces observed on a hafted cleaver after its use in percussion for the disarticulation of a bison
Crédits Photographs and CAD: M. Deschamps
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-75.jpg
Fichier image/jpeg, 148k
Titre Table 20 - List of assemblages attributed to the Acheulean in which use-wear on bifacial pieces is mentioned, based on studies published as of 2010
Légende A cross indicates that the number of pieces or active zones is unknown; in this case, the minimum number (one) is used in the calculation of totals“
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-76.jpg
Fichier image/jpeg, 558k
Titre Table 21 - List of assemblages dated to the Early or Middle Palaeolithic (aside from the Acheulean) in which traces of use were mentioned on bifacial pieces, as published in 2010
Légende A cross indicates that the number of pieces or active zones is unknown; in this case, the minimum number (one) is used in the calculation of totals“.
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-77.jpg
Fichier image/jpeg, 610k
Titre Figure 93 - Schematic illustration of the macro-traces observed on a flint biface used to scrape and saw dry wood
Crédits Photographs and CAD: É. Claud and M. Coutureau
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-78.jpg
Fichier image/jpeg, 192k
Titre Figure 94 - Schematic illustration of the macro-traces observed on a flint biface used in percussion to fell a tree
Crédits Photographs and CAD: É. Claud and M. Coutureau
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-79.jpg
Fichier image/jpeg, 124k
Titre Figure 95 - Schematic illustration of the macro-traces observed on a flint biface used in cutting during butchery
Crédits Photographs and CAD: É. Claud and M. Coutureau
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-80.jpg
Fichier image/jpeg, 149k
Titre Figure 96 - Schematic illustration of the macro-traces observed on a flint biface used to cut leather thongs from dry hide
Crédits Photographs and CAD: É. Claud and M. Coutureau
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-81.jpg
Fichier image/jpeg, 117k
Titre Figure 97 - Microscopic view of the retouch removals from sharpening and microscopic traces of percussion on the experimental bifaces and biface manufacturing flakes
Légende a: sharpening retouch; b: sharpening retouch (aligned, invasive removals, bending initiation) (superimposed removals, cone initiation); c: additive streak related to percussion during sharpening retouch; d: spot of polish on an exposed zone related to contact with an organic hammer; e: smooth-bottomed striations related to percussion with a quartzite hammerstone
Crédits Photographs: É. Claud
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-82.jpg
Fichier image/jpeg, 98k
Titre Figure 98 - Photographs of the experimental macro- and micro-traces related to technical abrasion on the residual zones of the edges of bifaces and the platforms of biface manufacturing flakes
Légende a: crushing produced by scraping with a sandstone hammer (platform, biface manufacturing flake); b: striations produced by abrasion from scraping with a sandstone hammer (platform, biface manufacturing flake); c: detail of crushing produced by abrasion from scraping with a sandstone hammer (platform, biface manufacturing flake); d: microscopic striations produced by abrasion on schist; e: microscopic striations produced by abrasion with sandstone on the proximal end of a biface manufacturing flake e (platform); f: microscopic striations produced by abrasion with sandstone on the proximal end of a biface manufacturing flake (dorsal surface)
Crédits Photographs: É. Claud
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-83.jpg
Fichier image/jpeg, 116k
Titre Figure 99 - Overview of the characteristic macro-traces produced during experimental use of bifaces, according to the hardness and nature of the materials worked and the mode of action
Crédits Photographs and CAD: É. Claud
URL http://journals.openedition.org/palethnologie/docannexe/image/4070/img-84.jpg
Fichier image/jpeg, 371k
Haut de page

Pour citer cet article

Référence électronique

Émilie Claud, Céline Thiébaut, Aude Coudenneau, Marianne Deschamps, Vincent Mourre, Michel Brenet, Maria Gema Chacón‑Navarro, David Colonge, Cristina Lemorini, Serge Maury, Christian Servelle et Flavia Venditti, « Stone tool reference collection »Palethnologie [En ligne], 10 | 2019, mis en ligne le 01 novembre 2019, consulté le 19 mars 2024. URL : http://journals.openedition.org/palethnologie/4070 ; DOI : https://doi.org/10.4000/palethnologie.4070

Haut de page

Auteurs

Émilie Claud

University of Bordeaux, UMR 5199 – Pacea / INRAP
emilie.claud[at]inrap.fr

Articles du même auteur

Céline Thiébaut

University Toulouse Jean Jaurès, UMR 5608 – Traces
celine.thiebaut[at]wanadoo.fr

Articles du même auteur

Aude Coudenneau

coudenneau.aude[at]orange.fr

Articles du même auteur

Marianne Deschamps

UNIARQ – Centro de Arqueologia da Universidade de Lisboa / Université Toulouse Jean Jaurès / CNRS, UMR 5608 – Traces
mardesch1690[at]gmail.com

Articles du même auteur

Vincent Mourre

University Toulouse Jean Jaurès, UMR 5608 – Traces / INRAP
vincent.mourre[at]inrap.fr

Articles du même auteur

Michel Brenet

University of Bordeaux, UMR 5199 – Pacea / INRAP
michel.brenet[at]inrap.fr

Articles du même auteur

Maria Gema Chacón‑Navarro

Institut Català de Paleoecologia Humana i Evolució Social / Universitat Rovira i Virgili / UMR 7194 – HNHP
gchacon[at]prehistoria.urv.cat

Articles du même auteur

David Colonge

University Toulouse Jean Jaurès, UMR 5608 – Traces / INRAP
david.colonge[at]inrap.fr

Articles du même auteur

Cristina Lemorini

LTFAPA Laboratory – University of Rome “La Sapienza”
cristina.lemorini[at]uniroma1.it

Articles du même auteur

Serge Maury

serge.maury24[at]wanadoo.fr

Christian Servelle

christian.servelle[at]gmail.com

Flavia Venditti

Department of Archaeology and Near Eastern Cultures, Tel Aviv University
flavia.venditti[at]gmail.com

Articles du même auteur

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