Skip to navigation – Site map

HomeIssues34The use of splits and flakes as b...

The use of splits and flakes as blanks for debitage or shaping: examples from the late Lower and Middle Palaeolithic of Southern France

L’utilisation d’éclats et de galets fendus (« splits ») comme supports de débitage ou façonnage : exemples de la fin du Paléolithique ancien et du Paléolithique moyen dans le sud de la France
Cyrielle Mathias, Sébastien Bernard-Guelle and Cyril Viallet
p. 200-233

Abstracts

Our study investigates the use of flakes and splits as matrices (i.e., blanks) for debitage or shaping in the context of (mainly) cobble-based lithic assemblages from the late Lower and Middle Palaeolithic in southern France. Here, we present our findings from seven lithic assemblages comprising two main strategies: the ramification of chaînes opératoires, where flakes from primary production are repurposed, and integrative strategies, where blanks such as flakes or splits are specifically produced for further debitage. Through the comparative analysis of lithic assemblages, we evaluate the technical and economic implications of raw material morphology, local availability, and site-specific adaptations.
Our results highlight the diverse applications of flake and split matrices, for example, in Discoid and Levallois debitage methods, as well as their use in shaping Large Cutting Tools. Additionally, this research highlights the adaptability of bipolar-on-anvil methods in processing pebbles and cobbles with specific shapes and morphologies. The findings contribute to broader discussions on technological innovation, resource management, and hominin behaviours during the Lower and Middle Palaeolithic.

Top of page

Full text

Introduction

1Flakes and splits (split cobbles/pebbles) are preferential blanks produced and selected by hominins for the production of tools, whether by debitage or shaping. They were used in varying proportions throughout the Palaeolithic, according to chronological periods and local traditions.

2The use of flakes as matrices for debitage can be considered as the recycling of fully planned production, and is part of a broader techno-economic management system (fig. 1 c2). We thus distinguish two different phenomena: the ramification of chaînes opératoires or branching strategies (Bourguignon et al. 2004; Faivre 2008), and the integrative strategy (Mathias and Bourguignon 2020).

3Ramification corresponds to the selection of a flake issued from the main production (waste as well as end-products), whereas the integrative strategy corresponds to the specific production of flakes to be used as blanks for debitage. The integrative strategy has been identified at Late Acheulean or Early Middle Palaeolithic sites in southern France, such as Barbas C’ 3 base or Orgnac 3 level 2 (Boëda et al. 1996, 2004; Mathias 2018; Mathias and Bourguignon 2020). On the other hand, ramification processes are much more common, especially during the Middle Palaeolithic (Bourguignon and Turq 2003; Bourguignon et al. 2004; Faivre 2008; Rios-Garaizar et al. 2015; Vaquero et al. 2015; Turq et al. 2017; Romagnoli et al. 2018; Vaissié 2021). They are attested since the Lower Palaeolithic (Barsky et al. 2015; Aureli et al. 2016), particularly in the Levant (Agam et al. 2015; Parush et al. 2015; Wojtczak 2015; Agam and Barkai 2018; Venditti et al. 2019).

4The use of flakes as matrices for shaping Large Cutting Tools (LCTs) is also widespread throughout the Lower and Middle Palaeolithic (fig. 1 c1). During the Acheulean and Vasconian periods, for example, the use of large flakes (i.e., flakes more than 10 centimeters in length) as blanks has been documented for the production of handaxes and cleavers (Tavoso 1975; Sharon 2008, 2010; Deschamps 2011, 2017; Rubio-Jara et al. 2016; Baena Preysler et al. 2018; García-Vadillo et al. 2022; Wilson et al. 2024). Flakes were also preferentially selected to shape small handaxes during the Mousterian in Western Europe (Cliquet, Lautridou 1988; Turq 2000; Jaubert 2001; Soressi 2002; Claud 2008). The use of flakes as matrices for shaping can be seen as a specific initialisation method, but also as part of the volumetric configuration of the tool. In the case of cleavers, the use of flakes is inherent to their definition and corresponds to an anticipation of the morpho-technical characteristics of the end-product (Tixier 1956; Tavoso 1975; Inizan et al. (dir.) 1995; Mourre 2003; Deschamps 2014; Capdevielle, Colonge 2023). In this sense, it can be assimilated to the integrative strategy.

Figure 1. Schematic representation of the technical processes discussed in the article. A: Split cobbles along different axes, then used as debitage or shaping matrices (blanks); B: Fracturing on an anvil in the central part of the cobble, with the possibility of using the fragments as tools (retouching) or matrices for debitage; C: Core-on-Flake, use of flakes as matrices for debitage or shaping (drawing C. Mathias).
Représentation schématique des procédés techniques discutés dans l'article. A : Galets fendus selon différents axes, puis utilisés comme matrices de débitage ou de façonnage (supports) ; B : Fracturation sur enclume dans la partie centrale du galet, avec possibilité d'utiliser les fragments comme outils (par la retouche) ou matrices de débitage ; C : Nucléus sur éclats, utilisation des éclats comme matrices de débitage ou de façonnage (dessin C. Mathias).

Figure 1. Schematic representation of the technical processes discussed in the article. A: Split cobbles along different axes, then used as debitage or shaping matrices (blanks); B: Fracturing on an anvil in the central part of the cobble, with the possibility of using the fragments as tools (retouching) or matrices for debitage; C: Core-on-Flake, use of flakes as matrices for debitage or shaping (drawing C. Mathias). Représentation schématique des procédés techniques discutés dans l'article. A : Galets fendus selon différents axes, puis utilisés comme matrices de débitage ou de façonnage (supports) ; B : Fracturation sur enclume dans la partie centrale du galet, avec possibilité d'utiliser les fragments comme outils (par la retouche) ou matrices de débitage ; C : Nucléus sur éclats, utilisation des éclats comme matrices de débitage ou de façonnage (dessin C. Mathias).

5Until recently however, little attention had been paid to the status of splits. This has changed of late with increasing focus on the bipolar-on-anvil (both as a technique and a method), highlighting adaptability and diversified application modes (Donnart et al. 2009; Mourre and Jarry 2010; Mourre et al. 2010; Diez-Martín et al. 2011; De la Peña 2015; Byrne et al. 2016; De Lombera-Hermida et al. 2016; Horta et al. 2019; Arrighi et al. 2020; Yeşilova et al. 2024). Here, we consider the split technique to consist of the fracturing of a pebble/cobble (regardless of morphology) into two half-pebbles/cobbles by bipolar percussion on an anvil. The fractured surfaces are flat, often of equal size, and may bear traces of a blow and a counter-blow. It is not possible to use conventional terminology (i.e., flake, core) as the two parts are indistinguishable. Depending on whether it is fractured along its longitudinal or transversal axis (fig. 1 a1, a2) or on its central face (fig. 1 a3), the morphology of the products differs. When violent percussion is applied to the centre of a cobble, producing several products and by-products, we can also refer to fractured cobbles. A spherical pebble/cobble fractured in two will also offer a different volume (fig. 1 b).

6In this paper, we focus on determining the proportion of products made on flakes and splits at several late Lower and Middle Palaeolithic sites in southern France, in contexts where cobbles and pebbles are the main available raw materials. Pebbles and cobbles are frequently used during Prehistory, in all geographical regions (Collina-Girard, Turq 1991; de Beaune 1997; Collina-Girard 1997; Nicoud 2010; Pérez-Balarezo, Guibert 2023). The aim here is to address technical and economic issues, in relation to the morphologies of the available materials, but also to discuss potential differential treatments, and possible correlations with distances from raw material outcrops.

1 | Selected sites

7We selected several lithic assemblages from sites attributed to the late Lower Palaeolithic and Middle Palaeolithic in southern France (fig. 2), where cobbles and pebbles are the main types of raw materials (tab. 1). The sites are dated from MIS 7 to MIS 3 and correspond mainly to open-air sites excavated during preventive archaeology operations. Here, we briefly describe the context of the different assemblages:

Figure 2. Map showing the location of the studied sites in southern France (made using QGIS 3.16.16, data ESRI - C. Mathias).
Carte montrant la localisation des sites étudiés dans le sud de la France (réalisée avec QGIS 3.16.16, données ESRI - C. Mathias).

Figure 2. Map showing the location of the studied sites in southern France (made using QGIS 3.16.16, data ESRI - C. Mathias). Carte montrant la localisation des sites étudiés dans le sud de la France (réalisée avec QGIS 3.16.16, données ESRI - C. Mathias).

Table 1. Main characteristics of the selected sites. *As the site is currently excavated, this count is not definitive.
Principales caractéristiques des sites sélectionnés. *Le site étant en cours de fouille, ce décompte n'est pas définitif.

Table 1. Main characteristics of the selected sites. *As the site is currently excavated, this count is not definitive. Principales caractéristiques des sites sélectionnés. *Le site étant en cours de fouille, ce décompte n'est pas définitif.

8Le Mas des Caves I (Lunel-Viel, Hérault) is the oldest site, dated to MIS 7 and attributed to the Early Middle Palaeolithic (Boutillier-Uzunidis 2017; Brugal et al. 2021; Falguères et al. 2024). The cave was initially excavated by E. Bonifay and is now undergoing new excavations and studies directed by J.-Ph. Brugal, in order to better understand site formation processes and the relationship between carnivore and hominin activities (Bonifay 1968; Le Grand 1993; Fosse 1996; Brugal et al. 2021). New spatial analysis of faunal and lithic refits indicates that some anthropic activities took place in the cave (Giuliani et al. in prep.).

9Le Cassé (Cornebarrieu, Haute-Garonne) was excavated in 2020 by Paléotime under the supervision of P. Tallet (6000 m²). The excavation yielded a residual lithic assemblage from the Acheulean (“Acheuléen Pyrénéo-Garonnais”), attributed to MIS 5 (Tallet et al. 2022; Viallet et al. 2024).

10Le Bois de l’Hôpital (Saint-Sulpice-la-Pointe, Tarn) was excavated for six months between 2015 and 2016 by Paléotime, under the supervision of S. Bernard-Guelle and P. Tallet. Several levels were uncovered during the excavation of a vast area of 3.5 ha (Bernard-Guelle et al. 2016a, 2019; Viallet et al. 2022a). The lower archaeological level was attributed to the Acheulean lato sensu and dated to the end of MIS 6 or MIS 5e, whereas the Mousterian level 3 is dated to MIS 3.

11In Corrèze, three lithic assemblages unearthed during preventive archaeology operations are studied. Brive-Laroche North, Brive-Laroche South (Brive-la-Gaillarde, Corrèze) and Les Hauts de Lestrade (Saint-Pantaléon-de-Larche, Corrèze) were respectively excavated by Paléotime under the supervision of C. Viallet and G. Monin in 2019 (2800 m²) and C. Viallet in 2021 (10,000 m²). The two excavated areas at Brive-Laroche are dated between MIS 5 and 4 (probably 5a-c for BLA North; Viallet et al. 2022b; Mathias et al. 2023), whereas the lithic assemblage from Les Hauts de Lestrade (which appears to be a palimpsest of two layers) probably dates to MIS 5 (Viallet et al. 2023).

2 | Results

2.1 | Selected raw materials

Bois de l’Hôpital

12Quartz cobbles and pebbles are the main raw materials used in the Acheulean and Mousterian levels at Bois de l’Hôpital. In archaeological unit 3, quartz represents the vast majority of the lithic assemblage (88.2 %). The second most represented category consists of flint and lydian, with 8.6 % (No. = 1346). The other diversified raw materials comprise quartzite cobbles/pebbles (No. = 267, 1.7 %), and other rocks, such as sandstone, schist, conglomerates, etc.

13The Acheulean level is mainly composed of quartz cobbles/pebbles (90.8 %), followed by flint (4.6 %), quartzite (2.7 %) and other types of rare raw materials (lydian, sandstone, schist, etc.).

14Quartz was selected from alluvial deposits and colluvial outwash rich in pebbles/cobbles, resulting from the erosion of the Massif central and its sedimentary cover. This term encompasses a wide variety of lithological types, all collected in the immediate vicinity of the site (Fernandes and Minet in Bernard-Guelle et al., 2019). These formations also contain other raw materials, such as lydian, gneiss or sandstone.

15The flint category also comprises highly diversified siliceous materials from secondary contexts. These include flints from the Castrais, Verdier, the Tarn basin and the Pyrenees (Fernandes and Minet in Bernard-Guelle et al. 2019). It is thus difficult to define the exploited mineral domain.

Brive-Laroche and Les Hauts de Lestrade

16As the sites are located just a few kilometres apart, the raw materials from the three lithic assemblages of Brive-Laroche North, South, and Les Hauts de Lestrade are presented together in this section.

17At Brive-Laroche North, out of a total of 333 pieces, quartz cobbles and pebbles are the main raw materials (No. = 217; 65.2 %), followed by flint nodules (No. = 105; 31.5 %), and more rarely gneiss cobbles (No. = 4; 1.3 %). The lithic assemblage at Brive-Laroche South, is larger (No. = 972), and consists mainly of quartz (No. = 777; 79.9 %), followed by flint (No. = 184; 18.9 %), as well as small proportions of other rocks (gneiss, granite and undetermined metamophic rocks, No. = 11; 1.1 %).

18The lithic assemblage of Les Hauts de Lestrade yielded 1981 pieces, dominated by quartz cobbles and pebbles (No. = 1820; 92 %), followed by flint nodules (No. = 149; 7 %). The other diverse raw materials consist of gneiss, granite, hornfels and limestone.

19Quartz pebbles and cobbles were collected locally, within a few kilometers of the site, in several sedimentary formations in the Brive basin (Permian sandstone). They were also collected from various surface formations in a secondary position. Quartz cobbles and pebbles, as well as gneiss, are also present in various alluvial formations, including the Corrèze and Vézère alluviums.

20Broadly speaking, the flints from all three assemblages were sourced from the same outcrops, with some variations (flint from the Haut-Agenais at Brive-Laroche and flint from the Bergeracois at Les Hauts de Lestrade). Flints were preferentially collected in weathered formations located on the plateaus (mainly flint nodules or small rounded nodules), mainly in a radius of 30 km of the site (Viallet et al. 2022; Mathias et al. 2023).

Le Cassé

21The lithic assemblage of Le Cassé (Cornebarrieu) yielded a total of 1717 pieces from UPS 6.1. Quartzite cobbles and pebbles are the main raw materials (No. = 1494; 87 %), followed by quartz cobbles and pebbles (No. = 162; 9.4 %). Flints and lydian are also used to a lesser extent (respectively, No. = 51; 3 % and No. = 10; 0.6 %).

22The quartzites consist mainly of “cold-toned” quartzites from the Garonne, and to a lesser extent, light-coloured quartzites. Raw materials were almost exclusively sourced from the alluvial formations of the Garonne, with, in addition, two groups of flints from the Tarn basin, and one quartz group from the right bank of the Garonne. This lithological range roughly corresponds to the raw material variability observed on the left bank of the river near the site (Tallet et al. 2022, Viallet et al. 2024).

Mas des Caves I

23Flint cobbles and pebbles are the main raw materials used at Mas des Caves I (Lunel-Viel), followed by quartz, quartzite and sandstone cobbles and pebbles.

24All the raw materials mentioned above can be found in the alluvial formations in the vicinity of the site. The flints are known as “Costières du Gard” flints (Grégoire and Bazile 2005; Lebègue 2012), comprising various flint facies from different geological formations, but all bearing a neocortex typical of the Durance and Rhône alluviums. These alluviums comprise one of the main secondary flint outcrops in the Languedoc region, and also contain quartz, quartzite and sandstone cobbles.

2.2 | The use of splits and flakes as blanks

Bois de l’Hôpital, Acheulean level

25The number of lithic artefacts found in this archaeological level is relatively low, and it is thus attributed to the Acheulean lato sensu.

26This level comprises 16 cores in total (15 quartz cores and one cinerite core), including two quartz split cobbles with two removals. The other methods correspond to unifacial exploitation (including unifacial Discoid) or S.S.D.A. debitage (i.e. Alternating Surface Debitage System or “Système par surface de débitage alternée”, Ashton 1992; Forestier 1993), with sometimes three debitage surfaces.

27Nine Large Cutting Tools were identified in this level, two of which were shaped on split quartz cobbles (fig. 3), and three on large quartzite flakes (one of which was transformed into a cleaver).

Figure 3. Large Cutting Tools in quartz and quartzite from the Acheulean level of Bois de l'Hôpital. A-B: handaxes made on splits; C: handaxe made on a large flake; a shaping flake is refitted at the base of the ventral surface (drawings R. Picavet).
Grands outils façonnés en quartz et quartzite du niveau acheuléen de Bois de l'Hôpital. A-B : bifaces sur galets fendus; C: biface sur grand éclat ; un éclat de façonnage remonte sur la base de la surface ventrale (dessins R. Picavet).

Figure 3. Large Cutting Tools in quartz and quartzite from the Acheulean level of Bois de l'Hôpital. A-B: handaxes made on splits; C: handaxe made on a large flake; a shaping flake is refitted at the base of the ventral surface (drawings R. Picavet). Grands outils façonnés en quartz et quartzite du niveau acheuléen de Bois de l'Hôpital. A-B : bifaces sur galets fendus; C: biface sur grand éclat ; un éclat de façonnage remonte sur la base de la surface ventrale (dessins R. Picavet).

28Despite the low quantity of lithics in this level, it is noteworthy that flakes or splits were primarily used as blanks for shaping and rarely for debitage.

Bois de l’Hôpital, Level 3

29Considering the high number of lithics in this level (No. = 15,460), split pebbles/cobbles in quartz are rare (No. = 23, fig. 4, 6). They were knapped along their longest axis on an anvil. Some of them, in particular small flattened spheroid pebbles, were refitted, but the reasons underlying such breakage (use of splits as matrices for debitage?) could not be confidently ascertained. It is possible that some of the split pebbles (and fragments) result from accidental breakage during initial use as hammerstones. This may also be the case for the rare impacted pebbles (No. = 17) with one or more removals along their longest axis. In all cases, bipolar-on-anvil debitage is the predominant production method identified on cores (No. = 243). Discoid debitage sensu lato completes the series (No. = 161).

Figure 4. Refit of a split cobble, both parts of which were selected for secondary knapping (unifacial discoid intent?) from level 3 of Bois de l’Hôpital (photograph and CAD S. Bernard-Guelle).
Remontage d'un galet fendu, dont les deux parties ont été sélectionnées pour un débitage secondaire (intention d'une production discoïde unifaciale ?) provenant du niveau 3 du Bois de l'Hôpital (photographie et DAO S. Bernard-Guelle).

Figure 4. Refit of a split cobble, both parts of which were selected for secondary knapping (unifacial discoid intent?) from level 3 of Bois de l’Hôpital (photograph and CAD S. Bernard-Guelle). Remontage d'un galet fendu, dont les deux parties ont été sélectionnées pour un débitage secondaire (intention d'une production discoïde unifaciale ?) provenant du niveau 3 du Bois de l'Hôpital (photographie et DAO S. Bernard-Guelle).

30The main category in this assemblage consists of fractured pebbles (No. = 515; 24.4 % or 3.7 % of the total quartz elements - fig. 5). This category is sometimes assimilated to split pebbles, but it refers to intentionally fractured pebbles/cobbles on an anvil. This is not an organised debitage method, but rather a fragmentation method for various secondary objectives (initialisation for another debitage sequence). These pebbles/cobbles are hit with a hammer stone, usually on an anvil, along a perpendicular axis to their greatest elongation and through the thickness of the pebble (one or more removals). The morphological and dimensional characteristics of these products are variable: (1) thin (< or = 5 cm) and generally flattened or flat, sometimes spheroid with facets, (2) or even thick (> 5 cm), flattened, more rarely ovoid or with facets. They probably generally stem from production, but are separated from the cores here because most of them show traces of fracturing/impacts but no series of removal scars. Percussion can also be more or less centred on the flat face of the pebble, causing it to fracture into several large fragments (with or without surface fractures). These fragments are not always identifiable as flakes and often bear no potential cutting edges (fig. 5). The numerous cases and refits tend to show that these pebbles/cobbles do not exhibit diagnostic impact marks resulting from a particular activity that led to accidental fracturing (such as use as an anvil in the context of debitage or following direct percussion on their surface in connection with the crushing or grinding of fruit, plants, etc.). The intention to obtain large matrices for debitage is evident in the case of violent percussion in the centre of the cobbles, or along an apparent surface fracture. Some refits of two hemi-cobbles show the beginnings of unifacial Discoid type debitage (fig. 4). Some of the fragments may also have been used for domestic or culinary purposes, or in connection with heating/combustion activities. Traces of heating are unfortunately rare, but are nevertheless visible on certain fragments. Finally, some of the fragments seem to have been intended for retouching; some were transformed into scraper-type tools by continuous reverse retouching of the cutting edge opposite a cortical back.

Figure 5. Refit of a large ovoid quartz cobble, fractured by bipolar percussion on an anvil in its centre (photograph Paléotime).
Remontage d'un gros galet de quartz ovoïde, fracturé par percussion bipolaire sur enclume en son centre (photographie Paléotime).

Figure 5. Refit of a large ovoid quartz cobble, fractured by bipolar percussion on an anvil in its centre (photograph Paléotime). Remontage d'un gros galet de quartz ovoïde, fracturé par percussion bipolaire sur enclume en son centre (photographie Paléotime).

31Some quartz flakes were occasionally used as matrices for debitage. A fine-grained flake was knapped on the ventral surface (secant and convergent debitage, i.e., Discoid flaking?), another flake was knapped in its thickness. Two flake-cores were knapped on their ventral surface, using a method similar to the Levallois flaking method (same volumetric construction, Levallois-like or “type-Levallois”).

32Large Cutting Tools have also been identified (No. = 111; 89 of which are in quartz). Most of them correspond to unifacially shaped cobbles/pebbles, but five are handaxes or handaxe fragments in quartz. They are frequently shaped on large flakes (fig. 7).

Figure 6. Split cobble, one fragment of which was secondarily knapped (photograph Paléotime).
Galet fendu, dont un fragment a été secondairement taillé (photographie Paléotime).

Figure 6. Split cobble, one fragment of which was secondarily knapped (photograph Paléotime). Galet fendu, dont un fragment a été secondairement taillé (photographie Paléotime).

Figure 7. LCTs on flakes from Bois de l’Hôpital, level 3 (photograph C. Viallet).
Grandes pièces façonnées sur éclat du Bois de l'Hôpital, niveau 3 (photographie C. Viallet).

Figure 7. LCTs on flakes from Bois de l’Hôpital, level 3 (photograph C. Viallet). Grandes pièces façonnées sur éclat du Bois de l'Hôpital, niveau 3 (photographie C. Viallet).

33Diversified flints make up the second main lithic raw material category, sometimes in the form of cobbles or pebbles (Bernard-Guelle et al. 2019). Ramification was clearly identified for these raw materials, as illustrated by several levels of ramification as part of Discoid debitage (fig. 8 A and B). In this case, a large ovoid flint pebble was fractured to produce large cortical flakes, two of which were reused as discoidal cores. A smaller cortical flake (with a Siret fracture), issued from the debitage of one of these secondary cores, was subsequently reused as a matrix for debitage (fig. 8 C).

Figure 8. Ramification process as part of discoidal debitage, flint cobble (photograph Paléotime).
Processus de ramification dans le cadre d'un débitage Discoïde, galet de silex (photographie Paléotime)

Figure 8. Ramification process as part of discoidal debitage, flint cobble (photograph Paléotime). Processus de ramification dans le cadre d'un débitage Discoïde, galet de silex (photographie Paléotime)

Brive-Laroche and Les Hauts de Lestrade

34The small lithic assemblage from Brive-Laroche North (217 quartz pieces, including 35 cores) is homogeneous. The cores consist of single-surface exploitation (S.S.D.A.), the bipolar-on-anvil technique and Discoid flaking. However, three Levallois-type cores attest to flaking methods using two hierarchically organized surfaces. Two of these were produced from flattened split pebbles (fig. 9 B). Another core on a split bears several secant removal scars (fig. 9 A). In addition, one globular cobble was split into two parts, with one of the surfaces showing the presence of several removals (short debitage or first stages of a more complex debitage - fig. 10). Several cores on quartz flakes were also identified (8.5 %). They correspond to a Kombewa-type debitage, as well as the exploitation of both ventral and dorsal surfaces (No. = 2; including bipolar-on-anvil technique).

Figure 9. Two cores on splits from Brive-Laroche North (photograph C. Mathias).
Deux nucléus sur galets fendus de Brive-Laroche Nord (photographie C. Mathias).

Figure 9. Two cores on splits from Brive-Laroche North (photograph C. Mathias). Deux nucléus sur galets fendus de Brive-Laroche Nord (photographie C. Mathias).

Figure 10. Refitted split cobble with a globular shape from Brive-Laroche North. One of the splits bears some removals (photograph C. Mathias).
Remontage d'un galet fendu de forme globulaire provenant de Brive-Laroche Nord. Un des hémi-galets présente des négatifs d'enlèvements (phototographie C. Mathias).

Figure 10. Refitted split cobble with a globular shape from Brive-Laroche North. One of the splits bears some removals (photograph C. Mathias). Remontage d'un galet fendu de forme globulaire provenant de Brive-Laroche Nord. Un des hémi-galets présente des négatifs d'enlèvements (phototographie C. Mathias).

35The chaînes opératoires on flint are fragmented (e.g., Turq et al. 2013), owing undoubtedly to the distance of the exploited outcrops from the site. Only eight cores were identified among the 105 flint pieces. Nearly all of these (No. = 6, possibly 7) correspond to cores-on-flakes illustrating various flaking modes (including one Kombewa-type core, two exploited on two surfaces, and Levallois cores).

36At Brive-Laroche North, evidence of chaîne opératoire ramification on flint and quartz (sometimes discrete for the latter) was identified, as well as the integrative strategy, composed of the integrated production of splits for debitage.

37In contrast, the lithic assemblage from Brive-Laroche South is more abundant. Among the 150 cores on quartz, four are cores-on-flakes illustrating a Kombewa-type mode or debitage in the thickness of the matrix (fig. 11). At least two cores from other categories are made on flakes, and probably four on splits. The knapping method used for these cores is Discoid, and the removals are relatively invasive (fig. 12). The assemblage also includes eight splits (hemi-cobbles).

Figure 11. Secondarily knapped quartz flake from Brive-Laroche South. The ventral surface is flaked by a large removal, made after the exploitation of the thickness (proximal bipolar-on-anvil exploitation - photograph C. Mathias).
Éclat de quartz secondairement débité, provenant de Brive-Laroche Sud. La surface ventrale présente un grand négatif d'enlèvement, réalisé après une première exploitation de l'épaisseur par percussion bipolaire sur enclume en partie proximale de l'éclat-matrice (photographie C. Mathias).

Figure 11. Secondarily knapped quartz flake from Brive-Laroche South. The ventral surface is flaked by a large removal, made after the exploitation of the thickness (proximal bipolar-on-anvil exploitation - photograph C. Mathias). Éclat de quartz secondairement débité, provenant de Brive-Laroche Sud. La surface ventrale présente un grand négatif d'enlèvement, réalisé après une première exploitation de l'épaisseur par percussion bipolaire sur enclume en partie proximale de l'éclat-matrice (photographie C. Mathias).

Figure 12. Two discoidal quartz cores probably made on splits from Brive-Laroche South, unifacial mode (photograph C. Mathias).
Deux nucléus de conception Discoïde unifaciale, probablement réalisés sur deux galets fendus provenant de Brive-Laroche Sud (photographie C. Mathias).

Figure 12. Two discoidal quartz cores probably made on splits from Brive-Laroche South, unifacial mode (photograph C. Mathias). Deux nucléus de conception Discoïde unifaciale, probablement réalisés sur deux galets fendus provenant de Brive-Laroche Sud (photographie C. Mathias).

38Nine of the 20 flint cores are cores-on-flakes (Kombewa, No. = 8; Dorsal surface exploitation, No. = 1). Four Levallois cores are also made on flakes. One of the Kombewa cores corresponds to the recycling of a side scraper. In total, at least 13 of the 20 cores were made on flakes (65 % - fig. 13).

Figure 13. Ramification processes on flint at Brive-Laroche South. A: refit of a broken core-on-flake, ventral unipolar convergent debitage from the distal part; B: Kombewa recurrent unipolar exploitation from the proximal part; C: Kombewa lateral flake; D: Kombewa lineal flaking and dorsal unipolar exploitation (after the truncation; photograph C. Mathias).
Processus de ramification identifiés sur silex à Brive-Laroche Sud. A : remontage d'un nucléus sur éclat fracturé (raccord), débitage unipolaire convergent sur la face inférieure depuis la partie distale ; B : nucléus sur éclat Kombewa récurrent, exploitation unipolaire depuis la partie proximale ; C : éclat Kombewa latéral ; D : nucléus Kombewa linéal et exploitation de la face supérieure après aménagement d'une troncature (photographie C. Mathias).

Figure 13. Ramification processes on flint at Brive-Laroche South. A: refit of a broken core-on-flake, ventral unipolar convergent debitage from the distal part; B: Kombewa recurrent unipolar exploitation from the proximal part; C: Kombewa lateral flake; D: Kombewa lineal flaking and dorsal unipolar exploitation (after the truncation; photograph C. Mathias). Processus de ramification identifiés sur silex à Brive-Laroche Sud. A : remontage d'un nucléus sur éclat fracturé (raccord), débitage unipolaire convergent sur la face inférieure depuis la partie distale ; B : nucléus sur éclat Kombewa récurrent, exploitation unipolaire depuis la partie proximale ; C : éclat Kombewa latéral ; D : nucléus Kombewa linéal et exploitation de la face supérieure après aménagement d'une troncature (photographie C. Mathias).

39Thus, as in the northern sector, a ramification of the chaînes opératoires is attested on both flint and quartz, with the occasional use of splits. However, in this context, quartz flakes or splits are specifically used as matrices for Discoid debitage.

40The technological composition of the lithic assemblage from Les Hauts de Lestrade differs significantly from those of Brive-Laroche North and South. Quartz cores are abundant (No. = 419, 444 including tested cobbles). This has been linked to post-depositional processes but also to the importance of bipolar debitage on anvils (No. = 113), the characteristic recurrent flakes (No. = 148), but also the presence of unknapped splits (No. = 81). Four refits confirm the production of these split cobbles on site. The high proportion of longitudinal splits is particularly significant. Most of the quartz matrices are whole cobbles/pebbles, followed by splits and flakes. Cores-on-flakes were identified on quartz (No. = 20) and correspond mainly to recurrent Kombewa-type flaking (fig. 14 A). Two Kombewa-type flakes were also observed, despite the difficulties surrounding the identification of these products on quartz. Some Discoid production initialisations were also possibly identified (fig. 14 B), as well as three Discoid cores-on-flakes, and one unifacial Discoid core on a split.

Figure 14. Cores on quartz flakes from Les Hauts de Lestrade. A: Kombewa recurrent; B: partially alternating COF (discoidal initialisation? - drawings C. Mathias).
Nucléus sur éclats en quartz des Hauts de Lestrade. A : exploitation Kombewa récurrente ; B : exploitation alternante partielle, initialisation d'un débitage Discoïde ? (dessins C. Mathias).

Figure 14. Cores on quartz flakes from Les Hauts de Lestrade. A: Kombewa recurrent; B: partially alternating COF (discoidal initialisation? - drawings C. Mathias). Nucléus sur éclats en quartz des Hauts de Lestrade. A : exploitation Kombewa récurrente ; B : exploitation alternante partielle, initialisation d'un débitage Discoïde ? (dessins C. Mathias).

41Flint chaînes opératoires attest to ramification processes. Seven of the nineteen identified cores are cores-on-flakes (Kombewa, dorsal exploitation), while two represent combined matrices (large notches that can be considered both as tools and cores), and two others derive from Levallois and Discoid flaking (No. = 11/19). The blanks selected as matrices are mostly cortical flakes of various sizes (fig. 15, 16). Four Kombewa-type flakes were identified. It should be noted that one of the Kombewa-type flakes presents the characteristics of a pseudo-Levallois point removed from the ventral surface of a flake. Such features have been identified in southwest France at Champs de Bossuet and Combe-Grenal, for example (Bourguignon and Turq 2003). A Discoid core on a flake also attests to the presence of this flaking mode at Les Hauts de Lestrade.

Figure 15. Flint core on a flake from Les Hauts de Lestrade, Unipolar Levallois (photographs C. Mathias).
Nucléus sur éclat en silex des Hauts de Lestrade, Levallois unipolaire (photographies C. Mathias).

Figure 15. Flint core on a flake from Les Hauts de Lestrade, Unipolar Levallois (photographs C. Mathias). Nucléus sur éclat en silex des Hauts de Lestrade, Levallois unipolaire (photographies C. Mathias).

Figure 16. Flint Discoid core on a flake, from Les Hauts de Lestrade (photographs C. Mathias).
Nucléus Discoïde bifacial sur éclat en silex des Hauts de Lestrade (photographies C. Mathias).

Figure 16. Flint Discoid core on a flake, from Les Hauts de Lestrade (photographs C. Mathias). Nucléus Discoïde bifacial sur éclat en silex des Hauts de Lestrade (photographies C. Mathias).

42At Les Hauts de Lestrade, Large Cutting Tools made of quartz (No. = 35) correspond mainly to unifacially shaped items, and, to a lesser extent, to unifaces, handaxes and picks. Most of the raw material volumes used for shaping were pebbles, although splits and flakes were also selected. One of the four handaxes was shaped on a flake. Two additional flint handaxes were identified, but neither was made on a flake or split.

Le Cassé

43The lithic assemblage from Le Cassé incorporates flaking and shaping concepts, with emphasis on the use of splits and especially flakes as matrices for debitage (No. of flakes = 110; 18.4 %). Simple flaking concepts (short unifacial sequences) as well as Discoid flaking (also mainly unifacial) were applied to these flakes (fig. 17). Splits were also used as matrices for debitage (No. = 36; 6%). Debitage methods for splits are simple (unifacial and S.S.D.A. knapping – i.e., simple debitage with a change of knapping surface on the core, “Système par Surfaces de Débitages Alternées”) and more rarely Discoid. Altogether, 24.4 % of the production matrices are flakes or splits. By contrast, no cores-on-flakes or splits were identified in flint.

Figure 17. Discoid cores on large quartzite flakes from Le Cassé unifacial modality (photographs C. Viallet).
Nucléus de conception Discoïde sur gros éclats de quartzite provenant du Cassé, modalité unifaciale (photographies C. Viallet).

Figure 17. Discoid cores on large quartzite flakes from Le Cassé unifacial modality (photographs C. Viallet). Nucléus de conception Discoïde sur gros éclats de quartzite provenant du Cassé, modalité unifaciale (photographies C. Viallet).

44Large Cutting Tools at Le Cassé are composed of unifacial and bifacial shaped tools, as well as unifaces, handaxes, cleavers and picks (No. = 81). Twenty-five LCTs are made on flake-blanks (fig. 18), representing 30.8 % of the total number of LCTs (including the eight cleavers). Ten other LCTs are made on splits (12.3 %).

Figure 18. Shaped quartzite tool on a large flake from Le Cassé. Transversal cutting-edge (photograph C. Viallet).
Outil façonné sur éclat du Cassé, à tranchant transversal (photographie C. Viallet).

Figure 18. Shaped quartzite tool on a large flake from Le Cassé. Transversal cutting-edge (photograph C. Viallet). Outil façonné sur éclat du Cassé, à tranchant transversal (photographie C. Viallet).

45The flake-matrices (for both debitage and shaping) appear to have been produced on cores exhibiting low productivity (fig. 19), bearing between one and three removals (“galets à enlèvements”; (Jarry, 2010)). They are produced with local raw materials at Le Cassé and represent 20% of production matrices (No = 122).

Figure 19. Large quartzite primary flake («entame») from Le Cassé (photographs C. Viallet).
Gros éclat d'entame en quartzite du Cassé (photographies C. Viallet).

Figure 19. Large quartzite primary flake («entame») from Le Cassé (photographs C. Viallet). Gros éclat d'entame en quartzite du Cassé (photographies C. Viallet).

46The use of flakes and splits at Le Cassé can thus be interpreted as an integrative strategy, rather than the result of ramification processes. It is therefore a means of initialising cobbles/pebbles, but also of shaping volumes with specific morpho-technical characteristics, as is the case for cleavers for example.

Mas des Caves I

47At Mas des Caves I, flint cobbles/pebbles are thus the main raw materials employed (cf. Costières du Gard). The initialisation of pebbles displays differential treatment. Flat pebbles/cobbles, or those with angles adapted to knapping, are initialised by direct freehand percussion (fig. 20), while more globular or angular shaped cobbles were frequently split longitudinally or transversally using the bipolar-on-anvil technique. Some examples even show knapping accidents during the process, related to irregularities in raw materials (fig. 20).

Figure 20. The two different options to open flint cobbles at le Mas des Caves I: splitting with an accident linked to an irregularity, bipolar on anvil (A) and freehand percussion (B), depending on the cobble morphology (photographs C. Mathias).
Les deux options utilisées pour entamer les galets de silex au Mas des Caves I : fendage du galet par percussion bipolaire sur enclume, avec un accident lié à une irrégularité de la matière première (A), et percussion à main levée (B), en fonction de la morphologie du galet (photographies C. Mathias).

Figure 20. The two different options to open flint cobbles at le Mas des Caves I: splitting with an accident linked to an irregularity, bipolar on anvil (A) and freehand percussion (B), depending on the cobble morphology (photographs C. Mathias). Les deux options utilisées pour entamer les galets de silex au Mas des Caves I : fendage du galet par percussion bipolaire sur enclume, avec un accident lié à une irrégularité de la matière première (A), et percussion à main levée (B), en fonction de la morphologie du galet (photographies C. Mathias).

48Among the 88 identified cores, 80 are in flint, five in quartz and three in quartzite. Almost all of the quartz cores are associated with bipolar on anvil percussion to obtain flakes (No = 4). Roughly a third of the flint cores, 30 %, are made on flakes (No. = 24), whereas the use of flint splits is marginal (No. = 3; 3.7 %).

49Refits associated with the production phase also indicate the use of splits. One of the refits is a core with hierarchical exploitation, associated with a fragment and a debris (accident during preparation). The blank is a split flint cobble. The striking platform is slightly prepared (plain striking flake platforms). The final product is very similar to a Kombewa-type flake (fig. 21), with the detachment plane parallel to the core surface. This case is not unique as four other refits illustrate the use of secondarily knapped splits, with short unipolar sequences or even Levallois-type and Levallois flaking modes (fig. 22).

Figure 21. Split secondary knapped on flint from le Mas des Caves I, two flakes are refitted (photographs C. Mathias).
Galet fendu en silex secondairement taillé provenant du Mas des Caves I, deux éclats sont remontés (photographies C. Mathias).

Figure 21. Split secondary knapped on flint from le Mas des Caves I, two flakes are refitted (photographs C. Mathias). Galet fendu en silex secondairement taillé provenant du Mas des Caves I, deux éclats sont remontés (photographies C. Mathias).

Figure 22. Refitting illustrating a Levallois core made on a split flint cobble from le Mas des Caves I (photograph C. Mathias).
Remontage montrant un nucléus Levallois sur galet fendu du Mas des Caves I (photographie C. Mathias).

Figure 22. Refitting illustrating a Levallois core made on a split flint cobble from le Mas des Caves I (photograph C. Mathias). Remontage montrant un nucléus Levallois sur galet fendu du Mas des Caves I (photographie C. Mathias).

50Ramification processes are clearly identified for flint productions, as 30 % of flint cores are flake matrices. This corresponds to specific cores-on-flakes, as well as Levallois methods applied to flakes (fig. 23). The use of Levallois flaking economises gestures and can be considered as a technical shortcut, especially in cobble/pebble-based assemblages.

Figure 23. Core-on-Flake from le Mas des Caves I (photograph C. Mathias).
Nucléus sur éclat du Mas des Caves I (photographie C. Mathias).

Figure 23. Core-on-Flake from le Mas des Caves I (photograph C. Mathias). Nucléus sur éclat du Mas des Caves I (photographie C. Mathias).

3 | Discussion

51The use of splits as blanks can be considered as an initialisation method, and as such is similar to the integrative strategy for flakes. The use of the bipolar-on-anvil technique in this context is related to the use of cobbles or pebbles as raw materials. Thus, the use of this technique is particularly effective for initialising cobbles/pebbles with specific morphologies (i.e., globular cobbles), where the percussion angle is not suitable for freehand percussion. At the same time, this technique produces particularly sought-after plano-convex matrices for Levallois/Discoid productions, but also for shaping Large Cutting Tools. In comparison to a flake, the potential useful volume is greater, and is conducive to several knapping cycles or the preparation of the striking platform, for example.

52At Saint-Sulpice, as at Brive-Laroche and Hauts de Lestrade, differences emerge in the processing of quartz pebbles, depending on their morphology. In this way, flat oval pebbles were initialised and knapped by direct freehand percussion or by bipolar percussion on an anvil in the centre of the cobble/pebble. On the other hand, thicker pebbles ‘with sides’ (“galets à pans”), or globular shapes were initialised by longitudinal bipolar percussion on an anvil (generating splits). Discoid debitage was thus frequently applied to quartz splits. This technical behaviour has been documented in many other contexts. For example, in Africa at Ounjougou (Mali), these strategies were used and documented by a detailed technological analysis with sets of refits (Soriano et al. 2010).

53Analysis of the Lunel-Viel lithic assemblages shows the use of bipolar percussion on an anvil to initialise the flint cobbles/pebbles for Levallois debitage. Once again, this type of initialisation is used in other flint cobble/pebble contexts, such as Thomas Quarry I – L1 in Morocco (Gallotti et al. 2020). It shows that anticipation skills are clearly established for debitage in early hominin communities around 1 My. At the Moroccan site, pebbles were split along the transversal axis and then exploited to produce bladelet-like tools. More recently, during the early Middle Palaeolithic in Italy, specific methods on flints have been described and identified as “Pontinian anvil flaking” (Soriano and Villa 2017). This involves the selection of hemi-pebbles for retouch in a second stage, as at Sedia del Diavolo and Monte delle Gioie during MIS 9/8, and ramified chaines opératoires have also been identified in the form of Kombewa exploitations (Soriano and Villa 2017).

54On the other hand, ramification (i.e., the selection of flakes from the main production systems for use as blanks) is not exclusively linked to raw material characteristics. Our analysis shows that the use of flakes and ramified strategies is not solely restricted to flint productions, although it is more widespread on these raw materials. Some quartz flakes were also secondarily knapped at Brive-Laroche, Bois de l’Hôpital Unit 3 or Les Hauts de Lestrade. In the same way, large quartzite flakes were used for Discoid productions at Cornebarrieu. Ramification is thus not linked to the type of raw material, and entails a genuine volumetric advantage, as shown by its use for local materials such as quartzite or quartz.

55For flint, the identified debitage ramification processes probably stem from different factors. In the Correzian lithic assemblages, they appear to be linked to an economy of raw materials, since flint outcrops are some thirty kilometers away from sites (and sometimes more) in Dordogne. Cortical flakes were probably part of the tool kits transported by Neanderthals, and some tools were recycled (see the secondarily knapped scraper at Brive-Laroche). However, the functional aspect should not be overlooked, as the morpho-technical characteristics of flakes produced on the ventral surface in particular may have been sought after for specific needs. At Mas des Caves I, the identified ramification occurs in a context rich in raw materials. The use of flakes in these series is mixed. One of the objectives is to produce flakes with specific attributes. However, in a Levallois context, it also serves as a shortcut in the initialisation of Levallois debitage. It is more efficient in this respect than the use of splits, since in this case, in addition to the plano-convex section, the convexities of the debitage surface are adapted to this type of sequence. In this case, we find a combination of constraints linked to the morphology of the raw material, associated with specific objectives and an economy of gestures. These examples illustrate that the use of flake-matrices offers technical solutions and flexibility, as has been shown for the different Mousterian LCTs (Bourguignon et al. 2004; Turq et al. 2013; Turq et al. 2017).

56In assemblages such as Le Cassé or Bois de l’Hôpital (in both Mousterian and Acheulean levels), large flakes are commonly used to produce various Large Cutting Tools, including cleavers. In the case of Bois de l’Hôpital, these large flake matrices are accompanied by splits. Splits are produced on less homogeneous quartz, for which the detachment of large flakes is complex. Here, the use of an anvil and the choice of a thick pebble to be fractured longitudinally, corresponds to the technical anticipation of the blank with a plano-convex cross-section and a sufficiently large volume of material to shape cutting edges by bifacial shaping. The split can therefore imitate the production of large flakes for lower-quality raw materials. The use of large flakes for shaping is common during the Early Palaeolithic and the transitional phase with the Middle Palaeolithic in Western Europe, until the end of the Middle Pleistocene (Mourre and Colonge 2010; Santonja et al. 2016; Méndez-Quintas et al. 2018, 2020). However, the use of flakes as blanks did not disappear in the Mousterian period. Handaxes made from smaller flakes are found in contexts where flint is abundant (Soressi 2002; Claud 2008). In addition, new radiometric dating of the Garonne series tends to show that this technical tradition continued throughout the Upper Pleistocene (some sites in this paper; route de Brial (Montbartier), ongoing study led by C. Viallet). The Vasconian is another very pertinent example of the use of large flakes in the recent Middle Palaeolithic (Deschamps 2014, 2017).

Conclusion

57Based on this brief overview, illustrated by several examples from the late Lower Palaeolithic and Middle Palaeolithic of southern France, we have highlighted the importance of an initial stage of blank (or matrix) production prior to the debitage/shaping phase. While this is particularly widespread for raw materials in the form of pebbles or cobbles (whatever the material; quartz, quartzite, flint), certain economic contingencies (distance from raw material deposits, for example) also come into play in this segmentation of chaînes opératoires.

58The two technical solutions using splits or flakes as blanks often constitute a form of initialisation and contribute to integrated technical and economic management. They sometimes involve specific production methods, as is the case for large quartzite flakes, or the longitudinal splitting of quartz associated with the shaping of Large Cutting Tools. Moreover, in the case of cleavers, these solutions fulfil a specific and anticipated morpho-technical objective. The ramification of chaînes opératoires is also attested, regardless of the type of raw material considered.

59One notable difference between split and flake blanks is that the useful volume of splits is greater, which facilitates certain technical operations or production concepts.

60Generally speaking, these strategies demonstrate an ability to anticipate, even for apparently simple production behaviours. From an economic point of view, the use of hemi-cobbles or pebbles maximises the useful volume of these blanks, and circumvents the considerable initialisation efforts required for direct knapping sequences.

Acknowledgements

61Many sites presented here were excavated during preventive archaeology operations and we would like to thank all the teams who participated in fieldwork. In particular, we would like to thank Pascal Tallet, excavation director at Bois de l’Hôpital and Le Cassé. The description and identification of the raw materials at these sites is the work of Paul Fernandes and Théo Minet, to whom we also express our thanks. We would like to thank Jean-Philip Brugal and Carla Giuliani for allowing us to study the atypical Lunel-Viel site, and for all the information they provided on the new excavations. We extend warm thanks to all the MNP team for their hospitality at les Eyzies during the study of the E. Bonifay collections. We are grateful to Laurence Bourguignon for her advice and for past discussions of the ramification concept. Many thanks also to Nathalie Fourment and Peggy Bonnet-Jacquement. Finally, we would like to thank the two reviewers whose comments have helped to improve this text, and Louise Byrne for the English editing.

Top of page

Bibliography

AGAM A., BARKAI R. 2018 - Small Flake Acheulian: Further Insights into Lithic Recycling at Late Acheulian Revadim, Israel. Tel Aviv, 45, 2, p. 170-192.

AGAM A., MARDER O., BARKAI R. 2015 - Small flake production and lithic recycling at Late Acheulian Revadim, Israel. Quaternary International, 361, p. 46-60.

ARRIGHI S., MARCIANI G., ROSSINI M., PEREIRA SANTOS M.C., FIORINI A., MARTINI I., AURELI D., BADINO F., BORTOLINI E., FIGUS C., LUGLI F., OXILIA G., ROMANDINI M., SILVESTRRINI S., RONCHITELLI A., MORONI A., BENAZZI S. 2020 - Between the hammerstone and the anvil: bipolar knapping and other percussive activities in the late Mousterian and the Uluzzian of Grotta di Castelcivita (Italy). Archaeological and Anthropological Sciences, 12, 11, p. 271.

ASHTON N. 1992 - The High Lodge flint industries. In: N. ASHTON, J. COOKE, S.J. LEWIS, J. ROSE (Éds.), High Lodge. Excavations by G. de Sieveking, 1962-8, and J. Cook, 1988. Bristish Museum Press, p. 124-168.

AURELI D., ROCCA R., LEMORINI C., MODESTI V., SCARAMUCCI S., MILLI S., GIACCIO B., MARANO F., PALOMBO M.R., CONTARDI A. 2016 - Mode 1 or mode 2? “Small tools” in the technical variability of the European Lower Palaeolithic: The site of Ficoncella (Tarquinia, Lazio, central Italy). Quaternary International, 393, p. 169-184.

BAENA PREYSLER J., TORRES NAVAS C., SHARON G. 2018 - Life history of a large flake biface. Quaternary Science Reviews, 190, p. 123-136.

BARSKY D., SALA R., MENENDEZ L., TORO-MOYANO I. 2015 - Use and re-use: Re-knapped flakes from the Mode 1 site of Fuente Nueva 3 (Orce, Andalucía, Spain). Quaternary International, 361, p. 21-33.

BEAUNE S. de (1997) - Les galets utilisés au Paléolithique supérieur. Approche archéologique et expérimentale. Gallia Préhistoire, 32e supplément à Gallia Préhistoire, 1, p. 3-298.

BERNARD-GUELLE S., TALLET P., AJAS A., BOUFFARD L., CHASSAN N., CHESNAUX L., FERNANDES P., GRIGGO C., BREUTZER S., MERCIER N., MINET T., MONIN G., RUÉ M., VIALLET C. 2019 - Saint-Sulpice-la-Pointe, Le Bois de l’Hôpital (Tarn). Rapport final d’opération, Villard-de-Lans, Paléotime.

BERNARD-GUELLE S., TALLET P., AJAS A., RUÉ M., FERNANDES P. 2016 - Une opportunité inédite d’étudier des occupations du Paléolithique moyen sur plusieurs hectares dans la vallée du Tarn : la fouille préventive du site du « Bois de l’Hôpital » à Saint-Sulpice (Tarn). Actualités Scientifiques du Bulletin de la Société préhistorique française, 113, 2, p. 375-377.

BOËDA É., KERZAVO B., MERCIER N., VALLADAS H. 1996 - Barbas C’3 base (Dordogne), une industrie bifaciale contemporaine des industries du Moustérien ancien ; une variabilité attendue. Quaternaria Nova, VI, p. 465-504.

BOËDA É., SORIANO S., NOËL-SORIANO S. 2004 - Fonction et fonctionnement d’un site à la fin du Pléistocène moyen. Le niveau acheuléen C’3 base de Barbas I (Creysse, Dordogne). In: Approches fonctionnelles en préhistoire. Actes du XXVe Congrès préhistorique de France, Nanterre, 24-26 novembre 2000, Paris. 

BONIFAY E. 1968 - Stratigraphie et industries lithiques de la grotte no 1 du Mas des Caves à Lunel-Viel (Hérault). In: La Préhistoire : Problèmes et tendances, p. 37-46.

BOURGUIGNON L., FAIVRE J.-P., TURQ A. 2004 - Ramification des chaînes opératoires : une spécificité du Moustérien ? Paleo 16, p. 37-48. 

BOURGUIGNON L., TURQ A. 2003 - Une chaîne opératoire de débitage discoïde sur éclat du moustérien à denticulés aquitain :  les exemples de Champs de Bossuet et de Combe-Grenal. In: M. PERESANI (dir.), Discoïd Lithic Technology. Advances and Implications, BAR International Series, 1120, p. 131-152.

BOUTILLIER-UZUNIDIS A. 2017 - Grands herbivores de la fin du Pléistocène moyen au début du Pléistocène supérieur dans le sud de la France : implications anthropologiques pour la lignée néandertalienne. Thèse de doctorat, Aix-Marseille.

BRUGAL J.-P., GIULIANI C., FOSSE P., FOURVEL J.-B., MAGNIEZ P., PELLETIER M., UZUNIDIS A. 2021 - Preliminary data on the Middle Pleistocene site of Lunel-Viel I (Hérault, France). Alpine and Mediterranean Quaternary, 34, 1, p. 75-88.

BYRNE F., PROFFITT T., ARROYO A., DE LA TORRE I. 2016 - A comparative analysis of bipolar and freehand experimental knapping products from Olduvai Gorge, Tanzania. Quaternary International, 424, p. 58-68.

CAPDEVIELLE J., COLONGE D. 2023 - Contributions and Limitations of a Technomorphometric Approach for Cleavers: The Case of Lanne-Darré (Hautes-Pyrénées). Lithic Technology, p. 1-26.

CLAUD É. 2008 - Le statut fonctionnel des bifaces au Paléolithique moyen récent dans le Sud-Ouest de la France. Étude tracéologique intégrée des outillages des sites de La Graulet, La Conne de Bergerac, Combe Brune 2, Fonseigner et Chez-Pinaud/Jonzac. Thèse de doctorat, Université de Bordeaux I, Bordeaux, 546 p.

CLIQUET D., LAUTRIDOU J.-P. 1988 - Le Moustérien à petits bifaces dominants de Saint-Julien de la Liègue (Eure). Revue archéologique de Picardie, 1, 1, p. 175-185.

COLLINA-GIRARD J. 1997 - Les outillages naturels sommaires sur supports naturels tenaces (quartz et quartzites). Technomorphologie et évolution psychique. Préhistoire Anthropologie Méditerannéennes, 6, p. 211-226.

COLLINA-GIRARD J., TURQ A. 1991 - Le Paléolithique moyen sur galets de la station des Planes, commune de Montayral (Lot-et-Garonne). Paleo 3, p. 49-74.

DESCHAMPS M. 2011 - Le Vasconien : révision de sa signification à partir des industries lithiques d’Olha I et II, d’Isturitz et de Gatzarria. Paleo 21, p. 103-126.

DESCHAMPS M. 2014 - La diversité culturelle au Paléolitique moyen récent : le Vasconien et sa signification au sein des faciès moustériens. Thèse de doctorat, Université Toulouse le Mirail - Toulouse II, Toulouse, 589 p.

DESCHAMPS M. 2017 - Late Middle Palaeolithic assemblages with flake cleavers in the western Pyrenees: The Vasconian reconsidered. Quaternary International, 433, p. 33-49.

DIEZ-MARTIN F., YUSTOS P.S., DOMINGUEZ-RODRIGO M., PRENDERGAST M.E. 2011 - An Experimental Study of Bipolar and Freehand Knapping of Naibor Soit Quartz from Olduvai Gorge (Tanzania). American Antiquity, 76, 4, p. 690-708.

DONNART K., NAUDINOT N., LE CLEZIOT L. 2009 - Approche expérimentale du débitage bipolaire sur enclume : caractérisation des produits et analyse des outils de production. Bulletin de la Société préhistorique française, 106, 3, p. 517-533. 

FAIVRE J.-P. 2008 - Organisation techno-économique des systèmes de production dans le Paléolithique moyen récent du Nord-est Aquitain : Combe-Grenal et les Fieux. Thèse de doctorat, Bordeaux 1, Bordeaux.

FALGUERES C., LAHAYE C., TOMBRET O., GARBE L., LEBRUN B., BAHAIN J.-J., FREREBEAU N., GIULIANI C., BRUGAL J.-P. 2024 - ESR/U-series and pIR-IR290 dating of the Middle Pleistocene site of Lunel-Viel (LV I), Hérault, Southern France. Quaternary Geochronology, 81, p. 101516.

FORESTIER H. 1993 - Le Clactonien : mise en application d’une nouvelle méthode de débitage s’inscrivant dans la variabilité des systèmes de production lithique du Paléolithique ancien. Paleo 5, p. 53-82. http://doi.org/10.3406/pal.1993.1104

FOSSE P. 1996 - La grotte n° 1 de Lunel-Viel (Hérault, France) : Repaire d’hyènes du Pléistocène moyen. Etude taphonomique du matériel osseux. Paleo 8, p. 47-79.

GALLOTTI R., MOHIB A., FERNANDES P., EL GRAOUI M., LEFEVRE D., RAYNAL J.-P. 2020 - Dedicated core-on-anvil production of bladelet-like flakes in the Acheulean at Thomas Quarry I - L1 (Casablanca, Morocco). Scientific Reports, 10, 1, p. 9225.

GARCIA-VADILLO F.-J., CANALS-SALOMO A., RODRIGUEZ-ALVAREZ X.-P., CARBONELL E. 2022 - The large flake Acheulean with spheroids from Santa Ana Cave (Cáceres, Spain). Journal of Archaeological Science: Reports, 41, p. 103265.

GREGOIRE S., BAZILE F. 2005 - La diffusion du silex des Costières du Gard au Paléolithique supérieur. Comptes Rendus Palevol, 4, 5, p. 413-419.

HORTA P., CASCALHEIRA J., BICHO N. 2019 - The Role of Lithic Bipolar Technology in Western Iberia’s Upper Paleolithic: The Case of Vale Boi (Southern Portugal). Journal of Paleolithic Archaeology, 2, 2, p. 134-159.

INIZAN M.-L., REDURON M., ROCHE H., TIXIER J. 1995 -Technologie de la pierre taillée. Meudon, CREP (Préhistoire de la pierre taillée 4), 199 p.

JARRY M. 2010 - Les groupes humains du Pléistocène moyen et supérieur en Midi toulousain : contextes, ressources et comportements entre Massif Central et Pyrénées. Thèse de doctorat, Université de Toulouse Le Mirail - Toulouse II, Toulouse, 470 p.

JAUBERT J. 2001 - Industries à outils bifaciaux du Paléolithique moyen entre Massif central et Pyrénées. In: Les industries à outils bifaciaux du Paléolithique moyen d’Europe occidentale, Liège (ERAUL 98), p. 151-161.

LE GRAND Y. 1993 - Approche méthodologique et technologique d’un site d’habitat du Pléistocène moyen : la grotte n° 1 du Mas des Caves (Lunel-Viel, Hérault). Thèse de doctorat, Université Aix Marseille.

LEBEGUE F. 2012 - Le Paléolithique moyen récent entre Rhône et Pyrénées : approche de l’organisation techno-économique des productions lithiques, schémas de mobilité et organisation du territoire (Les Canalettes, L’Hortus, Bize-Tournal, La Crouzade et La Roquette II). Thèse de doctorat, Perpignan Via-Domitia, Perpignan, 797 p.

LOMBRERA-HERMIDA A. de, RODRIGUEZ-ALVAREZ X.P., PENA L., SALA-RAMOS R., DESPRIEE J., MONCEL M.-H., GOURCIMAULT G., VOINCHET P., FALGUERES C. 2016 - The lithic assemblage from Pont-de-Lavaud (Indre, France) and the role of the bipolar-on-anvil technique in the Lower and Early Middle Pleistocene technology. Journal of Anthropological Archaeology, 41, p. 159-184.

MATHIAS C. 2018 - Les phases anciennes du Paléolithique moyen dans le Sud-Est et le Sud-Ouest de la France : étude des systèmes techniques lithiques. Thèse de doctorat, Université de Perpignan Via-Domitia, Perpignan, 730 p.

MATHIAS C., BOURGUIGNON L. 2020 - Cores-on-flakes and ramification during the middle palaeolithic in southern France: A gradual process from the early to late middle palaeolithic? Journal of Archaeological Science: Reports, 31, p. 102336.

MATHIAS C., VIALLET C., DELVIGNE V., FERNANDES P., GAUVRIT-ROUX E., LAHAYE C., LEBRUN B., RAYNAL J.-P., RUE M., TALLET P. 2023 - Le Paléolithique moyen des sites de Brive-Laroche-Aérodrome (Corrèze, France) : choix techno-économiques entre Périgord et Massif central. Bulletin de la Société préhistorique française, 120, 4, p. 571-602.

MENDEZ-QUINTAS E., SANTONJA M., ARNOLD L.J., CUNHA-RIBEIRO J.P., DA SILVA P.X., DEMURO M., DUVAL M., GOMES A., MEIREILES J., MONTEIRO-RODRIGUES S., PEREZ-GONZALEZ A. 2020 - The Acheulean Technocomplex of the Iberian Atlantic Margin as an Example of Technology Continuity Through the Middle Pleistocene. Journal of Paleolithic Archaeology, 3, 4, p. 918-943.

MENDEZ-QUINTAS E., SANTONJA M., PEREZ-GONZALEZ A., DUVAL M., DEMURO M., ARNOLD L.J. 2018 - First evidence of an extensive Acheulean large cutting tool accumulation in Europe from Porto Maior (Galicia, Spain). Scientific Reports, 8, 1, p. 3082.

MOURRE V. 2003 - Implications culturelles de la technologie des hachereaux. Thèse de doctorat, Université Paris Nanterre, France, 3 tomes, 299 p., 257 p. et 327 p.

MOURRE V., COLONGE D. 2010 - La question du débitage de grands éclats à l’Acheuléen. Paleo, Numéro spécial, p. 35-48.

MOURRE V., JARRY M. 2010 - Avant-propos. Paleo, Numéro spécial, p. 9-11.

MOURRE V., JARRY M., COLONGE D., LELOUVIER L.-A. 2010 - Le débitage sur enclume aux Bosses (Lamagdelaine, Lot, France). Paleo, Numéro spécial, p. 49-62.

NICOUD É. 2010 - Les chaînes opératoires sur galets en roches volcaniques et quartz dans l’industrie lithique du gisement Moustérien de Champ Grand (Saint-Maurice-sur-Loire, Loire). Paleo, Numéro spécial, p. 107-122.

PARUSH Y., ASSAF E., SLON V., GOPHER A., BARKAI R. 2015 - Looking for sharp edges: Modes of flint recycling at Middle Pleistocene Qesem Cave, Israel. Quaternary International, 361, p. 61-87.

PEŇA P. (DE LA) 2015 - The Interpretation of Bipolar Knapping in African Stone Age Studies. Current Anthropology, 56, 6, p. 911-923.

PEREZ-BALAREZO A., GUIBERT J. 2023 - Quand la pierre fait l’outil : (Pré)histoires de galets. Techniques & Culture, 79, 1, p. 56-75.

RIOS-GARAIZAR J., EIXEA A., VILLAVERDE V. 2015 - Ramification of lithic production and the search of small tools in Iberian Peninsula Middle Paleolithic. Quaternary International, 361, p. 188-199.

ROMAGNOLI F., GOMEZ DE SOLER B., BARGALLO A., CHACON M.G., VAQUERO M. 2018 - Here and now or a previously planned strategy? Rethinking the concept of ramification for micro-production in expedient contexts: Implications for Neanderthal socio-economic behaviour. Quaternary International, 474, p. 168-181.

RUBIO-JARA S., PANERA J., RODRIGUEZ-DE-TEMBLEQUE J., SANTONJA M., PEREZ-GONZALEZ A. 2016 - Large flake Acheulean in the middle of Tagus basin (Spain): Middle stretch of the river Tagus valley and lower stretches of the rivers Jarama and Manzanares valleys. Quaternary International, 411, p. 349-366.

SANTONJA M., PEREZ-GONZALEZ A., PANERA J., RUBIO-JARA S., MENDEZ-QUINTAS E. 2016 - The coexistence of Acheulean and Ancient Middle Palaeolithic techno-complexes in the Middle Pleistocene of the Iberian Peninsula. Quaternary International, 411, p. 367-377.

SHARON G. 2008 - The impact of raw material on Acheulian large flake production. Journal of Archaeological Science, 35, 5, p. 1329 -1344.

SHARON G. 2010 - Large flake Acheulian. Quaternary International, 223-224, p. 226-233.

SORESSI M. 2002 - Le Moustérien de tradition acheuléenne du sud-ouest de la France : discussion sur la signification du faciès à partir de l’étude comparée de quatre sites : Pech-de-l’Azé I, Le Moustier, La Rochette et la Grotte XVI. Thèse de doctorat, Bordeaux 1, Bordeaux.

SORIANO S., ROBERT A., HUYSECOM É. 2010 - Percussion bipolaire sur enclume : choix ou contrainte ? L’exemple du Paléolithique d’Ounjougou (Pays dogon, Mali). Paleo, Numéro spécial, p. 123-132.

SORIANO S., VILLA P. 2017 - Early Levallois and the beginning of the Middle Paleolithic in central Italy. PLOS ONE, 12, 10, p. e0186082.

TALLET P., VIALLET C., FERNANDES P., RUE M. 2022 - Entre Paléolithique inférieur et moyen : le site du Cassé à Cornebarrieu. Rapport final d’opération, Villard-de-Lans, Paléotime.

TAVOSO A. 1975 - Les hachereaux sur éclats de l’Acheuléen montalbanais. Quartär 26, p. 13-31.

TIXIER J. 1956 - Les Hacheraux dans L’Acheuléen Nord-Africain. Notes Typologiques, p. 15-22.

TURQ A. 2000 - Le Moustérien de tradition acheuléenne. Paleo 2, p. 244-273.

TURQ A., FAIVRE J.-P., GRAVINA B., BOURGUIGNON L. 2017 - Building models of Neanderthal territories from raw material transports in the Aquitaine Basin (southwestern France). Quaternary International, 433, p. 88-101.

TURQ A., ROEBROEKS W., BOURGUIGNON L., FAIVRE J.-P. 2013 - The fragmented character of Middle Palaeolithic stone tool technology. Journal of Human Evolution, 65, 5, p. 641-655.

VAISSIE E. 2021 - Géographie culturelle du Paléolithique moyen récent dans le Massif central et ses marges : territoires, mobilités et systèmes techniques lithiques. Thèse de doctorat, Université de Bordeaux I, Bordeaux, France, 1161 p.

VAQUERO M., BARGALLO A., CHACON M.G., ROMAGNOLI F., SANUDO P. 2015 - Lithic recycling in a Middle Paleolithic expedient context: Evidence from the Abric Romaní (Capellades, Spain). Quaternary International, 361, p. 212-228.

VENDITTI F., NUNZIANTE-CESARO S., PARUSH Y., GOPHER A., BARKAI R. 2019 - Recycling for a purpose in the late Lower Paleolithic Levant: Use-wear and residue analyses of small sharp flint items indicate a planned and integrated subsistence behavior at Qesem Cave (Israel). Journal of Human Evolution, 131, p. 109-128.

VIALLET C., AJAS-PLANTEY A., BERNARD-GUELLE S., DELVIGNE V., FERNANDES P., GAUVRIT-ROUX E., LAHAYE C., LEBRUN B., MATHIAS C., MONIN G., NAVENNEC G., PLATEL J.-P., PIBOULE M., RAYNAL J.-P., ROBBE J.,RUE M., TALLET P., TURQ A. 2022b -  Une nouvelle occurrence du Paléolithique moyen en Corrèze : les occupations du site de Brive-Laroche aérodrome (Brive-la-Gaillarde/Saint-Pantaléon-de-Larche). Rapport final d’opération, Villard-de-Lans, Paléotime.

VIALLET C., BURCET T., CONTE P., DELVIGNE V., FERNANDES P., GROSFILLEY S., LAHAYE C., MATHIAD C., RUE M. 2023 - Les Hauts de Lestrade : de nouveaux vestiges du Paléolithique moyen en Corrèze (Saint-Pantaléon-de-Larche). Rapport final d’opération, Villard-de-Lans, Paléotime.

VIALLET C., FERNANDES P., LAHAYE C., LEBRUN B., RUE M., TALLET P. 2024 - Une nouvelle occurrence de l’Acheuléen pyrénéo-garonnais et la question de la régionalisation des productions lithiques à la fin du Pléistocène moyen : le site du Cassé à Cornebarrieu (Haute-Garonne). L’Anthropologie, 128, 1, p. 103236. 

VIALLET C., MINET T., FERANDES P., RUE M., AJAS A., TALLET P., BERNARD-GUELLE S. 2022a - L’outillage façonné utilisé en percussion lancée du site paléolithique moyen du Bois de l’Hôpital (Saint-Sulpice-la-Pointe, Tarn). Comptes Rendus Palevol, 21, 5, p. 123-143.

WILSON C.G., CARUANA M.V., BLACKWOOD A.F., ARNOLD L.J., HERRIES A.I.R. 2024 - Why large Flakes? Later Acheulian handaxe manufacture at Amanzi Springs, Area 2 (Eastern Cape, South Africa). Journal of Archaeological Science: Reports, 53, p. 104393.

WOJTCZAK D. 2015 - Cores on flakes and bladelet production, a question of recycling? The perspective from the Hummalian industry of Hummal, Central Syria. Quaternary International, 361, p. 155-177.

YESILOVA G.C., ARROYO A., VERGES J.M., OLLE A. 2024 - New Approaches to the Bipolar Flaking Technique: Qualitative, Quantitative, and Kinematic Perspectives. Journal of Archaeological Method and Theory, 31, 3, p. 1333-1382.

Top of page

List of illustrations

Title Figure 1. Schematic representation of the technical processes discussed in the article. A: Split cobbles along different axes, then used as debitage or shaping matrices (blanks); B: Fracturing on an anvil in the central part of the cobble, with the possibility of using the fragments as tools (retouching) or matrices for debitage; C: Core-on-Flake, use of flakes as matrices for debitage or shaping (drawing C. Mathias). Représentation schématique des procédés techniques discutés dans l'article. A : Galets fendus selon différents axes, puis utilisés comme matrices de débitage ou de façonnage (supports) ; B : Fracturation sur enclume dans la partie centrale du galet, avec possibilité d'utiliser les fragments comme outils (par la retouche) ou matrices de débitage ; C : Nucléus sur éclats, utilisation des éclats comme matrices de débitage ou de façonnage (dessin C. Mathias).
URL http://journals.openedition.org/paleo/docannexe/image/10157/img-1.jpg
File image/jpeg, 220k
Title Figure 2. Map showing the location of the studied sites in southern France (made using QGIS 3.16.16, data ESRI - C. Mathias). Carte montrant la localisation des sites étudiés dans le sud de la France (réalisée avec QGIS 3.16.16, données ESRI - C. Mathias).
URL http://journals.openedition.org/paleo/docannexe/image/10157/img-2.jpg
File image/jpeg, 668k
Title Table 1. Main characteristics of the selected sites. *As the site is currently excavated, this count is not definitive. Principales caractéristiques des sites sélectionnés. *Le site étant en cours de fouille, ce décompte n'est pas définitif.
URL http://journals.openedition.org/paleo/docannexe/image/10157/img-3.png
File image/png, 105k
Title Figure 3. Large Cutting Tools in quartz and quartzite from the Acheulean level of Bois de l'Hôpital. A-B: handaxes made on splits; C: handaxe made on a large flake; a shaping flake is refitted at the base of the ventral surface (drawings R. Picavet). Grands outils façonnés en quartz et quartzite du niveau acheuléen de Bois de l'Hôpital. A-B : bifaces sur galets fendus; C: biface sur grand éclat ; un éclat de façonnage remonte sur la base de la surface ventrale (dessins R. Picavet).
URL http://journals.openedition.org/paleo/docannexe/image/10157/img-4.jpg
File image/jpeg, 716k
Title Figure 4. Refit of a split cobble, both parts of which were selected for secondary knapping (unifacial discoid intent?) from level 3 of Bois de l’Hôpital (photograph and CAD S. Bernard-Guelle). Remontage d'un galet fendu, dont les deux parties ont été sélectionnées pour un débitage secondaire (intention d'une production discoïde unifaciale ?) provenant du niveau 3 du Bois de l'Hôpital (photographie et DAO S. Bernard-Guelle).
URL http://journals.openedition.org/paleo/docannexe/image/10157/img-5.jpg
File image/jpeg, 916k
Title Figure 5. Refit of a large ovoid quartz cobble, fractured by bipolar percussion on an anvil in its centre (photograph Paléotime). Remontage d'un gros galet de quartz ovoïde, fracturé par percussion bipolaire sur enclume en son centre (photographie Paléotime).
URL http://journals.openedition.org/paleo/docannexe/image/10157/img-6.jpg
File image/jpeg, 740k
Title Figure 6. Split cobble, one fragment of which was secondarily knapped (photograph Paléotime). Galet fendu, dont un fragment a été secondairement taillé (photographie Paléotime).
URL http://journals.openedition.org/paleo/docannexe/image/10157/img-7.jpg
File image/jpeg, 532k
Title Figure 7. LCTs on flakes from Bois de l’Hôpital, level 3 (photograph C. Viallet). Grandes pièces façonnées sur éclat du Bois de l'Hôpital, niveau 3 (photographie C. Viallet).
URL http://journals.openedition.org/paleo/docannexe/image/10157/img-8.jpg
File image/jpeg, 608k
Title Figure 8. Ramification process as part of discoidal debitage, flint cobble (photograph Paléotime). Processus de ramification dans le cadre d'un débitage Discoïde, galet de silex (photographie Paléotime)
URL http://journals.openedition.org/paleo/docannexe/image/10157/img-9.jpg
File image/jpeg, 724k
Title Figure 9. Two cores on splits from Brive-Laroche North (photograph C. Mathias). Deux nucléus sur galets fendus de Brive-Laroche Nord (photographie C. Mathias).
URL http://journals.openedition.org/paleo/docannexe/image/10157/img-10.jpg
File image/jpeg, 296k
Title Figure 10. Refitted split cobble with a globular shape from Brive-Laroche North. One of the splits bears some removals (photograph C. Mathias). Remontage d'un galet fendu de forme globulaire provenant de Brive-Laroche Nord. Un des hémi-galets présente des négatifs d'enlèvements (phototographie C. Mathias).
URL http://journals.openedition.org/paleo/docannexe/image/10157/img-11.jpg
File image/jpeg, 324k
Title Figure 11. Secondarily knapped quartz flake from Brive-Laroche South. The ventral surface is flaked by a large removal, made after the exploitation of the thickness (proximal bipolar-on-anvil exploitation - photograph C. Mathias). Éclat de quartz secondairement débité, provenant de Brive-Laroche Sud. La surface ventrale présente un grand négatif d'enlèvement, réalisé après une première exploitation de l'épaisseur par percussion bipolaire sur enclume en partie proximale de l'éclat-matrice (photographie C. Mathias).
URL http://journals.openedition.org/paleo/docannexe/image/10157/img-12.jpg
File image/jpeg, 76k
Title Figure 12. Two discoidal quartz cores probably made on splits from Brive-Laroche South, unifacial mode (photograph C. Mathias). Deux nucléus de conception Discoïde unifaciale, probablement réalisés sur deux galets fendus provenant de Brive-Laroche Sud (photographie C. Mathias).
URL http://journals.openedition.org/paleo/docannexe/image/10157/img-13.jpg
File image/jpeg, 272k
Title Figure 13. Ramification processes on flint at Brive-Laroche South. A: refit of a broken core-on-flake, ventral unipolar convergent debitage from the distal part; B: Kombewa recurrent unipolar exploitation from the proximal part; C: Kombewa lateral flake; D: Kombewa lineal flaking and dorsal unipolar exploitation (after the truncation; photograph C. Mathias). Processus de ramification identifiés sur silex à Brive-Laroche Sud. A : remontage d'un nucléus sur éclat fracturé (raccord), débitage unipolaire convergent sur la face inférieure depuis la partie distale ; B : nucléus sur éclat Kombewa récurrent, exploitation unipolaire depuis la partie proximale ; C : éclat Kombewa latéral ; D : nucléus Kombewa linéal et exploitation de la face supérieure après aménagement d'une troncature (photographie C. Mathias).
URL http://journals.openedition.org/paleo/docannexe/image/10157/img-14.jpg
File image/jpeg, 816k
Title Figure 14. Cores on quartz flakes from Les Hauts de Lestrade. A: Kombewa recurrent; B: partially alternating COF (discoidal initialisation? - drawings C. Mathias). Nucléus sur éclats en quartz des Hauts de Lestrade. A : exploitation Kombewa récurrente ; B : exploitation alternante partielle, initialisation d'un débitage Discoïde ? (dessins C. Mathias).
URL http://journals.openedition.org/paleo/docannexe/image/10157/img-15.jpg
File image/jpeg, 276k
Title Figure 15. Flint core on a flake from Les Hauts de Lestrade, Unipolar Levallois (photographs C. Mathias). Nucléus sur éclat en silex des Hauts de Lestrade, Levallois unipolaire (photographies C. Mathias).
URL http://journals.openedition.org/paleo/docannexe/image/10157/img-16.jpg
File image/jpeg, 256k
Title Figure 16. Flint Discoid core on a flake, from Les Hauts de Lestrade (photographs C. Mathias). Nucléus Discoïde bifacial sur éclat en silex des Hauts de Lestrade (photographies C. Mathias).
URL http://journals.openedition.org/paleo/docannexe/image/10157/img-17.jpg
File image/jpeg, 112k
Title Figure 17. Discoid cores on large quartzite flakes from Le Cassé unifacial modality (photographs C. Viallet). Nucléus de conception Discoïde sur gros éclats de quartzite provenant du Cassé, modalité unifaciale (photographies C. Viallet).
URL http://journals.openedition.org/paleo/docannexe/image/10157/img-18.jpg
File image/jpeg, 212k
Title Figure 18. Shaped quartzite tool on a large flake from Le Cassé. Transversal cutting-edge (photograph C. Viallet). Outil façonné sur éclat du Cassé, à tranchant transversal (photographie C. Viallet).
URL http://journals.openedition.org/paleo/docannexe/image/10157/img-19.jpg
File image/jpeg, 280k
Title Figure 19. Large quartzite primary flake («entame») from Le Cassé (photographs C. Viallet). Gros éclat d'entame en quartzite du Cassé (photographies C. Viallet).
URL http://journals.openedition.org/paleo/docannexe/image/10157/img-20.jpg
File image/jpeg, 264k
Title Figure 20. The two different options to open flint cobbles at le Mas des Caves I: splitting with an accident linked to an irregularity, bipolar on anvil (A) and freehand percussion (B), depending on the cobble morphology (photographs C. Mathias). Les deux options utilisées pour entamer les galets de silex au Mas des Caves I : fendage du galet par percussion bipolaire sur enclume, avec un accident lié à une irrégularité de la matière première (A), et percussion à main levée (B), en fonction de la morphologie du galet (photographies C. Mathias).
URL http://journals.openedition.org/paleo/docannexe/image/10157/img-21.jpg
File image/jpeg, 328k
Title Figure 21. Split secondary knapped on flint from le Mas des Caves I, two flakes are refitted (photographs C. Mathias). Galet fendu en silex secondairement taillé provenant du Mas des Caves I, deux éclats sont remontés (photographies C. Mathias).
URL http://journals.openedition.org/paleo/docannexe/image/10157/img-22.jpg
File image/jpeg, 388k
Title Figure 22. Refitting illustrating a Levallois core made on a split flint cobble from le Mas des Caves I (photograph C. Mathias). Remontage montrant un nucléus Levallois sur galet fendu du Mas des Caves I (photographie C. Mathias).
URL http://journals.openedition.org/paleo/docannexe/image/10157/img-23.jpg
File image/jpeg, 308k
Title Figure 23. Core-on-Flake from le Mas des Caves I (photograph C. Mathias). Nucléus sur éclat du Mas des Caves I (photographie C. Mathias).
URL http://journals.openedition.org/paleo/docannexe/image/10157/img-24.jpg
File image/jpeg, 282k
Top of page

References

Bibliographical reference

Cyrielle Mathias, Sébastien Bernard-Guelle and Cyril Viallet, “The use of splits and flakes as blanks for debitage or shaping: examples from the late Lower and Middle Palaeolithic of Southern France”PALEO, 34 | 2025, 200-233.

Electronic reference

Cyrielle Mathias, Sébastien Bernard-Guelle and Cyril Viallet, “The use of splits and flakes as blanks for debitage or shaping: examples from the late Lower and Middle Palaeolithic of Southern France”PALEO [Online], 34 | 2024, Online since 01 September 2025, connection on 10 December 2025. URL: http://journals.openedition.org/paleo/10157; DOI: https://doi.org/10.4000/1516s

Top of page

About the authors

Cyrielle Mathias

UPVD, UMR 7194 - HNHP; défi-clé Sciences du Passé en Occitanie ; Centre de Recherche français à Jérusalem.

By this author

Sébastien Bernard-Guelle

Paléotime, UMR 6972 - LAMPEA.

By this author

Cyril Viallet

Paléotime, UMR 7194 - HNHP.

By this author

Top of page

Copyright

CC-BY-NC-ND-4.0

The text only may be used under licence CC BY-NC-ND 4.0. All other elements (illustrations, imported files) may be subject to specific use terms.

Top of page
Search OpenEdition Search

You will be redirected to OpenEdition Search