Artisanats et territoires des chasseurs moustériens de Champ Grand Ludovic Slimak (dir.) Artisanats et territoires No. 1. Aix-en-Provence: a&t Editions, 2008, 432 pp. (paperback), £50.00. ISBN-13: 9782952958707.

Reviewed by ALEXANDRE STEENHUYSE Department of Evolutionary Anthropology, P.O. Box 90383, Science Drive, Duke University, Durham, NC 27708-383, USA; [email protected]

rtisanats et territoires des chasseurs mousteriens de Champ considered so self-evident that the general lack of biblio- AGrand is best described as a site monograph summa- graphical references and methodological justifications may rizing the results of a series of analyses conducted on ar- be a major obstacle for readers not familiar with French chaeological samples collected at the site of Champ Grand Paleolithic studies or structuralist anthropology (e.g., Le- (France) during the excavation directed by A. Popier from roi-Gourhan 1943, 1945; Mauss 1936). The reader also may 1968 to 1983. Divided into 15 chapters, all in French, this regret a similar bibliographical aridity concerning key ar- monograph is the result of the collaborative work of a team chaeological concepts such as mobility, raw material maxi- of specialists, each focusing on a specific aspect of the col- mization, or tool design. This may in fact reflect the over- lected material. Champ Grand is a late Mousterian open- all narrow theoretical focus of the book. Similarly, many air site located in the central Massif Central mountain re- interpretive conclusions rely upon complex and multi-se- gion of France in a section of the Loire Valley known for mantic concepts (“culture,” “cultural traditions,” or “social its relatively high density of Paleolithic sites, including the representations”) which are never clearly defined. As a re- Magdalenian site of Rocher de la Caille and Goutte Rouf- sult, this book reflects both the strengths and weaknesses fat and Mousterian, Gravettian, and Magdalenian occupa- of current French Paleolithic archaeology. For the wealth tions at the site of Vigne Brun. Two distinct Mousterian lay- of information it contains, this monograph is both an im- ers separated by sterile sediments were initially identified portant contribution to our current understanding of the by the excavators (see Chapter 1 by J. Combier), although late Middle Paleolithic of Western Europe and an oppor- only one single, thick Mousterian layer was identified in tunity to critically revisit some fundamental aspects of the the lower section of the site. Following the excavation of theoretical and methodological foundations of Paleolithic large contiguous areas, Popier and his colleagues hypoth- archaeology. esized that the preferential distribution of certain artifact The first two chapters highlight the history of research types and concentrations of charcoal and faunal material at the site (Chapter 1 by J. Combier) and the overall meth- might reflect the spatial organization of certain activities odology used during the excavation and the choices made within the site. This monograph is an opportunity to test for the analysis of the archaeological material (Chapter 2 by some of those hypotheses and to conduct an initial analy- L. Slimak). Initiated in 1979, the systematic excavation of sis of a sample of the lithic collection composed of nearly large portions of the site was organized as a salvage opera- 90,000 lithic artifacts (including pebbles). Although struc- tion aimed at collecting as much material and contextual tured as a typical site monograph, this book also contains information as possible before the site was submerged due methodological reflections regarding some aspects of lithic to the construction of a dam on the Loire River in 1983. As analysis and our current understanding of late Mousterian a result, the Popier collection is composed of an extensive technological strategies. This review will focus on the re- lithic artifact assemblage (n=89,330 lithic artifacts including sults of those studies and on a critical assessment of the pebbles and non-siliceous artifacts). The analysis published theoretical and methodological foundations supporting in this monograph was conducted on a sample represent- the analysis and interpretation of the archaeological data. ing about 10% of this initial collection (total of 1,622 arti- This monograph is the first volume of a new collection facts and 3,052 “micro-éclats et débris”; p. 37). Although titled Artistanats et territoires directed by L. Slimak. As em- described as “random” (p. 37), the sampling method is not phasized by J. Jaubert in the preface, the overall aesthetic clearly described while the relative proportion of the core quality of the book is remarkable. It includes color pictures component in the studied sample is slightly higher than of artifacts, technical drawings, and effective visual aids that of the original collection. Those two chapters are fol- (diagrams and schemas) that all effectively support a rich lowed by a detailed analysis of the site deposits aimed at and detailed text. The preface (Jaubert) and the following identifying post-depositional processes and reconstruct- introduction (Geneste), whose purpose is essentially lau- ing the chronology of the site deposition (Chapter 3 by B. datory, provide the reader with an appreciation for the Kervazo, C. Duchadeau-Kervazo and N. Maumont). Most theoretical undertone of the book. Written in French, and of the site deposits did not suffer from major post-deposi- likely intended for a French audience, its argumentation tional processes, giving credence to the following chapters relies upon an implicit interpretive framework. The theo- focusing on the spatial organization of activities within the retical assumptions supporting many conclusions often are site’s boundaries based on the analysis of the spatial distri-

PaleoAnthropology 2011: 78−82. © 2011 PaleoAnthropology Society. All rights reserved. ISSN 1545-0031 doi:10.4207/PA.2011.REV101 BOOK REVIEW • 79 bution of certain artifact types (Chapter 4 by A. Popier and cal surfaces. As a result, Slimak is able to demonstrate that J. Combier; Chapter 5 by P. Yvorra). Yvorra conducted a most of the blank production methods (Levallois, discoid, statistical analysis (k-means clustering) of the spatial distri- and Kombewa) were geared toward the production of bution of lithic artifacts located in and near archaeological flakes with normalized, similar morphologies interpreted features previously identified by the excavators as possible here as reflecting cultural norms. The blade and bladelet habitation structures. This analysis supports some of those component is particularly interesting as prismatic blade observations and confirms that some activities, includ- production has long been considered a defining criterion of ing stone tool blank production, took place outside of the Upper Paleolithic lithic industries. According to Slimak’s structures interpreted as possible huts. The analysis of the demonstration, the comparison of the general dimensions manufacturing processes of the pebble component (Chap- of the blades and bladelets suggests that those two compo- ter 6 by E. Nicoud) and the use-wear analysis of a selected nents were manufactured following two distinct sequences sample of those pebbles (Chapter 7 by L. Roux) do not pro- (see also Slimak 1999). Slimak’s detailed descriptions of vide additional details regarding those activity areas. They selected blade cores also suggest that the blade produc- do, however, demonstrate the extensive use of pebbles as tion method at Champ Grand did not involve the careful both tools and flake cores. The scope of the faunal analysis planning and maintenance of the core often described for (Chapter 8 by E. Crégut-Bonnoure) was greatly reduced by Upper Paleolithic prismatic blade cores. Turning to the dis- poor preservation conditions due to the acidity of the site coid cores, Slimak’s lengthy discussion is merely intended sediments. Composed of 44 teeth and tooth fragments, and to support his own re-definition of the “Discoid concept.” 4 identifiable bone fragments, the faunal collection only al- The validity of the assumptions underlying his argumenta- lowed for the identification of four species. Although not tion needs, however, to be first evaluated. As developed be- representative, this sample is thought to indicate a general low, some of those assumptions rely upon undemonstrated cold steppic environment. Those results are surprising con- theoretical claims and concepts, inherently introducing a sidering that the excavators identified large concentrations great deal of interpretive bias into the analysis. of horse teeth and bone fragments, as mentioned in the first The following chapters (Chapter 13 by M. de Araujo- chapter. Igreja; Chapter 14 by H. Plisson) summarize the results of Based on extensive regional surveys, the identification the use-wear analysis of selected tools (n=153). The most of the lithic raw materials was more successful (Chapters 9 striking result is the identification of the use of small unre- and 12 by L. Slimak; Chapter 10 by Y. Giraud). The major- touched flakes (n=24) bearing micro-wears consistent with ity of the lithic raw materials used in the studied sample meat cutting. The systematic production of small, or micro, is of a local origin and reflects the intensive use of both lo- flakes using small discoid cores or truncated-facetted pieces cal pebbles (phonolith, quartz, basalt, granite), likely col- has been identified in several Middle Paleolithic contexts. lected from the nearby Loire riverbed, and local cherts of The demonstrated use of those tiny flakes at Champ Grand inconsistent qualities outcropping between 10 and 40km may further support the notion that an entire portion of the from the site and representing nearly 99% of the siliceous Mousterian technological repertoire has been underesti- raw materials. The remaining 1% is composed of non-local mated until recently (Dibble and McPherron 2006). If con- cretaceous cherts originating from various regions includ- firmed, this small flake component may further contribute ing the Parisian Basin (200km north from the site), the Cher to demonstrate that microlithism is an adaptive technologi- Valley (250km northwest), the Maconnais region (80km cal strategy that was not restricted to post-Middle Paleo- northeast), and the Rhône Valley (180km south). Such lithic contexts. Plisson’s use-wear analysis focuses on one long distance movements of raw materials are not uncom- single scraper made of translucent quartz interpreted as mon in Middle Paleolithic contexts in Central Europe (e.g., likely possessing special “meaning” or status, according to Féblot-Augustins 1999) but remain rare in Western Europe. the author, due to the rarity of its raw material. This short As such, these results support the notion of Mousterian chapter resonates as an attempt to demonstrate that a very groups moving over vast regions characterized by differ- detailed description of one single tool has the potential to ent ecosystems, possibly following the main river valleys. provide meaningful information regarding Neandertals’ Chapter 11 (L. Slimak) is dedicated to the typological and cognitive abilities, social organization, or symbolic expres- technological analysis of a sample of the original lithic col- sion. This daunting, and unconvincing, tour de force relies lection. Based on a chaîne opéraroire approach (Balfet 1991; upon a series of typical structuralist arguments concerning Inizan et al 1995), the analysis mainly focuses on the recon- the relationships between technology and its social environ- struction of the operational sequences associated with tool ment, further supported by Marxist anthropological refer- blank production. The operational sequences are divided ences (Godelier 1984; Lemonnier 1992). This short chapter into several phases focusing on the later stages of core re- is the only one attempting to provide the reader with direct duction (core shaping, core maintenance, and blank pro- theoretical references. Left open-ended, this demonstration duction per se) following Geneste’s methodology (1985). is also symptomatic of the underlying empiricist belief that These phases are isolated in the assemblage through the artifacts carry some intrinsic meaning directly reflecting a identification of morpho-technological flake categories de- broader system of social representations. fined by their overall morphology, the morphology of their In the opening sections of the book, Geneste initially dorsal surface, and the location and extent of their corti- defends the specificity of French Paleolithic archaeology 80 • PaleoAnthropology 2011 within a discipline he perceives as too often relying on the superficial penetration of foreign ideas” (Audouze and “now obsolete” materialist assumptions and plagued by Leroi-Gourhan 1981:182), while denouncing the “insular” “exclusively naturalistic” explanations (p. 17). The direct nature of the French archaeology of the early 1980s. This reference to a “Maussian anthropology” (p. 16) further con- reluctance to use ecological models, perceived as inher- tributes to delineate the theoretical paradigm of this book ently deterministic, better explains some of Slimak’s inter- as grounded in a structuralist anthropological approach pretation of the similarities in the distributions of lithic raw (Leroi-Gourhan 1943, 1945; Mauss 1936). Self-defined as material categories at Champ Grand and at the nearby Up- part of the social sciences, French Paleolithic studies have per Paleolithic site of Vigne Brun. Those similarities tend to always been reluctant to fully embrace a positive scientific demonstrate the continuity of procurement strategies be- paradigm. In brief, scientific reductionism is considered yond classic industrial boundaries that Slimak interprets as ill-equipped to fully grasp the complexity and subtleties of reflecting regional technological traditions. human social behavior, which can only be fully understood At a methodological level, those interpretive choices within its own system of social representation. At a fun- are supported by a chaîne opératoire approach of lithic as- damental level, French anthropology has suffered from a semblages (Balfet 1991; Inizan et al 1995). This approach rigid partition between hard (i.e., biologically based) and so- emphasizes the dynamic nature of the techno-economic cial sciences strongly influenced by humanistic undertones processes reflected in the archaeological record. As such, it and maintained by the internal organization of the French shares once again some methodological aspects of a broad- research and educative systems. As a result, French anthro- ly defined processual approach. However, the New Archae- pologists have systematically downplayed the role of bio- ology was built upon a critique of the previous cultural-his- logical and environmental factors in shaping or affecting torical interpretive paradigm as a whole. Leroi-Gourhan’s human behaviors and their evolution. At a paradigmatic new focus on technological processes did not require such level, this rebuttal may stem from the Cartesian distinction radical critique. His structuralist assumption that each between Body and Soul translated in the anthropological technological system is organized by and within its own realm into various dualistic distinctions such as nature-cul- social structure does not necessarily contradict cultural- ture, idéel-matériel (Godelier 1984), or milieu interne-milieu historical claims. As a result, while the goals of French Pa- externe (Leroi-Gourhan 1945). This philosophical bias has leolithic archaeology shifted toward the reconstruction of recently been under critique within French anthropology past technological behaviors, those behaviors were still in- (Shaeffer 2007) as it has led to a superficial, and often er- terpreted as cultural markers and broad cultural traditions. roneous, understanding of evolutionary processes and of In other words, the chaîne opératoire approach contributed the role played by the interaction between biological and to refurbishing some aspects of cultural-historical theory cultural/behavioral factors during our evolutionary his- by replacing traditional concepts, such as cultural norms, tory. This is illustrated by Slimak’s refusal to explain the by structuralist concepts such as “mental structures” or repeated occupation of Champ Grand as a direct conse- “mental templates.” Located at the very foundation of quence of the strategic location of the site for horse hunting, structuralist anthropology, this “mental template” gambit despite the accumulation of large amount of horse skeletal ultimately relies upon undemonstrable assumptions as it remains. For Slimak, “the idea of societies isolated or subju- fails to clearly tease out the complex relationships between gated by the exploitation of local resources is not here conceiv- techno-economic organization, social representations, and able.” (p. 405; my translation). Following his rationale, the cognitive mechanisms. Focusing on its methodological lithic raw material procurement strategies implemented at implications, the concept of chaîne opératoire used in this Champ Grand demonstrate the capacity for Neandertals to monograph ultimately requires accepting the existence of develop complex technological systems. Those systems ul- such “mental templates” in order to fully function as an timately reveal elaborate social and representational struc- interpretive tool. First intended to describe sequences of tures, which operated according to their own internal logic, technical actions (Balfet 1991), the chaîne opératoire concept unaffected by environmental constraints. Ironically, evolu- quickly became the keystone of French Paleolithic archae- tionary theories played a major role in French Paleolithic ology as it allowed linking detailed technological obser- archaeology as illustrated by Leroi-Gourhan’s comparative vations and structural explanations. Similarly, the term analysis of technology (Leroi-Gourhan 1943, 1945) based on of “concept” was later introduced in Paleolithic archaeol- evolutionary principles and sharing some similarities with ogy (e.g., Boëda 1993) to describe repetitive operational the contemporaneous, neo-evolutionist theses of L. White sequences identified in similar archaeological contexts. (1949). Similarly, Leroi-Gourhan’s palethnographie often Those sequences were therefore interpreted as the mate- has been compared to the American New Archaeology. The rial expressions of normative cultural templates determin- main differences between those two schools reside in the ing the modalities of tool blank production and ultimately scientific ambition of the New Archaeology, which quickly any technical activities. Practically speaking, a “concept” identified the need for middle range theories and for ecologi- is merely a representation of the unifying characteristics of cally based models directly relevant to the analysis of ar- an operational sequence identified in similar archaeologi- chaeological data. Leroi-Gourhan himself would describe, cal contexts. Those characteristics are not, however, solely 40 years later, the way the American Processual Archaeol- based on morphological criteria. Those “concepts” also ogy was perceived by French archaeologists as “typical of are based on the identification of redundant technical de- BOOK REVIEW • 81 cisions or mental processes inferred from the reconstruc- production or to secure reserves of blank volumes. Rather, tion of the steps followed during the operational sequence. Slimak concludes that the relative simplicity and the vari- Such methods have raised significant concern because the ability of the “Discoid concept” define it as a poor cultural validity of such inferences is based on our presumed abil- marker, thus avoiding constructing hypotheses here re- ity to identify the intentionality of technical behaviors in garding its adaptive characteristics. the archaeological record. Furthermore, as acknowledged Similarly, in the final chapter, Slimak compares some by Slimak, this approach tends to level archaeological vari- aspects of the lithic assemblages of Champ Grand with ability by introducing biased expectations during the initial that of the site of Néron Cave located further south in the descriptive steps of the analysis. Slimak’s re-evaluation of Rhône Valley (Slimak 2008). Slimak starts by stating that the Discoid concept is precisely motivated by the fact that both assemblages share very close similarities in terms of some published reconstructions of discoid blank produc- blank production, blank morphologies, and even metrics tion sequences (e.g., Lenoir and Turq 1995; Mourre 2003) (“to the millimeter;” p. 414), of the distribution of scraper do not necessarily meet expectations based on the original types, thinning techniques, recycling strategies, etc. Yet, definition of the “discoid concept” by Boëda (1993). Slimak then argues that these two sites are in fact differ- The core of Slimak’s critique hinges upon the notion of ent in terms of activities, environment and modalities of “predetermination” defined by Boëda as one of the criteria occupation. Slimak seems so inclined to demonstrate the defining Discoid and Levallois core reductions. According existence of technological traditions during the late Mous- to Boëda, true Levallois or discoid end products are pre- terian in southeastern France that he omits fully explain- determined by the removal of previous flakes whose func- ing how lithic assemblages composed of the same tool kits, tion is precisely to shape the volumes of the core in order bearing the same use-wear, supported by blank production to control the morphology of those final products. Prede- methods sharing “the same principles and the same objectives” termination thus requires assuming that a causal relation- (p. 414; my translation) can in fact reflect different activi- ship exists between the morphology of the product and the ties. Furthermore, while the faunal collection recovered at morphology of the surface from which the product is re- Néron Cave is relatively abundant and well preserved, the moved. Such relationship remains to be fully demonstrated faunal collection of Champ Grand greatly suffered from as it is so far mostly based on replicative experiments. Sli- poor preservation conditions (see Chapter 7). As a non-rep- mak’s main issue with Boëda’s definition is the fact that, resentative sample, the Champ Grand faunal assemblage for the Discoid method, any given flake is simultaneously cannot be used to support any meaningful reconstructions predetermined by previous removals and predetermines of the site activities and comparisons with other sites. It re- subsequent ones. In other words, he is not able to identify mains that the similarities of the assemblages from those flakes whose function would be to only maintain or adjust two sites are striking. As briefly mentioned by Slimak (p. the volumes of those discoid cores. Again, the prerequisite 196), Champ Grand also shares strong similarities with of both Slimak and Boëda’s argumentations remains that the site of Chez Pourré Chez Comte (France) located in the we are able to identify the specific purpose of each flake or western piedmont of the Massif Central in the Brive basin step of those core reduction sequences. This requires know- region. Chez Pourré Chez Comte is a large collapsed rock- ing the flake’s final intended use as well as assuming that shelter overlooking the Vézère Valley that yielded a very this intention existed prior to the flake’s actual production. large lithic collection likely reflecting the repeated occupa- The weakness of this demonstration resides in the fact that tion of the site. The lithic raw materials were primarily lo- this chaîne opératoire method focuses on the identification cal pebbles (quartz, quartzite) and semi-local Senonian and of repetitive or archetypal technological sequences and Jaspoid cherts collected within a 60km radius similar to the that it interprets such redundancy as reflecting culturally one identified for Champ Grand. The overall typo-techno- determined technological norms. As such, the intents be- logical makeup of both assemblages includes a large dis- hind those sequences of events are perceived as known or coid component, Levallois products, cores made of exotic even self-evident. Slimak demonstrates however that, if high-quality raw materials, and a small flake component archetypal blank production methods (or “concepts”) can (Lhomme 2000; Steenhuyse 2007). This monograph con- be identified in the archaeological record, the Discoid con- firms that looking at sites located on the margins of theclas - cept cannot be used as an effective cultural or industrial sic densely occupied regions may significantly contribute marker due to its overall simplicity. Although Slimak does to grasping the scope of Mousterian assemblage variability not push this demonstration any further, one could argue and how it was affected by regional factors. that the Discoid method is merely designed to produce a Beyond the narrow focus of its interpretive basis, this large amount of flakes following basic flaking principles. book remains an important contribution to our understand- Slimak does not formulate any hypotheses regarding pos- ing of the complexity of late Mousterian adaptive strate- sible economic constraints favoring the implementation of gies. A critical assessment of the theoretical assumptions the Discoid method over another blank production meth- supporting a broadly-defined chaîne opératoire approach is od. Considering the extent of the mobility patterns inferred long overdue. The evaluation of their validity and ability from the distribution of the lithic raw material categories to construct hypotheses testable using archaeological data at Champ Grand, we could expect to identify technologi- could in fact contribute to alleviating the weaknesses of cal strategies designed to maximize the return rate of blank this approach by strengthening its theoretical underlying 82 • PaleoAnthropology 2011 rationale. This monograph demonstrates, however, that the Leroi-Gourhan, A. 1945. Milieu et Techniques. Albin Michel, data thus produced can be integrated into alternative inter- Paris. pretive frameworks as long as it relies upon objective and Lhomme, V. 2000. Industries lithiques de Paléolithique moyen à consistent descriptive criteria and quantified observations. nombreuses fractures: les exemples du secteur sud de Cham- Other recent French archaeological publications (Bon 2009; plost (Yonne) et de la Couche inférieure de «Chez-Pourré- Valentin 2008) further demonstrate that the construction of Chez-Comte» (Corrèze). Unpublished PhD Dissertation, a more elaborate theoretical foundation for Paleolithic ar- Universite des Sciences et Technologies, Universite de chaeology is clearly perceived as one of the challenges of Lille I, Lille. the discipline. Mauss, M. 1936. Les techniques du corps. Journal de Psycho- logie 32(3–4): 271–293. REFERENCES Mourre, V. 2003. Discoïde ou pas Discoïde ? Réflexions sur Audouze, F. and A. Leroi-Gourhan. 1981. France: a conti- la pertinence des critères techniques définissant le dé- nental insularity. World Archaeology 1: 170–189. bitage Discoïde. In Discoid Lithic Technology - Advances Boëda, E. 1994. Le concept Levallois: Variabilité des Méthodes. and implications, M. Peresani (ed.), pp. 1–18. BAR Inter- Monographie du CRA. CNRS, Paris. national Series, Oxford. Bon, F. 2009. Préhistoire. La fabrique de l’homme. l’Univers Shaeffer, J.-M. 2007. La fin de l’exception humaine. NRF Es- historique, Paris. sais. Gallimard, Paris. Dibble, H. and S. McPherron. 2006. The Missing Mouste- Slimak, L. 1999. Mise en evidence d’une composante lami- rian. Current Anthropology 47(5): 777–803. naire et lamellaire dans un complexe mousterien du Féblot-Augustins, J. 1999. Raw Material transport patterns Sud de la France. Paleo 11: 89–109. and settlement systems in the European Lower and Slimak, L. 2008. Artisanats et territoires des chasseurs mouste- Middle Palaeolithic: continuity, change and variability. riens de Champ Grand. Artisanats et territoires 1. MMSH, In The Middle Paleolithic Occupation of Europe, C. Gam- Aix-en-Provence. ble and W. Roebroeks (eds.), pp. 193–214. University of Slimak, L. 2008. The Neronian and the historical structure Leiden, Leiden. of cultural shifts from Middle to Upper Palaeolithic in Geneste, J.-M. 1985. Analyse Lithique d’Industries Mousté- Mediterranean France. Journal of Archaeological Science riennes du Périgord: une Approche Technologique du Com- 35(8): 2204–2214. portement des Groupes Humains au Paléolithique Moyen. Steenhuyse, A. 2007. Continuity and change in technological Unpublished PhD Dissertation, Université de Bor- and economic strategies: the Middle and Upper Paleolithic deaux I, Bordeaux. industries of the Brive basin (France). Unpublished Dis- Godelier, M. 1984. L’idéel et le matériel. Fayard, Paris. sertation, Department of Anthropology, University of Inizan, M.-L., M. Reduron-Ballinger, H. Roche and J. Tixier. Pennsylvania, Philadelphia. 1995. Préhistoire de la Pierre Taillée - t. 4: Technologie de la Valentin, B. 2008. Jalons pour une paléohistoire des derniers pierre taillée. CREP, Meudon. chasseurs (XIVe-VIe millénaire avant J.-C). Cahiers Ar- Lemonnier, P. 1992. Elements for an Anthropology of Technol- chéologiques de Paris 1. Publications de la Sorbonne, ogy. Anthropological Paper, 88. Museum of Anthropol- Paris. ogy, University of Michigan, Ann Arbor. White, L. 1949. The Science of Culture: A study of man and Leroi-Gourhan, A. 1943. L’Homme et la Matiere. Albin Mi- civilization. Farrar, Straus and Giroux. chel, Paris.